DOMAIN: ELECTRICAL ENGINEERING Prof. univ. dr. ing. Marian ...

HABILITATION THESIS

RESEARCHES AND CONTRIBUTIONS IN THE FIELD OF ROBOTICS AND NEUROPROSTHESES

DESIGN AND CONTROL

DOMAIN: ELECTRICAL ENGINEERING

Prof. univ. dr. ing. Marian-Silviu POBORONIUC

GHEORGHE ASACHI Technical University of Iaşi

April, 2016

HABILITATION THESIS: Researches and contributions in the field of robotics and neuroprostheses design and control

2

Table of contents Chapters Page

1. Motivation of the request for obtaining the habilitation certificate in Electrical Engineering

4

2. Research directions and competences 6

2.1. Contributions to the development of new control strategies in robotics (especially mobile robotics)

8

2.2. Contributions on robotics modelling and control strategies to human biomechanics and rehabilitation

8

2.3. Contributions on FES-based neurorehabilitation by means of neuroprosthesis control

9

2.4. Contributions to Brain-Computer Interfaces based rehabilitation 10

2.5. Contributions to improving the European Union Electrical and Information Engineering Higher Education

11

3. Domains of competence complementary to Electrical Engineering 12

4. Technical-scientific report on the activities and research results 14

4.1. Mobile robotics control and applications 14

4.1.1. Strategies to control mobile robots on path tracking and obstacles avoidance

14

4.1.2. Steps towards virtual laboratories on mobile robots control 21

4.1.3. Voice controlled mobile robots 22

4.2. Neuroprostheses design and control 23

4.2.1. Neuroprostheses test benches 25

4.2.1.1. Lower limb neuroprostheses test benches 25

4.2.1.2. Upper limb neuroprosthesis test benches and hybrid FES&Mechatronic devices

32

4.2.2. Neuroprostheses control strategies 36

4.2.3. BCI techniques in neuroprostheses 40

4.3. Contributions to the EU Higher Education programmes oriented to the Renewable Energies and ICT Securities technical challenges – The EU SALEIE project

43

5. Carrier development directions that require the habilitation 50

Annex 1 - References 51


3

Abbreviations Medical Abbreviations: FES Functional Electrical Stimulation BCI Brain-Computer Interfaces MI Motor Imagery

CSP Common Spatial Patterns ERP Event Related Potentials BMI Brain-Machine Interfaces SSVEP Steady state visual evoked potential SCI Spinal Cord Injured MS Multiple Sclerosis CVA Cardiovascular Accident (or Stroke) SEMG Surface EMG EMG Electromyographic CNS Central Nervous System MD Medical Doctor EEG Electroencephalography PT Physical Therapist SU Standing-Up SD Sitting-down Technical Abbreviations: GUI Graphical User Interface PC Personal Computer ANN Artificial Neural Network HMM Hidden Markov Modelling

SR Speech Recognition VR Voice Recognition


4

1. Motivation of the Request for Obtaining the Habilitation Certificate in Electrical Engineering

Since October 1991 I was firstly holding an Assistant Professor position for Robotics and Mechatronic Systems Control at the Faculty of Electrical Engineering of GHEORGHE ASACHI Technical University of Iasi. Following all the university steps, since October 2014 I am holding a Professor position for Robotics & Neuroprostheses Control at the same faculty.

I was born in Piatra-Neamţ, Romania, in 1967. I have studied electrical engineering at the Faculty of Electrical Engineering of GHEORGHE ASACHI Technical University of Iasi, Romania from 1985 till 1991, and I have received the Dipl.-Ing. degree in 1991. In parallel with my duties as Tutor and Assistant Professor at the Faculty of Electrical Engineering of Iasi I have performed my doctoral studies and I have received the Dr. degree from the POLITEHNICA University of Bucharest in 2000. In 2001 I have joined the Institute of Automatic Control Engineering, Technical University of Munich, where I have pursued postdoctoral research in the field of modelling and control of neuroprostheses as FP6 NeuralPRO TMR Fellow. I have continued my postdoctoral work as NeuralPRO Control Engineer within the Department of Medical Physics and Biomedical Engineering, University College of London & Department of Medical Physics and Engineering – Salisbury Districy Hospital, UK, from 2002 till 2003. In 2003, once returned to the GHEORGHE ASACHI Technical University of Iasi to act as Associate Professor, I have coordinated several research projects dealing with neuroprostheses control algorithms and being also active in the Neurology Clinic of the Rehabilitation Hospital of Iasi by jointly teaching Functional Electrical Stimulation courses and developing neuroprostheses supporting gait and arm therapies.

My current research interests involve mobile robots control algorithms, human motion analysis and synthesis, neuroprostheses design and control, biomechanics, brain-computer interfaces (BCI), FES&BCI based rehabilitation and rehabilitation robotics. I have authored and co-authored more than 140 journal and conference articles (4 in ISI journals, 21 indexed in ISI Web of Science, 14 indexed in other databases as Scopus, IEEExplore, INSPEC, Copernicus, EMBASE, EBSCO, Scirius, WAME), 2 license patents and 4 new license patents are pending. I am a member of IFESS, EAEEIE and SETIS, a speaker of the RAAD2009, EPE2006, and a member of several conferences International Program Committees (e.g. IASTED International Conferences on Modelling, Identification). The performed research has been awarded with several prizes including the bronze medal to the 3rd International Exhibition of Invents, Research and Technological Transfer, October 9-13, 1996, Iasi, Romania, the Jean Peperstraete AWARD of EAEEIE (European Association for Education in Electrical and Information Engineering) for the best presented paper, during The 17th EAEEIE Annual International Innovation in Education for Electrical and Information Engineering, June 1st-3rd, 2006, Craiova, Romania, GOLD medals and CYBERLIFE AWARD at EUROINVENT2014- European Exhibition of Creativity and Innovation, 22-24 May 2014, Iasi, Romania.

My first steps in research have been mostly related to my PhD degree ("Industrial Robot Identification and Control") and touched the robotics control field by applying advanced method controls (e.g. fuzzy and ANN control, H-infinity control, control methods based on the Lie Algebra, modelling and control of parallel robots [81]÷[89], [93], and newly [39], [44],[45], [49], [50], [56], [57], [59], [64], [69], [74] ÷ [77]).


5

Starting with 2001 during my involvement in a European project NeuralPRO (Marie Curie Research Training Network Neural Prostheses -NeuralPRO - HPRN-CT-2000-00030) and after, since nowadays, my research has been focused to apply robotics modelling and control theory to develop neuroprostheses aiming to restore walking in paraplegia or to help the rehabilitation processes in disabled people due to a central nervous system lesion (e.g. stroke, multiple sclerosis, cerebral palsy, etc.). The electrical engineering knowledge have been specifically used to:

Develop hardware platforms (FES, BCI, embedded systems [47], hybrid FES&exoskeletons based rehabilitation systems of the upper limbs in CVA [8], [10], [11], [17] - Gold Medal Awarded at EuroInvent2014, [35], [24]-OSIM Licence patent request no.693/15.09.2014, hybrid FES & mechanic glove to rehabilitate the hand in people with CVA [22], [36], [38] si [25] – Gold Medal Awarded at EuroInvent2014, laboratory tests benches [37], [51], [54], [58], [61], [62], [63] devices to assess the rehabilitation processes outcomes [65], etc.);

Develop functional clothes integrating the surface stimulation electrodes to be used together with the developed neuroprostheses (director of UEFISCDI grant NOVAFES 267/2014, [7], [15], [31], [42], pending patent licenses [OSIM-A00673/21.09.2015; OSIM-A00787/03.11.2015], [16])

Develop new control strategies that command the neuroprostheses to support standing and walking in paraplegia, [46], [24] - pending licence patent – OSIM request RO-129704-A2/29.08.2014, [23] – CYBERLIFE Award at EuroInvent 2014, [52], [53], [60] and to rehabilitate the upper limbs in CVA patients [19], [21], [48].

Performing the clinical trials together with MDs ([12], [18], [19], [20], [43]). It is worth to notice that during the past few years I was strongly involved in cooperating

and leading the ALGCON group which supported three thesis on the FES&BCI neurorehabilitation systems:

1. Danut Irimia-“Neuroprostheses control and experimental testing” (Controlul şi testarea experimentală a neuroprotezelor)- presented in October 2012 (PhD coordinator Professor Gheorghe Livint);

2. Serea Florin – “Researches on designing a robotic FES&exoskeleton system to rehabilitate the upper limbs in disabled people” (Cercetări privind realizarea unui sistem robotic FES-exoschelet pentru reabilitarea membrelor superioare la persoanele cu handicap) – presented in September 2014 (PhD coordinator Professor Radu Olaru);

3. Sergiu Hartopanu – “Researches to design and control an intelligent FES&robotic glove aiming to rehabilitate the hand in disabled people” (Cercetări privind realizarea și controlul unui echipament de tip FES-mănușă robotică inteligentă pentru recuperarea funcțiilor motorii ale mâinii) - presented in September 2014 (PhD coordinator Professor Gheorghe Livint). I was acting as member of the PhD thesis analysis Commission during their public

presentation (Rector Order nr.1970/19.09.2012 (first one), nr.1481/01.09.2015 (last two)). The SALEIE1 (No. 527877-LLP-1-2012-1-UK-ERASMUS-ENW) project, EU Lifelong

Learning Programme funded project, aims “to explore and then provide models for ways in which Higher Education Institutions of Europe in the Electrical and Information Engineering disciplines can respond to current global technical challenges”. I have been acting as workpackage (WP3) leader and together with another 45 partners across Europe. The first

1 http://www.saleie.co.uk/index.php - [accessed on February 4th, 2016].

http://www.saleie.co.uk/index.php


6

stage in the project was to undertake a ‘desk research’ exercise to identify existing programmes orientated to the key global challenges. These technical challenges are in line with the challenges that the new launched Horizon20202 EU Research and Innovation programme for 2014-2020 addresses. Based on the survey’s techniques, brainstorming sessions and partners’ research and expertise we finally sorted out two new curricula on Renewable Energies and ICT Securities that we consider to be able to respond to global technical challenges the EIE graduates may face nowadays. To gain confidence in the model curricula a sample of the required modules were tested by being delivered and evaluated in partner institutions. Five modules were selected for evaluation and each has been report on individually. The reports have to be found on the project website and enlist all these findings.

The present submitted habilitation thesis has been motivated by all my undertaken research and accomplishments in the Electrical Engineering field. This step forward would enable me:

to contribute, by developing new knowledge and sharing acquired knowledge, to benefit any debutant researchers;

to further contribute to the development of the Romanian research and Electrical Engineering education;

to increase the impact and visibility of Romanian research at the international level.

to strengthen the interdisciplinary research with other research centres from abroad (e.g. by jointly coordinating PhD thesis).

2. Research Directions and Competences Since the year 2000, when I have received my PhD degree, the main research has been

directed towards:

Development of new control strategies in robotics (especially mobile robotics);

Applying robotics modelling and control strategies to human biomechanics and rehabilitation;

FES-based neurorehabilitation by means of neuroprosthesis control;

Brain-Computer Interfaces based rehabilitation.

Improving European Union Electrical and Information Engineering Higher Education (e.g. EU SALEIE project3)

After finalizing my PhD, I have been involved in a total 17 research projects (4 EU

funded projects, 4 bilateral projects with Slovenia and Turkey, 9 national grants (PNII, CNCSIS) awarded by competition). I was acting as project responsible –WP3 leader within the SALEIE3 EU funded project, director of 7 national awarded grants (e.g. PNII-NOVAFES 267/2014, PNII-NEUROTEH 31CB/2008, PNII-SINPHA-D11-068/2008, PNII-ARMS-D71-095/2008), TUIASI project responsible of 2 national awarded grants (e.g. IHRG-150/2012; ROKEY- 'Robot Technologies as a Key for Learning in Future' - COMENIUS project COM-08-PM-653-VS-DE).

2 http://ec.europa.eu/programmes/horizon2020/en/what-horizon-2020 3 “Strategic Alignment of Electrical and Information Engineering in European Higher Education Institutions (SALEIE)” No. 527877-LLP-1-2012-1-UK-ERASMUS-ENW, 2012-2015, Coodinator: University of York, UK; Project responsible P12-TUIASI: Marian Poboroniuc (WP3 leader).

http://ec.europa.eu/programmes/horizon2020/en/what-horizon-2020


7

The outcomes of the projects in which I have acted as director/project responsible or as member have been disseminated in 7 books (publishers with ISBN), more than 120 journals and conferences published papers (after 2000), among them 4 being published in ISI journals, 21 indexed in ISI Web of Science, 14 indexed in other databases (e.g. Scopus, IEEExplore, INSPEC, Copernicus, EMBASE, EBSCO, Scirius, WAME). I own 2 license patents and 4 new license patents are pending. It is worth to mention a number of 3 gold medals, CYBERLIFE AWARD4, Jean Peperstraete AWARD of EAEEIE (European Association for Education in Electrical and Information Engineering) awarded to the different international trades on inventics and research arising ideas.

My deep knowledge and skills are related to:

Robotics modelling and control;

Modelling and simulation in Electrical Engineering;

Theory of Systems and Modelling and Identification of Systems;

Programming in C++, Matlab&Simulink;

FES-based rehabilitation (modelling and control of the human body by means of FES);

Neuroprostheses design and control;

BCI&FES based rehabilitation. I would also like to believe that I have the potential to greatly potentiate the students

and young researchers to perform within the Electrical Engineering field and interdisciplinary projects (e.g. student Chirila Marian - first prize at “Salonul National de Creatie si Inventica pentru Tineret, Sectiunea: Tehnologia Informatiei”, Bucharest 18-22 November 2010, Romania – see figure 2.1 -left; PhD student Hartopanu Sergiu (PhD thesis coordinator Professor Gheorghe Livint), PhD student Danut Irimia (PhD thesis coordinator Professor Gheorghe Livint), PhD student Serea Florin (PhD thesis coordinator Professor Radu Olaru), gold medals at Euroinvent 2014 – see figure 2.1-right)

Fig.2.1 Medals and diplomas obtained together with students and young researchers which I jointly tutored during their studies/research (e.g. CYBERLIFE award at Euroinvent2014). First picture with

Marian Chirila; Second picture- from left to right: the PhD students Sergiu Hartopanu (first) and Danut Irimia (last) received their PhD thesis coordination from Professor Gheorghe Livint and Serea Florin

(third) received his PhD thesis coordination from Professor Radu Olaru

4 CYBERLIFE AWARD, Future Medical Devices controlled by means of Brain-Computer Interfaces, Marian Poboroniuc, Dănuț Irimia, Florin Serea, Sergiu Hartopanu, (6 th Edition of the European Exhibition of Creativity and Innovation), 22-24 May, 2014, Iasi, Romania;


8

2.1 Contributions to the development of new control strategies in robotics (especially mobile robotics)

Since the year 2000, I was constantly involved in developing new robotics applications.

That involved new control strategies (e.g. fuzzy and neural control in path tracking [64], [82], [84], vehicles stability and control [69], [74], [75]), hybrid robotic systems [81], robotic assisting systems in physically impaired people [51], [62], [76], [77], voice recognition based control in robotics [50]. The developed control strategies have provided a solid basis for continuing research within the neuroprosthesis field, proposing new research themes and being granted with national and international bilateral cooperation grants (e.g. Slovenia, Turkey, UK, Germany), improving the higher education lectures (e.g. proposed new Master lecture :Control algorithms in mobile robotics) and most of all coagulating a young research group of students to perform within this field (e.g. [56], [57], [59], 1st prize at StudING2009, 1st and 2nd prize at “Salonul National de Creatie si Inventica pentru Tineret, Sectiunea: Tehnologia Informatiei”, Bucharest 18-22 Noiembrie 2010).

2.2 Contributions on robotics modelling and control strategies to human biomechanics and rehabilitation

All my knowledge within the robotics field proved to be a solid background to perform

in a very challenging field of neuroprosthesis aiming to rehabilitate people with disabilities due to a central nervous system lesion. That chance arise once being involved as postdoc researcher within the EU NeuralPRO project (Research Training Network Neural Prostheses - HPRN-CT-2000-00030: March 2001-February 2003) and based on my research results I have been Nominated for the -Marie Curie Excellence Award- (FP6-509704 NeuralPRO), 2003. The nomination has been based on the active role in managing the cooperation between the Technical University of Munich and University College London groups and the research activity and results obtained within the field of biomedical system engineering and neuroprosthetics.

Paralysis due to spinal cord injury leaves the muscles and their innervating motoneurons below the level of the lesion, largely intact. Functional Electrical Stimulation (FES) is a means of producing useful movement in paralyzed limbs. A prerequisite for walking is standing. Standing in paraplegia can be achieved if an appropriate stimulation pattern is at least provided to the knee extensors. The joint movement can be achieved and controlled by modulating the amplitude of stimulation over the flexor and/or extensor muscles of the joint. Prior applying any electrical stimulation over the muscles of disabled people in order to induce standing and walking, the neuroprostheses have to be intensively laboratory tested and the ethical approvals steps fulfilled. Since 2001 I have developed different human body models under Matlab&Simulink environment ([67], [68], [70], [71], [78], [79]) as well as test benches ([61], [62], [77]) where human-like robots replicated the possible movements of the human body supposed to be controlled by a neuroprosthesis. For example, in [78] a new proposed fuzzy controller modelling the voluntary upper limb contribution of paraplegic patients when performing FES-based sitting-down motion tasks has been presented. It has been integrated in a more comprehensive patient model [91] and it is currently used to test control strategies aiming to support FES-based standing-up and sitting-down in paraplegia. It is integrated as


9

well in the developed test benches ([62], [77]) aiming to validate the performances in FES-based controlled SU and SD in paraplegia by means of robotic devices which mimic the human body movements. All these achievements have been presented also during different teaching courses which I have initiated during the last two decades (e.g. Functional Electrical Stimulation course - (The FES Single-Channel Course -Essex, UK, February 6th and 7th, 2003); Post university courses: FES-based rehabilitation in stroke patients, jointly teaching with people form Salisbury District Hospital, UK, “Gr.T Popa” Medicine and Pharmacy University of Iasi, 10 annual editions, held at Rehabilitation Hospital of Iasi).

2.3 Contributions on FES-based neurorehabilitation by means of neuroprosthesis control

Rehabilitation of paraplegic patients revolves around the goal of providing or restoring

their ability to walk. A prerequisite for walking is standing. It benefits the patient physiologically and psychologically. The physiological benefits are in terms of reduced risk in disuse osteoporosis and subsequent fractures, and of improved digestion, respiration and urinary drainage. The upright posture is associated with fewer decubitus ulcers and lessened spasticity [46], [73]. The psychological benefits of even temporary social interaction at normal eye level should not be underestimated. Based on specific courses which I followed during my stage within the EU NeuralPRO project (e.g. "Reducing Time to Market through Model-Based Design" Seminar, The MathWorks, Cambridge Business Park, September 19th, 2002, Cambridge, UK; "Neurorehabilitation of Movement for Humans with Central Nervous System Injury or Disease" course, Trade Union BEC, Kotor, Yugoslavia, July 1st-5th, 2002; The Functional Electrical Stimulation 2-Channel Course, Salisbury, UK, July 12th-13th, 2002) as well as the clinical experience earned at Grosshadern Hospital, Munich and Salisbury District hospital, UK, I have been able to propose and embed neuroprosthesis with new control strategies aiming to help paraplegics in performing SU, SD (e.g. new ONZOFF controller [46] ISI indexed journal article, awarded with CYBERLIFE AWARD at Euroinvent2014 [23], licence patent request RO-129704-A2/29.08.2014- International Patent Classification: A61N-005/08 as enlisted within Web of Science& THOMSON REUTERS [24], [79]) and even walking ([80], final tests together with dr. Thomas Fuhr at Grosshadern Rehabilitation Hospital, Munich, Germany).

Of great importance has been to transfer the research knowledge towards clinical practice. For example, new proposed control strategies (e.g. ONZOFF [46], ISI indexed journal paper), as well as some other proposed at international level ([92], [95]) have been integrated into a new application programme CLSTDSD aiming to sustain a standing-up, standing and sitting-down chained motion in paraplegia [72]. It has been implemented within the 8-channel Stanmore stimulator and tested in a clinical environment (Salisbury District Hospital & University College London, UK, Rehabilitation Hospital of Iasi, [72]). New hardware prototypes based of FPGA implementation of neurostimulators have been proposed and tested too ([47], ISI indexed journal paper).

The strong connection with the applicative part of the neuroprosthesis (Rehabilitation Hospital of Iasi, Rehabilitation Hospital of Cluj-Napoca), as well as top research which has been proposed within few awarded UEFISCDI grants (e.g. NOVAFES-PCCA267/2014- project director; IHRG-PNII-150/2012-project responsible; EXOSLIM-PNII-180/2012- scientific director), lead us to results which have been well received by the scientific community and rehabilitation professionals and awarded at different exhibitions and innovation trades (e.g.


10

EUROINVENT2014- 4 gold medals, Cyberlife Award; Invest- Invent medal Excellence diploma (International Trade on Inventics and Practical Ideas- Bucharet, june 17-23, 2013) for the idea of the requested patent OSIM no. A00170/25.02.2013, gold medal at Inventica’2011, etc.).

Restoring the normal upper limb motor functions for stroke patients has not shown major improvements with standard rehabilitation methods. However some new concepts of rehabilitation have shown great potential for restoring motor functions to some subjects, like Functional Electrical Stimulation (FES) and robotics or FES along with intensive motor learning [49]. During the last few years we have proposed new approaches as well and designed new devices like hybrid exoskeleton-FES aiming to improve the overall upper limb rehabilitation process in stroke patients. For example, the EXOSLIM ([10], [14] [17]-gold medal, [32], [35]) system consists in its balanced control among FES muscle activation and the lightweight exoskeleton itself creating the premises for a better recovery of the upper-limb functions. A patent license is pending (request no.693/15.09.2014 [13]). The hand rehabilitation and the aiming to improve its dexterity in patient which suffered a stroke have been addressed by means of a new proposed IHRG system [11], [12], [14], [22], [25]-gold medal, [36].

The devices that deliver electrical stimulation and aim to substitute for the control of body functions that have been impaired by neurological damage are termed ‘neuroprostheses’. Regular standing (e.g. by means of FES) in spinal cord injured subjects is thought to help in preventing osteoporosis, preventing contracture by preserving the range of movement at lower limb joints, improving digestion, respiration and urinary drainage etc. A new controller ONZOFF ([46]) has been proposed and clinically tested. It better controls the lowering of the person toward the seat, by reducing knee-end velocity and handle reaction forces. A patent licence on that idea is pending (Licence patent RO-129704-A2 / 29.08.2014; International Patent Classification: A61N-005/08 [24]).

Newly, the new project NOVAFES PCCA267/2014 which I am leading as director, aims to design functional clothes for persons with locomotor disabilities [7], [15] [16]. A possible solution for manufacturing incorporated textile electrodes consists in the insertion of some electro-conductive yarns onto textile surfaces by using a variety of technologies. The project approaches the use of knitting, a widespread textile technology. The knitted versions were tested by using a Microstim2v2 (PW=300μs, 40 Hz) neurostimulator for which the current parameter was adjusted to approximately 30mA. Clinical trials have been planned for the period 2016-2017. Two licence patents are pending ([1] OSIM nr.A00673/21.09.2015 – second prize at Practical Ideas-Invest-Invent2015; OSIM-A00787/03.11.2015 [2]).

2.4 Contributions to Brain-Computer Interfaces based rehabilitation. In Brain-Computer Interfaces (BCI) applications, users explicitly manipulate their brain

activity instead of motor movements to produce signals that control external devices. Usually, the brain signals are obtained non-invasively from the scalp through electroencephalography (EEG). The main idea of the proposed motor imagery (MI) based BCI&FES system is to decode the EEG signals and to provide the neurostimulator with the requested command in order to control the hands and the lower limbs in accordance with a rehabilitative pattern (stroke patients, paraplegics) [12], [14]-mention II, [39], [26]-Gold Medal. The method of common spatial patterns (CSP) has been used to discriminate among the motor imagery tasks. One of the studies evaluated the influence of choosing different time windows for calculating the CSP and a feedback strategy that can be used to initiate an FES-assisted movement task [40]. We have been able to discriminate between three classes and to control the upper limbs and lower limbs in paraplegics by means of a PC controlled neurostimulator [33].


11

BCI in robot control earned a lot of attention nowadays. Better EEG signals classifiers may lead to improved control in mobile robots to guide their movement in unknown environment and to perform specific tasks. Different BCI paradigms have been tested and new outcomes have been produced in terms of evaluating the influence of choosing different time windows for calculating the CSP and two different feedback strategies that can be used to initiate a robotic movement task have been compared [39].

Of great importance is the new system called recoveriX that is designed to detect movement imagery through EEG-based BCI technology, and use this imagery to directly control an FES system and avatar-based feedback [5]. It saw the daylight as a joint cooperation with a well-known company GTEC Medical Engineering from Austria and is currently under clinical testing phase. Tests are to be conducted at Rehabilitation Hospital of Iasi.

2.5 Contributions to improving the European Union Electrical and Information Engineering Higher Education

The need for competiveness of Europe’s higher education and lifelong learning

systems in a global perspective where competing economies (China, India, Korea) invest heavily in up skilling their new generations who largely outnumber EU populations of highly qualified engineers is essential. Against this backdrop, the Association for Education in Electrical and Information Engineer (EAEEIE5) has been for 20 years and still is dedicated to supporting Electrical and Information Engineering (EIE) across Europe [28], [29], [30]. As an EAEEIE member I have been actively involved in its last European project SALEIE6 (Strategic Alignment of Electrical and Information Engineering in European Higher Education Institutions) as work package leader (WP3- Global Challenges). One of the objectives of the SALEIE project, embedded within the WP3, is to enhance the competitiveness of Electrical and Information Engineering (EIE) education within Europe, especially in relation to modern global technical challenges [3]. In order to compare different EIE programmes which are supposed to respond at a certain extend to the identified key technical global challenges, a set of criteria have been defined and clearly explained [6], [34]. Finally, based on an extended research over Internet and on a base of a short questionnaire which has been discussed during workshops and circulated between SALEIE partners, a number of relevant data related to EIE programmes have been collected and commented [4], [9], [27], [28]. The analysis took into account EIE programmes from different higher education institutions on the topics regarding: Sustainable development and climate change, Energy, Clean Water, Global convergence of IT, Health Issues, Food security, and Green/Integrated Transport. The survey identified where expertise lie within the project consortium in sufficient depth to enable us to propose quality programmes and modules. Two of the above mentioned areas emerged as those in which we could work, Renewable Energies and ICT Securities. The survey also gathered more general information about programmes across Europe [6]. The main outcome of these initial reviews was information that enabled the project team to create model curricula, a secondary outcome was examples of existing programmes across Europe. Given the focus on the Renewable Energies and ICT Securities technical challenges, model curricula were designed for Bachelor and Master programmes for each challenge area [4], [27]. A

5 http://www.eaeeie.org/ [accessed on February 26th, 2016]. 6 SALEIE project - Project Reference No. 527877-LLP-1-2012-1-UK-ERASMUS-ENW; Project funded by the EU Lifelong Learning Programme. http://www.saleie.co.uk/ [accessed on February 26th, 2016].

http://www.eaeeie.org/

http://www.saleie.co.uk/


12

report on each of these curricula including the background and full programme specification and all the module specifications are available on the project website6. To gain confidence in the model curricula, a sample of the required modules were tested by being delivered and evaluated in partner institutions. Five modules were selected for evaluation and each has been report on individually.

3. Domains of competence complementary to Electrical Engineering Different activities which are complementary to Electrical Engineering field have been

performed, and they sustain the personal achievements recognition by different national and international institutions and specialists. A short selection may enlist:

Expert evaluator: o EU Expert: Vice-Chair – EU commission of H2020-FETOPEN-2014-2015; CT-

EX2002B055048-103/ November - December 2015, Brussels, Belgium; o EU Expert Evaluator: H2020-FETOPEN-2014-2015-RIA (7 proposals - remote); CT-

EX2002B055048-102 / May 2015. o EU Expert Evaluator - Horizon2020; Call: FET-OPEN- Novel Ideas for Radically New

Technologies H2020-FETOPEN-2014-2015-RIA (4 proposals - remote); 31.10.2014-30.11.2014 (Expert contract number: CT-EX2002B055048-101).

o External Expert- project ORTHO-eMAN (2011-1-RO1-LEO05-15321; contract LLP-LdV/ToI/2011/RO/008)-2014- evaluated curicula "Human Motion Analysis on-line course addressed to engineers" (March 2014).

o Evaluator COST (European Cooperation in Science and Technology): EU level- proposal OC-2013-2-17479 ( 9-30.01.2014).

o Expert evaluator EU - COST: COST Open Call – oc-2013-2 / 1 proposal/2013. o National expert project POSDRU 86/1.2/S/57748; contract 254/29.02.2012 at

UPB; o Expert evaluator EU – activities RTD (DIR F REVIEWS 2004 IST FP5)-Project TETRA

(Development of Tendon foRce TrAnsducer for neuroprotheses) – IST-2001-38948), 02-04.04.2004, Brussels, Belgium.

o Expert evaluator EU – activities RTD (DIR F REVIEWS 2004 IST FP6 – STREP) – Project DETECT (Development and Test of an implantable tendon forCe Transducer), 16.11.2004.

o Expert evaluator – CALIST programme – “Program National pentru Calitate si Standardizare” – CALIST, coordinated by CNCSIS – call C5-2005, July 7-9, 2004, Iasi, Romania.

“Peer review” activities: o IEEE Transactions on Biomedical Engineering – TBME-00247-2004 (2 papers),

November 2004 – appreciation letter (10.02.2005) received from Jose C. Principe, journal chef editor; Bulletin of the Polytechnic Institute of Iasi – 2004.

o Member of the International programme committee of IASTED Int. Conference on Modelling, Identification and Control, since 2006 (37 papers).

o Other conferences: EPE2004-2014, SIELMEN2005-2015, ISI indexed ‘Industria Textila’ journal – no.4/2015, ICSTCC13, SINTES17.


13

Invited speaker: Essex Rehabilitation Hospital, UK (teaching FES courses, 2002); Salisbury District Hospital, UK (neuroprosthesis control, April 25th, 2002); Plenary lecture – RAAD2009 - 18th Int. Workshop on Robotics in Alpe-Adria-Danube Region; Plenary lecture –EPE2004- 3rd International Conference on Electrical and Power Engineering; Valcean Medical Days 2013, Felix Medical Days -2012.

Jointly teaching postuniversity courses on “FES-based rehabilitation in disabled people due to a central nervous system lesion” (Salisbury District Hospital, UK, Rehabilitation Hospital of Iaşi, “Gr.T. Popa” Medicine and Pharmacy University of Iasi, 10 editions since 2005). Accredited FES trainer since 2002 – Salisbury District Hospital, UK.

Trained on: o "Neurorehabilitation of Movement for Humans with Central Nervous System

Injury or Disease" course, Trade Union BEC, Kotor, Yugoslavia, July 1st-5th, 2002; o The Functional Electrical Stimulation 2-Channel Course, Salisbury, UK, July 12th-

13th, 2002. o The Functional Electrical Stimulation Single-Channel Course, Stanmore, UK, April

6th-7th, 2001. o Course on human motor control (Brain motor control assessment, Motor effects

of brain stimulation in humans, Locomotion- Treadmill training in rehabilitation), September 11th-12th, Vienna, Austria.

o "Reducing Time to Market through Model-Based Design" Seminar, The MathWorks, Cambridge Business Park, September 19th, 2002, Cambridge, UK: 'Product Design & Development', 'Process Improvements using Model-Based Design', 'Realising a Design - case studies', 'Implementing a Design'.


14

4. Technical-scientific report on the activities and research results Nowadays, the robots started to play an essential role in our lives and almost any

human activity may benefit from their support. The bio-inspired robots, the human-like ones and those involved in rehabilitation are only a few examples of robots which are far away from their possible design limits, implemented artificial intelligence and integration within human real life. Often, only the imagination is the one which may offer an idea about the robotics future (e.g. the “Ex Machina” (2015) movie which won the 2016 Oscar for “Best Achievement in Visual Effects” and has been nominated for the “Best Writing, Original Screenplay”; It tells a story of a programmer which has to test an android with artificial intelligence). Besides the mobile robotics applications and research which will be presented below, the rehabilitation robotics covers an important part of the performed research since receiving the PhD degree in Automated Systems field in 2000.

4.1 Mobile robotics control and applications There is a tremendous potential of applications in robotics. The main research has

been performed on mobile robots path tracking control, voice controlled mobile robots, rescue robots (legged robots, with visual cameras; see figures 4.1 to 4.4) [56], [57], [59], [64], [66], [75], [82]. Most of these applications have been developed by attracting the young researchers and students towards mobile robotics research and the outcomes received numerous prizes (e.g. 1st and 3rd prize at StudING competition, Targu Jiu, 2008, 2nd prize at ELStudiIS2010, Iasi, 1st and 2nd prize, medal and cup at Salonul National de Creatie si Inventica pentru Tineret (National Saloon of Inventics and Creative Thinking), Bucuresti 18-22 Noiembrie 2010), [56], [57], [59].

4.1.1. Strategies to control mobile robots on path tracking and obstacles avoidance Strategies to control mobile robots on path tracking are of interest on different

applications (e.g. evolving in unknown environment and mapping small areas). Modern methods which involves neural and fuzzy controllers have been taken into account [64], [85], [86]. Good performances were obtained for trajectories like circles and sinusoids, but in hamper forms case they must be improved by using some prediction techniques. The mobile robot has two pairs of wheels with common axis: one pair of free rear wheels and the second one constituting the driving and steering front wheels. Its mathematical model has been deduced, the neural and fuzzy controllers have been proposed and tested in simulation and practice.

Fig.4.1 ALGCON research laboratory: mobile robotics applications (e.g. remote control of a robot by means of

visual cameras (middle); voice controlled robots (right side) [56]).


15

Fig.4.2 Mobile robots applications: trajectory tracking (left side), objects recognition and pick-up (middle),

hybrid robots (mobile platform & serial robot) cooperation (right sight)

Fig.4.3 Participating to different contests (Robochallenge 2011-2013, Bucharest, VIGOR 1 to 3 series of SUMO

robots – first two pictures; ROKEY- "European Robots Contest", Vocational High School "Stefan Procopiu", Vaslui, Romania- autonomous robots in obstacle avoidance tasks and reaching a target - right side)

Fig.4.4 Ensuring the connection research-industry (e.g. The Continental organized contest

Electromobility – left side two pictures); Humanoid robots control (Summer school SIROCO2010)- right side.

The mobile robot has two pairs of wheels with common axis: one pair of free rear

wheels and the second one constituting the driving and steering front wheels. Its mathematical model has been deduced, the neural and fuzzy controllers have been proposed and tested in simulation and practice. The space configuration of a mobile robot can be described by a vector qc=(xc, yc, φc)T where (xc, yc) are the coordinates of a reference point C, placed on the robot (e.g. on the middle of the axis between the free rear wheels), with respect to the world frame W and angle φc represents the direction of the car with respect to the X-axis. The robot's kinematics is described as follow:

T

L

cv

ccv

ccv

Tcc

yc

xtc

q

tansincos)( (1)

with the kinematic constraint (2)

0cossin yx (2)


16

where vc represents the linear robot velocity, supposed constant, L the robot length and ϴc steering angle.

In order to describe the path following task a virtual robot following the desired trajectory is considered. The path following problem consists in finding a controller ensuring a geometrical convergence of the real robot towards the path which is followed by the virtual robot. Assuming that O' is the frame of a kinematically equivalent virtual robot following the desired trajectory, the error configuration vector qe between the real and virtual robots with respect to O' is then:

22

cyrycxrx

cr

dd

deq

(3)

which will feed the fuzzy and neural controllers providing the steering angle ϴc to

control the robot towards the desired path [82].

d

dd

cFuzzy controller Mobile robot

Geometric transformation xcyc

xr

yr

r

c

Fig.4.6 Fuzzy controller structure.

The essential elements in designing a fuzzy controller include defining input and

output variables, choosing fuzzification and defuzzification methods and above all, determining the rule-base of the controller. As input for our fuzzy controller we have chosen

the variables d and dd, and as output the steering angle c. Triangular membership functions for the inputs and output are used. Minimum decomposition is chosen for fuzzy inference procedure and the center of gravity method for defuzzification. Neural networks allow

Fig.4.5 The description of the path following problem.

R

xc

yc

Y

X

W

C

dd

c

yr

xr

r

d

reference trajectory


17

reducing different systems to one simple structure composed of nodes and connections, representing respectively nonlinear functions and parameters. Some simple algorithms might be developed to designing neural models or neural optimal controllers. The objective of these algorithms is to identify the neural network parameters or weights, in the phase called the learning phase.

The simplest well known algorithm is the gradient backpropagation. The Levenberg-Marquardt algorithm avoid singularities, and permits a more rapid convergence of the parameters. Other algorithms have been developed with modifications, for optimizing the time of parameter convergence. The Levenberg-Marquardt rapid learning method has been used to design the mobile robot controller.

The learning process consist in calculating the weights W such that the estimated output y is close to the corresponding desired output y, like presented just below. To satisfy

this objective, learning algorithms that minimize the following quadratic criterion (4) are used:

),(ˆ)()(),(

),(N2

1=)(

1

2

WkykykWk

WkWEN

k

(4)

Thus, the error between the estimation and the desired state y is minimized. The

updating rules are as follows:

)()(11 iiii WEWRWW (5)

where, if adequately choosing the term it allows the convergence even for points far

from the minimum, Wi is the parameters vector after i updates, and )( iWE is the gradient of

the quadratic criterion to minimize, defined as :

nw

E

w

E

WE

1

)( with

N

k

N

k

N

k

k

iw

ky

N

k

iw

k

Nk

Ni

wi

w

E

1

11

2

)(.)(ˆ1

)(.)(1

)(1

(6)

dd

d

NL NM ZE PM PL

NL PL PL PL PM PM NM PL PM PM PM ZE ZE PL PM ZE NM NL PM ZE NM NM NL NL PL NM NM NL NL NL

d

Fig.4.7 The membership functions of the fuzzy controller and the controller's rule-base presented as a fuzzy

associative memory (FAM)-matrix.


18

T

NWE

1)( where

)(

)1(

;

)()(

)1()1(

1

1

Nw

N

w

N

ww

n

n

(7)

The term R depends on the learning algorithms (e.g. R=I identity matrix within the

gradient backpropagation). Within the Levenberg-Marquardt algorithm R=H+β*I; H is defined in (8) and β is a scalar term.

..1 T

NH (8)

The proposed fuzzy and neural controllers were firstly tested in simulation, under the

Matlab&Simulink environment and implemented on a real mobile robot with length L=25cm, breadth l=20cm, with two DC motors, one to impose the desired speed and the other controlling the desired steering angle. An encoder mounted on one of the wheel from the pair of free rear wheels provides position and speed. Both controllers have been implemented on a main board APTE 537, with a Siemens microcontroller 80C537 built in.

The Levenberg-Marquardt algorithm aiming to learn the neural network has been applied. We learned several networks with one hidden layer, changing the number of hidden nodes. The resulting criterions with respect to the hidden node number are shown in figure 4.8. We note a diminution of the criterion until about 18 hidden nodes. By augmenting the number of nodes, the learning can bring to local minimums, instead of global minimums. Thus, we chose a neural network with 14 hidden nodes. The control surface of the neural controller is shown in figure 4.9.

The initial mobile robot coordinates are: xc=2m, yc=0.2m, ϴc=0o. In order to prove the

fuzzy and neural controllers effectiveness they have been tested on a sinusoid-like curve trajectory (see figures 4.10).

5 10 15 20 254

4.5

5

5.5

6

6.5

7

7.5

8x 10

-4

Fig.4.8 Sum square error with respect to

the node numbers.

theta

ddd

Fig.4.9 The control surface of the neural controller,

c as a function of dd and d


19

Fig.4.10 Sinusoid-like curve reference path ( __ ) and robot trajectory (>>>) while implementing the fuzzy controller (left side) and neural controller (right side)

A good behavior of the fuzzy and neural controlled robot has been observed while

following sinusoid-like curves or circles imposed trajectories. Anyway, a hamper form trajectory (e.g. used in automotive industry tests) brings in new challenges due to the ninety degrees angles between the lines (trajectories) in a hamper form (see figure 4.11).

Fig.4.11 The hamper form reference path (yellow) and robot trajectory (green) while implementing a neural

controller without prediction horizon

Fig.4.12 The hamper form reference path (red spots) and robot trajectory (green) while implementing a fuzzy (left side) and neural controllers (right side) with a 25 points prediction horizon.


20

In order to overpass that challenge one of the ideas was to take into account future detected trajectory points on the robot trajectory. A prediction horizon of 25 points has been taken into account and the improved results while implementing both fuzzy and neural controllers are shown in figure 4.12. New reference points have been chosen as follow in (9):

n

kki

pointreferenceknn

pointreferencenew

1)(

_*)1(

2__ (9)

Mobile robot control strategies while performing tasks like obstacles avoidance in an

unknown environment have been proposed too [64]. The control strategy has been implemented within a 128 bit Motorola microcontroller kit mounted on a mobile platform. The system requires only a reduced number of proximity infrared sensors. The effectiveness of the implemented control strategy has been tested in our Robotics laboratory and proven during a student contest (e.g. Electromobility 2007- Continental Company, see figure 4.13).

Fig. 4.13a Mobile robots control strategies in obstacle avoidance tasks (top left- mobile robot platform; top right-control strategy while detecting obstacles; bottom – Matlab simulation tests of the proposed control

strategy)

Fig. 4.13b Mobile robots control strategies in obstacle avoidance tasks – tests in a real unknown environment.


21

4.1.2. Steps towards virtual laboratories on mobile robots control Teaching mobile robots control might be improved by creating interactive computer

simulations that can be used within a virtual laboratory. Interactive simulations in Java by implementing Easy Java Simulations (Ejs) concepts, in connection with Matlab and Simulink, have been proposed and implemented (see figure 4.14, [82]). Ejs offers the possibility to build interactive user-interfaces, whose properties are linked to the Matlab/Simulink model variables, an easy way to model automatic control systems and to remote control dynamic systems via Internet.

Fig.4.14 The Simulink model and the Ejs application in trajectory tracking for mobile robots

It relates to Ejs application within the field of remote control for mobile robots. The

main advantage of the implemented method is that the tools offer a simple way to develop interactive simulations in three steps: by describing a Matlab/ Simulink mathematical model, by building an off-the-shelf based user interface, and by connecting Matlab/Simulink model variables to the Ejs interface to remote control a physical system via Internet. The paper [82] which describes the implementation of a three wheeled mobile robot control via Internet by means of Ejs has been awarded with Jean Peperstraete AWARD of EAEEIE (European Association for Education in Electrical and Information Engineering), at EAEEIE2006 Conference.


22

4.1.3. Voice controlled mobile robots Nowadays, the human-machine interaction became a very important research topic

[5], [21], [33], [39]. The communication in a natural language increases the speed of performing tasks and it is user friendly. Practically, any device which interact with the humans can be provided with a voice recognition system making its use easier. Speech recognition (SR), also known as automatic speech recognition or computer speech recognition is a technology which converts spoken words to machine-readable input (e.g. VR stamp-RSC-4128 based technology). Speech recognition applications include voice dialing, call routing, domestic appliance control and content-based spoken audio search, simple data entry, preparation of structured documents, and in aircraft cockpits. In [50] a mobile platform which emulates the automobile cabin controls has been presented. A voice recognition system is implemented in order to control the lights, the steering, the air conditioner and the heating within the vehicle cabin (see fig. 4.15).

Fig.4.15 The hardware connections diagram of the designed mobile platform which emulates the voice controlled devices within an automobile cabin

The voice controlled mobile robot platform contains: 1) The VR Stamp Tool-kit – based on RSC family of microcontrollers. The VR Stamp is a completely modularized, production ready speech recognition system that allows system to speak and hear with minimal development time and low system cost. 2) The Motorola HCS12E128 kit based on the MC9S12E128 microcontroller which implements the overall control strategy. 3) The Sabertooth speed controller – a drive unit engineered and tested to provide superior performance and control, with a 10A continuous current on each Left and Right drive channels.


23

4) A 4WD1 mobile platform equipped with four GHM14 12VDC motors, 200 RPM; the motors on the same side are wired in parallel making the robot system to steer like a tank.

The words spoken by a human operator are recorded into the VR Stamp kit and compared with a predefined vocabulary. A grammar model has been created and engrave data files that describe the individual sounds of each word of the vocabulary and the relations between them. The process of voice recognition (VR) is based on a Hidden Markov Modeling (HMM) to adapt to different users spoken words and uses these data files. When a spoken word is recognized like being part of a predefined vocabulary, the processor sends a 4 bits coded signal towards the Motorola microcontroller. Besides the robot trajectory control the microcontroller reacts to the VR stamp control coded data and perform the required tasks (e.g. Lights ON/OFF, cabin heating, etc.). The results have been published in proceedings of international conferences ([50]) and presented to companies organized events (e.g. Electromobility contest, Continental Company, Iasi, 2009). Furthermore the developed research has been a basis for original applications of voice controlled neuroprostheses ([48], [50] – see figure 4.16). People with disabilities which use artificial devices during their rehabilitative process may benefit by voice controlling them instead of using switches, instrumented crutches, etc. Original research on neuroprosthesis design and control will be addressed within next few subchapters.

Fig.4.16 Neuroprosthesis control based on a voice recognition system

4.2. Neuroprostheses design and control

Functional electrical stimulation (commonly abbreviated as FES) is a technique that uses electrical currents to activate nerves innervating extremities affected by paralysis


24

resulting from spinal cord injury (SCI), head injury, stroke or other neurological disorders, restoring function in people with disabilities. The quality of paraplegics’ lives can be improved by daily standing exercises. Regular standing in spinal cord injured subjects is thought to help in preventing osteoporosis; in preventing contracture by preserving the range of movement at lower limb joints; in improving digestion, respiration and urinary drainage; in reducing the chance of decubitus ulcers by relieving pressure; and contributing to the psychological benefit by enhancing personal esteem. During the last few decades different FES-based control methods that aim to restore standing have been proposed [24], [46], [78], [79], [90], [94]. Nowadays, the main research aims to restore the broken intention-action-perception loop in disabled people due to a central nervous system lesion. For now the FES has been shown to improve muscle tone and researchers are working now to enhance the evidence that it contributes to improvements in functional ability. The devices that control the electrical stimulus delivery over the targeted muscles are termed as neuroprostheses. The current challenges are to provide the patients with a neuroprosthesis to be used at home for daily exercises. Such a device has to be reliable, with only few parameters to be adjusted by the patient and the supplementary equipment (e.g. sensors) to be ease in donning and doffing.

Future perspectives to develop practical FES systems rely on our ability to generate new knowledge and produce new technological components, and most of all on our ability to integrate our knowledge and technologies into useful systems that meet the need of users. In the same time we need to bear our minds that the SCI patients’ priorities are independence, ease of movement, ease of control and being able to do recreational activities as before the injury. Much more, the SCI subjects expressed a strong desire to have the ability to be spontaneous and to participate in new activities without a lot of pre-planning and preparation. Different research groups are currently involved in providing SCI people with FES systems for exercises and recreational activities. FES cycling and rowing are among the most preferred activities, bringing both, cardiopulmonary fitness and fun.

FES systems that respond to the physical movement as initiated by the patient are to be further investigated. If technology of this type is implemented in future neuroprostheses, it will allow the patient to perform more natural body movements while using FES, and it would also allow more accurate patient control. Technologies as Brain-Computer Interface (BCI) may bring new breakthrough within the field of neuroprostheses. Scientists discovered that some people are able to control the intensity of electrical signals emitted from certain parts of their brain. The signals can be recorded at the skull surface by means of electroencephalograms (EEGs). It follows that a SCI subject might be able to control an electrical device that helps his gait by only thinking about it. Of course, the problems are far away to be solved (invasive electrodes tend to move due to the very soft and mobile nature of the brain, models for coordinating muscle movement are not yet the better ones). However, it is a nice idea to think that someday the paraplegics will regain the complete use of their body through functional electrical stimulation.

One step forward towards these aims is related to neuroprostheses test benches (e.g. proposed EMULOBODY [60], [61], [62]) that offers a feeling on what might be the human body induced motion while controlling it by means of a FES-based device. It follows to propose new control methods (e.g. proposed ONZOFF controller [46] – patent license pending) to be embedded within neuroprostheses, as well as the new BCI techniques to close the loop between the disabled people thoughts and the controlled devices (e.g. neuroprostheses [5], [12], [21], [26], [37]). Following the next subchapters all the contributions related to that research path will be emphasized.


25

4.2.1. Neuroprostheses test benches

4.2.1.1. Lower limb neuroprosthesis test benches

The FES-based rehabilitative techniques are not used at their maximum potential and it is estimated that only a small percent of potential users are currently treating their illness by means of a FES-based rehabilitative device. Therefore it is a great interest among the physiotherapists, physicians, and anyone else performing research within the FES field, to learn more about the way to adapt the neuroprostheses parameters for individuals that might benefit from using FES. Existing FES courses (e.g. post university course -“Stimularea electrică funcţională în recuperarea mersului neurologic (FES-based gait rehabilitation)”- ten annual editions- jointly teaching it with Rehabilitation Hospital of Iasi, Gr.T.Popa Medicine and Pharmacy University of Iaşi) requires the presence of the participants and only a limited number of them are able to attend such a course. A web-site that offers FES tutorials, an e-learning method to control a neuroprostheses and much more a set up environment to implement new control strategies would bring benefits for many interested specialists within the FES-based rehabilitative techniques. Prior to develop an e-learning environment to teach neuroprosthesis control, a mechatronic device, that offers a feeling on what might be the human body induced motion while controlling it by means of a FES-based device, is required.

Fig.4.17 The EMULOBODY test bench system of neuroprostheses control

Within the past few projects (e.g. SINPHA D11-068/2008; ARMS D71-095/2008;

NEUROTECH 31CB/2008, IHRG-150/2012, NOVAFES-267/2014, 24-CEEX-I03/2005) which I was leading as director/project responsible we developed test benches aiming to test neuroprostheses control strategies, as well as hybrid FES-mechatronic systems to rehabilitate


26

the upper limb. The EMULOBODY test bench system [61], [62], [63], , contains: the mechatronic device and the programming environment that allows building any control strategy for transfer strategies as sit-to-stand, standing, or stand-to-sit in SCI people (see figure 4.17).

The test bench that contains a robot-like human body has been built around of a Lynxmotion Lynx6 robot structure that uses pulse-proportional servos HITEC HS-422 to impose the angle and the angular velocity for each joint. The HITEC HS-422 constructive characteristics (range: 0 to 180°, voltage: 4.8 - 6.0vdc, torque: 57 oz.-in., speed: 0.16s / 60 degrees) provides the desired speed and torque to control any of the driven joints. A Lynxmotion SSC-32 board which can control up to 32 servos, has been used to control the four servos that are required to emulate the human body movements around the ankle, knee and hip joints. The movement around the ankle joint is driven by two servos, while each of the other two knee and hip joints are driven by one servo. Neuroprosthesis aiming to control a chained motion like standing-up – standing – sitting-down in paraplegics’ people are to be tested on that kind of test bench.

The Simulink model implements a three segmental model with nine mono- and biarticular muscle groups, as described in [91]. These muscle groups are modeled in the sagittal plane inducing moments about the ankle, knee, and hip joints. The user can propose its own control strategy as a Matlab function that will be inserted within the controller red block (see figure 4.17). The controller provides the stimulation parameters for the selected muscle groups in accordance with the measured/computed joints variables (angles, angular velocities and accelerations). Muscle activation, muscle contraction and body segmental dynamics are the three main components of the implemented Matlab&Simulink model which calculates the angles and angular velocities to be prescribed to the robot-like human body joints. These control signals are converted in pulse width parameters for any of the joints servos. The Lynxmotion SSC-32 control board receives the calculated pulse width for any of the servos joints via a PC serial interface, and provides them with the required electrical signal. The animations plotted on the screen can be compared with the robot-like human body motions and both with experimental data obtained during trials on a SCI person that uses a neuroprosthesis to perform transfer motions as standing-up, standing and sitting-down.

For example, while implementing an ONZOFF control strategy (e.g. [46] ISI journal published paper; [60] ISI Web of Science indexed paper), which will be explained in a following subchapter, the results are shown as in figures 4.18-4.21.

Fig.4.18 ONZOFF controller output for the targeted muscles


27

Fig.4.19 Joints angles during a sitting-down motion task (simulation)

Fig.4.20 Pulse width for the driven servos of the EMULOBODY system

Fig.4.21 Sitting-down motion task controlled by means of an ONZOFF controller (Simulink simulation)

and shown by means of a mechatronic device (robot-like human body)

A sitting-down motion task which implements a control based on the ONZOFF

controller (results shown in figures 4.18 to 4.21), has been tested on our test bench (motion sub-phases: Buckle1=2° (knees are unlocked), SitDown=80o (ramp down stimulation to zero), ZONE=70o/s). The form of the switching curve in the second sub-phase (Buckle2) is defined by the maximum knee angular velocity (115o/s) and the Ox intersections (0o and 90o). The pulse width for the gluteal muscles has been considered constant within the subspaces, while the pulse widths for the quadriceps and hamstrings, have been calculated in accordance with


28

the ONZOFF controller diagram (see subchapter 4.2.2.), by choosing PW_quads_max=320 μs and PW_hams_max=150 μs.

The EMULOBODY system brings the following benefits:

offers a simple construction from the constructive, functional and technological point of view;

reduces the number of clinical experiments to be performed while testing new neuroprostheses;

helps the kinetotherapists, medical staff and biomedical engineers to acquire new knowledge in the field of neuroprosthesis control;

helps the control engineers to test new control strategies to be implemented within an embedded system as neuroprosthesis prior to test it on paraplegics, in a clinical environment.

Next research step (e.g. SINPHA D11-068/2008 project, acting as project director) aimed to build a mechatronic lower part of a human body to be embedded within a test bench for neuroprostheses. Usually, a half body robotic structure might be enough to test a neuroprosthesis controller which support the standing-up, standing and sitting-down maneuvers in paraplegia in the sagittal plane. The figure 4.22 shows the designed robotic leg (five revolute joints: 2 for hip, 1 for knee and 2 for ankle) with anthropometric dimensions.

Fig.4.22 The kinematic chain of the robotic leg (left side) and the robot leg with anthropometric dimensions (right side)

The actuating system of the robotic leg has been implemented by means of five

OMRON servomotors and harmonic gearboxes and the control system of the robotic leg with five OMRON servo drivers and Trajexia control unit. A feedback error based neural controller has been proposed in [41] (indexed SPRINGER-VERLAG, book chapter) and its effectiveness has been proven firstly in simulation.

The neuroprostheses control test bench which embeds the anthropometric dimensioned robotic leg is shown in figure 4.23.


29

Fig.4.23 The EMULOBODY2 test bench

The Simulink model implements a three segmental human body model with nine

mono- and biarticular muscle groups, as described in [91]. These muscle groups are modelled in the sagittal plane inducing moments about the ankle, knee, and hip joints. All muscle groups except monoarticular hip flexors can be activated in a real experiment by a proper arrangement of surface electrodes. Each modelled muscle group has its own activation and contraction dynamics. The inputs for the model are the stimulator pulse width and frequency. Muscle activation, muscle contraction and body segmental dynamics are the three main components of the implemented model. The forces computed for any of the nine muscle groups that are activated due to an applied electrical stimulus, are input to the body-segmental dynamics. The interaction (horizontal and vertical reaction forces) with a seat is modelled by means of a pair of nonlinear spring-dampers.

The upper body effort has to be taken into account into any model that aims to support FES-based controllers testing. Within the patient model as developed in [91], shoulder forces and moment representing the patient voluntary arm support are calculated on a basis of a look-up table, as functions deviations of horizontal and vertical shoulder joint position and trunk inclination from the desired values, and their velocities. In fact, the shoulder forces and moment model is based on a reference trajectory of the shoulder position and trunk inclination during the sit-to-stand transfer obtained during an experiment on a sole paraplegic patient. In our case the vertical shoulder forces are modelled as a function of measured knee angles by means of a fuzzy controller (published paper [78], as postdoc within NeuralPRO EU project).

In paraplegia, in the case of standing-up and sitting-down motion tasks we performed open loop control and different closed loop control trials in order to observe the voluntary contribution of the patient (see figure 4.24). Stimulation patterns for quadriceps and hamstrings were applied using surface electrodes.

When performing a SU task motion, the movement is faster (0.8÷1s) and the patient tries to use his/her hand forces to control the movement. In this case we found that vertical shoulder forces can be modelled as a function of measured knee angles (see figure 4.24). Oppositely, when accomplishing a SD motion task the main behavior of the patient relates to the acceleration of the body.


30

Figure 4.24 The SU motion tasks - Experimental recorded hand forces Fsh_y (red-left side, blue – right side,

green – approximated Fsh_y)

At the beginning of SD motion task, knee acceleration increases and the patient reacts

by increasing the hand forces. When the knee acceleration decreases (i.e. due to motion control), confidence of the patient increases and exerted hand forces decrease. In the SD middle and end phase, when patient feels a certain increase in knee acceleration, then he/she reacts, increasing the exerted hand forces. This behavior has been modelled as a fuzzy controller having as inputs (normalised): knee angle Θ and knee acceleration d2Θ/dt2, and as output vertical shoulder forces Fsh_y. It consists in a number of 25 linguistic control rules summarised in table 4.1. The defuzzification method is centroid. The fuzzy control surface is shown in figure 4.25 and the membership functions for inputs and output are shown in figure 4.26.

Θ_Knee d2Θ/dt2_Knee

S

SM

M

MB

B

NB

S

B

B

B

SM

NM

S

B

B

M

S

Z

SM

MB

MB

MB

S

PM

SM

MB

MB

MB

S

PB

SM

B

B

B

SM

Table 1 – The linguistic control rules. The abbreviations used above means: NB - Negative Big; NM – Negative Mediu; Z - Zero; PM – Positive Medium; PB – positive Big; S - Small; SM – Small Medium; M – Medium; MB –

Medium Big; B – Big. (Example of a rule: If [Knee_angle is SM] and [Knee_acceleration(qddot-knee) is NM] then Fsh_y is B.).

0 10 20 30 40 50 60 70 80 90 100-100

-50

0

50

100

150

200

250

300Hand forces (red - left side; blue - right side) - Standing Up

Knee angle [degrees]

Hand f

orc

es O

y [

N]


31

Fig.4.25 The fuzzy controller surface

a

b

c

Fig. 4.26 The membership functions: a) input1- q_knee, b) input2- qddot_knee, c) output- Fsh_y

0 0.5 1 1.5 2 2.5 3 3.5 -100

-50

0

50

100

150

Time [s]

q-

knee

[deg]

qd

ot[de

g/s]

qddot

-kn

ee[de

g/s

/s]

q-knee-stg qdot q-ddot-knee-stg /10

0 0.5 1 1.5 2 2.5 3 3.5 0

50

100

150

200

250

300

Time [s]

Force

[N]

Force-fuz Force-exp

Sit moment

Figure 4.27 – UPPER: Hand forces – blue: recorded within experiments; red: fuzzy controller output. LOWER:

Experimental data: red-knee angle, magenta – knee angular velocity, blue – knee angular acceleration

00.2

0.40.6

0.81

-1

-0.5

0

0.5

1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

q-kneeqddot-knee

Fyhand

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

0

0.2

0.4

0.6

0.8

1

q-knee

Degre

e o

f m

em

bers

hip

S SM M MB B

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1

0

0.2

0.4

0.6

0.8

1

qddot-knee

Degre

e o

f m

em

bers

hip

NB NM Z PM PB

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

0

0.2

0.4

0.6

0.8

1

Fyhand

Degre

e o

f m

em

bers

hip

S SM M MB B


32

The proposed fuzzy controller was validated based on experimental data.

Comparative results between the Fsh_y force prescribed by the proposed fuzzy controller and force recorded in experiments are shown in figure 4.27. The sitting moment was pointed by the fast decreasing of Fsh_y force to zero. We may observe that during the sitting-down motion task the main behavior of the vertical force exerted as voluntary action by the patient was modelled. There are still some errors between trajectories because of the impossibility to exactly distinguish the influence of patient voluntary action and controller in the relation Fsh_y=f(knee_angle, knee_acceleration). The proposed fuzzy controller was found helpful while testing new neuroprostheses control strategies and used accordingly (e.g. [46] ISI journal published paper; First prize and Cup of the Romanian Society of Oral Rehabilitation, at the International Fair INVEST-INVENT2013, Palace Hall, Bucharest, Romania).

4.2.1.2 Upper limb neuroprosthesis test benches and hybrid FES&Mechatronic devices

The intention-action-perception loop has to be closed in order to optimize

rehabilitation and it receives a great research attention nowadays. The user’s intention to perform an action can be detected by means of brain-computer interfaces [21], [26], [33], its action might be supported by means of a hybrid FES and mechatronic hand/arm orthosis [10], [11], [25], [32], [35], [36] and the effects on function and plasticity might be measured with non-invasive brain stimulation [20], [43], [55].

The EXOSLIM project (UEFISCDI PCCA 180/2012; leading as scientific director) deals with an FES-Exoskeleton system (see figure 4.28) embedding a combined method for actuating the human arm by electrically driven exoskeleton and electrical stimulus applied to the upper limb muscles [10], [35].

Fig. 4.28 The EXOSLIM system during laboratory tests (left side), the design (middle) and first clinical trials (right side)

The device is able to provide repetitive exercise training and it is suitable for long term

utilization in daily activities. Due to the ability to replicate forces, velocities and accelerations the device is superior to manual physiotherapy. The novelty of the system represents the combined control between the electrical drives of the exoskeleton and muscles drive through functional electrical stimulation. The complex upper limb movement by means of FES is expected to induce cortical reorganization and therefore to provide a major contribution to the upper limb rehabilitation process. In order to generate electrical stimulus and to induce


33

muscle contraction, a programmable MotionStim8 neurostimulator was used. A Matlab&Simulink GUI allows the control of the entire FES&Exoskeleton system.

Fig.4.29 ProTV news journal broadcasted (April 10th, 19:15, April 11th, 07:15) [96]

The neurostimulator controls up to 8 muscles. In order to operate the exoskeleton,

four DC motors with gear reducer were used. The idea and the EXOSLIM design received great attention during conferences presentations [10] and has been awarded at INVEST-INVENT2014 (diplom and medal), EUROINVENT2014 (gold medal [17], [23]), BringITon14 (mention II, [14]), ProTV journal broadcasted (April 10th, 19:15, April 11th, 07:15; see figure 4.29, [96]). A patent license is pending (OSIM patent license request no.693/15.09.2014).

Far more difficult problems arise while trying to rehabilitate the hand e.g. after stroke. The novelty idea of the SCECM-IEEIA team which I am leading on the PCCA IHRG-150/2012 grant (acting as TUIASI project responsible) is to detect the remaining potential of movement at the hand level (e.g. finger’s twitches), improve the hand movements by electrically stimulation the forearm and over the palm and guide the entire movement by means of a mechatronic glove (see figure 4.30-4.31, ProTV [96], Digi24 TV [97], [11], [12], [19], [22], [25], [36]). The entire research required mathematical modeling under Simulink&Matlab, and balanced control of FES provided by a MotionStim8 neurostimulator and a homemade mechatronic glove actuated with Firgelli linear actuators. Another glove with an adequate sensorial system might be placed on the non-affected hand and the patient can work to replicate the movement on the affected hand. It is expected to induce cortical reorganization and therefore regaining the hand dexterity.

The idea and the IHRG system design received great attention during conferences presentations [11], [22] and has been awarded at EUROINVENT2014 (gold medal [25]) and BringITon14 (mention II, [14]).


34

Fig. 4.30 The IHRG system laboratory tests (together with one of the PhD students, Sergiu Hartopanu, involved in the project research – I have been acting as member within the commission assisting its PhD thesis

defending – September 2015)

Fig.4.31 The actual experimental testing clinical procedure at the Rehabilitation Hospital of Iasi (March 2016) and the ProTV and Digi24 journal news broadcast during a project workshop [96]; [97]

FES uses various devices to produce muscle contractions by applying electrical pulses

to a specific nerve through skin surface or implanted electrodes. In skin surface FES, electrodes placed over the nerve are connected by leads to a stimulator unit and may be triggered with a foot switch (e.g. correcting a drop foot in stroke patients). The main disadvantages related to the conventional electrodes used up-to-date are: employment of disposable electrodes; employment of electro-gel that can cause skin irritation; need for every day positioning of the electrodes; high price; safety is not always guaranteed; the conditions are not always optimal for a maximum effect. Within the NOVAFES project (PCCA 267/2014, leading as project director) we aim at overcoming these shortcomings by developing innovative functional garments with textile electrodes for FES-based


35

rehabilitation with the following advantages: (1) easiest positioning of the electrodes; (2) garment customization to fit the patient and avoid use of gels; (3) easy handling; (4) no skin irritation and local burns; (5) fully reusable textile electrodes, fully integrated in durable, easy care garments. For that purpose, different elastic electrically conductive yarns will be tested and then knitted into textile electrodes. The therapeutic socks and tights with embedded electrodes will be validated by clinical tests performed in a rehabilitation hospital.

For now, one of the NOVAFES project ideas is to produce knitted electrodes within socks and tights (see figure 4.32). It is easier for a patient experiencing the drop foot problem to perform the set-up of a FES system, which improves his/her walking. By applying the knitting technology, we accomplished a tubular jersey containing two knitted electrodes and an electroconductive yarn acquired from the market (Shieldex® 117/17 dtex 2-ply HC+B); the electrodes were placed as following: one textile electrode will cover the head of fibula bone and another is closed to the tibia bone, but placed over the tibial muscle (Autex ISI journal [7], [15]).

Fig.4.32 FES tests of conventional electrodes and the novel knitted ones – improved foot dorsiflexion for stroke patients

A Microstim MS2v2 exercise stimulator (40 Hz, 350 μs) has been used. At about 50 mA

current, the required dorsiflexion has been produced similarly with the existing gel disposable electrodes. Until September 2017 these novel proposed electrodes will follow a clinical testing procedure (e.g. hand rehabilitation; see figure 4.33).

Fig.4.33 FES-based rehabilitation exercises of the hand

The results have been presented in conferences sessions, [16], and published in well-

known ISI indexed journals [7], [15]. The idea and the prototype product “Knitted product with embedded knitted electrodes as neuroprosthesis to rehabilitate the disabled people due to a neuromotor handicap” (OSIM license patent request A00673/21.09.2015) received the


36

second prize and diploma at INVEST-INVENT fair 2015 (Practical ideas session). Anther license patent (Textile electrode with improved conductivity) OSIM request A00787/03.11.2015 is pending too.

4.2.2 Neuroprostheses control strategies Functional electrical stimulation (FES) can be used to provide many functions to the

neurologically impaired population. The basic principle behind this stimulation is the use of small electric signals to act in place of nerves that no longer function. For people who have had an injury to their spinal cord, above the level of about T12, it is still possible to produce muscle contractions to their lower limbs by means of electrical stimulation. As functional application of FES, the lower-extremity mobility restoration in people with paraplegia has focused on achieving limited periods of standing, walking on level surfaces and climbing stairs. The current challenge is to provide the patients with a neuroprostheses that can be used at home for daily exercises. Such a device has to be reliable, with only few parameters to be adjusted by the patient and the supplementary equipment (e.g. sensors) to be ease in donning and doffing.

A prerequisite for walking is standing. FES-based standing in paraplegia has been achieved with relatively simple systems of surface stimulation in both laboratory and clinical settings. A closed-loop control system improves standing by reducing fatigue and increasing patient safety. Successful clinical implementation of FES-induced standing in paraplegia requires that simplicity in all aspects be maximized.

One of the proposed control strategy [46], [71] that aims to help paraplegics in performing transfer actions (e.g. wheel chair-toilet, wheel chair-bed) by providing FES-based standing-up and sitting-down control is the ONZOFF control strategy (see figure 4.34).

Fig.4.34 The ONZOFF controller state-space diagram

The ONZOFF (ON-Zone-OFF) controller [46], works according to a second order

switching curve in state space (see figure 4.34), with a smooth increase or decrease in pulse width (muscle stimulation) between the On and Off boundaries (Zone). Each motion task (standing-up (SU) or sitting-down(SD)) is divided in motion phases. For example, SD consists of three motion phases: Buckle1 (knee unlocking by shortly stimulating the hamstrings muscles), Buckle2 (controlled lowering) and SitDown (switching off stimulation). In phase Buckle2:

1. Gluteus is stimulated with constant pulse width.

-10 0 10 20 30 40 50 60 70 80 90 1000

20

40

60

80

100

120

Knee angle [degrees]

Kn

ee

an

gu

lar

ve

loc

ity

[d

eg

ree

s/s

]

Buckle1 Buckle2 SitDown

ON

ZONE

OFF

Zw


37

2. Quadriceps are activated at maximum pulse width into the ON subspace, at smoothly decreasing pulse width (depending of actual position into the zone) in ZONE and at minimum pulse width into the OFF subspace.

The overall ONZOFF control strategy is shown in figure 4.35.

Fig.4.35 The ONZOFF control strategy (left) and the CLSTDSD application program (right side) In the case of using the ONZOFF controller (see figure 4.35) there are only four main

parameters to be tuned: zone width Zw, maximum knee angular velocity KVmax of the switching curve and motion phase limits, Bkl1, SD. The pulse width levels for each group of stimulated muscles (quadriceps, gluteals, and hamstrings) for different subspaces can be determined for every patient and then maintained constant.

The ONZOFF controller has been tested in simulation and clinically. In [91], a model that describes the major properties of muscle and segmental dynamics occurring during FES, has been designed. Nine mono- and bi-articular muscle groups are modeled (Matlab&Simulink) in the sagittal plane inducing moments about the ankle, knee and hip joints due to surface electrical stimulation. The model has been improved with a fuzzy controller to model shoulder forces [78].

A novel CLSTDSD application program (see figure 4.35) has been implemented within a 8-channel Stanmore Stimulator [72]. The Stanmore Stimulator is a microcontroller-based (Motorola MC68HC11G5) electrical stimulator with eight stimulation channels. The CLSTDSD.C application program supports a chained motion standing up, standing and sitting down in paraplegia. The standing-up (SU) motion task has been accomplished with satisfactory performances by ramping up the pulse width of knee extensors. Any of the PID-based [92] or KEC [80] control strategies can be chosen for standing. The aim of the PID controller is to maintain the required knee angle by using the error in the knee angle to adjust the quadriceps stimulation. The KEC uses knee angle as input and according to three discrete states BUCKLE, EXTENSION and HYPEREXTENSION the stimulation is rapidly increased, maintained constant or slowly decreased. Within the SD motion task three FES-based control strategies are taken into account: open-loop control (ramping-down the pulse width of stimulated group of muscle), On/Off control [95] and ONZOFF control [46]. The proposed application program allows choosing any control strategy for a particular motion task within a chained motion SU-Standing-SD. Only the knee angles measurements are required as feedback for the system. The knee angles were measured using servo-potentiometers housed in a neoprene knee cuffs. This makes it particularly useful for daily exercises at home. The


38

patient that tested the system (see figure 4.36), has been able to place by himself the electrodes and the sensors housed in a neoprene knee cuffs.

Fig. 4.36 A SCI subject that uses the Stanmore Stimulator (Lesion: T7 –motor complete, sensory incomplete, YPI: 4, Age: 36)

The ONZOFF idea and the clinical test results have been presented in conferences (e.g.

[71] plenary session, [58], [66], [67], [70], [72]), published in ISI journals [46], [47], and has been awarded with CYBERLIFE AWARD and gold medal (EUROINVENT2014) [23]. The license patent is pending (OSIM request patent RO-129704-A2/29.08.2014; International Patent Classification: A61N-005/08 – ISI Web of Knowledge).

A slightly different controller aiming to support FES-based standing-up and sitting-down has been presented in [55], [62] and tested on the EMULOBODY test bench. The ONZOFF controller [46] uses two second-order switching curves in the state-space of knee angle against knee angular velocity. These two curves define a zone between the ON and OFF subspaces, where slightly decreasing or increasing pulse width for the quadriceps and hamstrings muscles can be taken into account during the sitting down phase. Sitting down is therefore divided into three motion sub-phases: Buckle1 (knees are unlocked), Buckle2 (controlled lowering) and SitDown (ramp down stimulation to zero). When sitting down has been initiated, i.e. the start of Buckle1, stimulation levels to the quadriceps and gluteal muscles ramp down to a minimum and simultaneously the hamstring muscles ramp up to maximum in a set time; these levels are then maintained. In the region Buckle2, the gluteal muscles are stimulated to increase hip stiffness. Quadriceps muscles are activated at maximum pulse width into the ON subspace, at smoothly decreasing pulse width (depending of actual position into the zone in the velocity direction) in ZONE and at minimum pulse width into the OFF subspace. The hamstrings are activated at minimum pulse width in the ON subspace, at smoothly increasing pulse width in ZONE and at maximum pulse width in the OFF subspace.


39

Fig.4.37 The state-space diagram for knee angle against knee angular velocity

The form of the switching curve in the second sub-phase (Buckle2) is defined by the

maximum knee angular velocity v_max and the Ox intersections (ka_s and ka_t, i.e. ka_s=0o, ka_t=90o). The modified algorithm (see figure 4.37) calculates the curve radius R and the pulse width to stimulate the muscles are calculated according to the difference between the radius and the distance d1 to any instantaneous point on the state space trajectory (see equations (10) and (11)).

max_v2

max_vs_kat_kaR

22

(10)

2

2

12

s_kat_kaang_KneeRmax_vvel_ang_Kneed

(11)

The Knee_ang_vel and Knee_ang are the instantaneous values of the knee angular velocity and knee angle, respectively, at any generic point P1 on the state space trajectory. Within the next step, the maximum values of the pulse width (PW_quads_max, PW_glu_maxt, PW_hams_max) which produce a sustained contraction for quadriceps, gluteal and hamstrings muscles are imposed. For a patient they are obtained during some clinical trials with a neurostimulator. For example, in accordance with the proposed algorithm, the quadriceps muscles are activated at maximum pulse width PW_quads_max into the ON subspace (d1≥R), and at minimum pulse width (≈ 0 μs) into the OFF subspace (d1≤R-Zw). For the trajectory points situated within the ZONE subspace the pulse width can be calculated as follow:

w1

w

ZRdZ

max_quads_PWquads_PW

(12) The same methodology can be applied to calculate the pulse width to be applied to

the hamstrings muscles. Within the ZONE subspace it can be calculated as follow:


40

1

w

dRZ

max_hams_PWhams_PW

(13) The testing results are similar with the one obtained with the ONZOFF controller but

it brings in an easier way to understand the controller and its tuning parameters.

4.2.3 BCI techniques in neuroprostheses Brain-computer interface (BCI) systems are tools that can provide communication

without movement. Unlike conventional means of communication, such as speech or gesture, and interfaces like mice or keyboards, BCIs rely on direct measures of brain activity [5]. EEG-based motor imagery BCIs can discern whether a user is imagining left vs. right hand movement. Thus, they may be helpful for rehabilitation of motor disabilities, when patients are typically required to imagine moving either hand or other movements. After calculating a classifier, the system can detect which limb was imaginary moved by the user and use that information to control for example a neurostimulator which activates the paralyzed upper limb. The proposed idea on how to involve the BCI within a FES-based rehabilitation process is presented in figure 4.38 and has been the basis for our research and several published papers and presentation in renowned conferences (Congress of Romanian Society of Neurology [19], [21], IFAC [39], WPA2015 [12], CYBERLIFE AWARD- Future Medical Devices controlled by means of Brain-Computer Interface – EUROINVENT2014 [23]).

If some steps towards understanding the EEG signals from the brain, collecting and treating them to obtain external devices controls, have been performed, the so-called brain computer “input” interfaces (which can send sensory information to the brain) are still in their infancy. The actual knowledge about how electric impulses in the nerves and brain are translated into the discrete sensations of vision, hearing, touch, and proprioception that we experience in everyday life are very limited and that is a research challenge for the years to come. Human-enhancing BCI will develop mainly in three phases:

Normal human capabilities will be surpassed by means of medical devices which conventionally replace normal human functions (e.g. military people with robotic legs/arms; eye prostheses enhancing eyesight);

Implanted medical devices with novel functions (e.g. visual prostheses allowing people to see in infrared spectra);

Nano-electronic interfaces implanted onto the brain to allow input-output communication (e.g. distributed wireless 3D implants). A BCI system has an input, the output and a signal processing algorithm that maps the

inputs to the output. The major strategies that are considered for the input of a BCI system are: the P300 wave of event related potentials (ERP), steady state visual evoked potential (SSVEP), slow cortical potentials and motor imagery (MI). The EEG is usually measured non-invasively, with electrodes mounted on the human scalp using conductive gel. We mainly used the MI-based BCIs method aiming to induce neural plasticity and thus serve as an important tool to enhance motor rehabilitation for different kind of patients [5].

For the effective training and use of a MI-based BCI, a crucial role in optimizing the user’s performance is played by the neurofeedback. Furthermore, the feedback provided to the user must reflect the user’s task in an appropriate way. When using the BCI for robotics


41

applications control, the feedback strategy can also initiate a pre-programmed motion task for a robot (e.g. control of a robotic arm).

Fig. 4.38 The proposed BCI & FES-based rehabilitation process

We evaluated the influence of choosing different time windows for calculating the CSP

and two different feedback strategies that can be used to initiate a robotic movement task [39]. The method of common spatial patterns (CSP) is used to discriminate of two motor imagery tasks. This method weights each electrode according to its importance for the discrimination task and suppresses noise in individual channels by using correlations between neighboring electrodes. The EEG data were recorded using 63 active electrodes (g.LADYbird, g.tec medical engineering GmbH, Austria) overlying the sensorimotor areas, which are activated during right- and left-hand movement imagination. The electrode positions are shown in figure 4.39, and the electrodes 29 and 33 correspond to C3 and C4 of the international 10-20 system.

Fig.4.39 The EEG electrodes positioning system (g.LADYbird electrodes mounted on a GTEC cap)


42

Two subjects (26-30 years old, male) participated in this study [39]. One of them was

left-handed, one right-handed and both free of medication and central nervous abnormality. Two sessions with different feedback were executed. The figure 4.40 presents the workflow within one session.

Fig.4.40 The workflow within one session (left side) and the timing within one trial (right side)

For each session and chosen window-length segment a map of the most significant

CSP has been built, as shown in figure 4.41.

Fig.4.41 An example of the maps of the most significant CSP (subject S1, session 2, time length: 3s, from 4s to

7s)


43

Based on the performed experiments we suggest using windows of four seconds,

especially it turned out that the window from 4s to 8s was the best one for the used paradigm. The longer the window is, the more important is of course to do an accurate artifact correction. This means that when taking a window of e.g. only one second then only artifacts that are within this specific second are important, when taking a window that covers almost the whole feedback period, then this whole period needs to be artifact-free.

Fig.4.42 One training session with recoveriX. The patient has FES pads on the forearms, which are connected to an FES control system in the left side of the picture. The amplifier is shown on the right. The monitor shows an arrow extending to the right, reflecting that the patient is imagining right hand movement. Concordantly, the

FES system is stimulating the right forearm, causing right wrist dorsiflexion [5].

Next major contribution relates to a FES&BCI system named RecoveriX which aims to

improve the upper limb rehabilitation in stroke patients. Tests have been performed together with the Rehabilitation Hospital of Iasi and the system will be marketable by means of the Austrian g.tec Medical Engineering GmbH company ([5]; see figure 4.42).

4.3 Contributions to the EU Higher Education programmes oriented to the Renewable Energies and ICT Securities technical challenges – The EU SALEIE project

“The aims and objectives of the SALEIE project were to investigate and explore some

of the challenges that face European Higher Education in Electrical and Information Engineering (EIE) in meeting the demands placed upon it now and into the future”7[4]. The

7 “Strategic Alignment of Electrical and Information Engineering in European Higher Education Institutions (SALEIE)” No. 527877-LLP-1-2012-1-UK-ERASMUS-ENW, 2012-2015, Coodinator: University of York, UK; Project responsible P12-TUIASI: Marian Poboroniuc (WP3 leader).


44

consortium comprised 44 European partners and 1 from Russia, and as member of the P12-TUIASI team I was leading the WP3-Technical Challenges.

Fig.4.43 The SALEIE Project Workpackages and Main Themes [4] We have been mainly concerned with:

The skills required of graduates able to help industry to respond to the current global technical challenges;

Programmes and modules that develop these key skills, the Institution offering them, their technical content and level of development.

In strength connection with WP4-Widening Participation and WP5-Governance or Policy (see figure 4.43), we addressed also:

The volume and types of learners with specific needs that are currently registered on EIE programmes across Europe.

Equal opportunities and diversity policies and practices.

The level and types of support systems in place for these students.

Policy and practices associated with programme and module specification including how well understood current specifications are to ERASMUS exchange partners and employers.

The right skills for employability are a compulsory requirement which the European education and training systems face in order to provide the learners with experience closer to the reality of the working environment. Among the objectives related to better education and training systems a headline target of 40% of young people completing higher education is envisaged at the European level. By 2020, 20% more jobs will require higher level skills. Education needs to drive up both standards and levels of achievement to match this demand, as well as encourage the transversal skills needed to ensure young people are able to be entrepreneurial and adapt to the increasingly inevitable changes in the labour market during their career.


45

Past project activity and much ongoing activity is developing specific modules, building our understanding of academic provision, the quality systems being followed, and ‘tuning’ programmes to meet the needs of industry, students, academics and graduates. A gap in this understanding is in a common understanding of how EIE can respond to the key global technical challenges and maximizing the pool of potential learners through the widening participation and student with specific needs agendas.

One of the main challenges addressed by the SALEIE project has been to ensure that graduates are prepared to enable Europe to respond to the current global technical challenges in the Green Energy the Environment and Sustainability, Communications and IT, Health, and Modern Manufacturing Systems (including Robotics), that is, a “new skills for new jobs” approach. This embrace conventional education, lifelong learning and training for entrepreneurship.

Based on the surveys, partners meetings, and participation to specific conferences ([9], [27], [28], [30], [34]) we came up with some useful conclusions that lead us to propose new curricula within Renewable Energies and ICT Security. The first few findings can be summarized as [3]:

A. A general agreement on the degree structures has been reached during the Bologna process. First degrees should require 180 to 240 credits (ECTS)8 (equivalent to 3 to 4 years fulltime) and the Masters should require 90 to 120 credits (ECTS)9 after the first degree, with a minimum of 60 credits at Master level. Based on the national reports for 2012 (see figure 4.2) related to the Bologna Process – EHEA10 an average of 66.78% of first cycle study programmes falls within the ones requiring 180ECTS. Some countries didn’t provide such information (e.g. Estonia, Spain). According to the same national reports for 2012 the responses to the question 5.7 (Please provide the (approximate) percentage of second cycle (master) programmes of the following length: 60÷75 ECTS, 90ECTS, 120ECTS, other) most of the second cycle programmes falls among those with 120ECTS (average of 56.18%). The percentages distribution against countries is presented in figure 4.3. To conclude, most of the first cycle programmes (Bachelor) accounts for 180 ECTS (European Credits Transfer System) and the main option for the Master level is for 120 ECTS.

B. A basic structure for a curriculum for three years Bachelor Degree in Electrical Engineering and Master perspectives have been presented in chapter 4.1. It represents a starting point for designing EIE programmes that respond to key global technical challenges.

C. EIE education is mainly supported by means of programmes in: a. Systems Engineering, Systems and Control, Computer and Systems

Engineering b. Biomedical Engineering c. Power Engineering, Renewable Energy Systems Technology

8 Conclusions and Recommendations of the International Seminar on Bachelor-Level Degrees, Helsinki, February 2001. 9 Conclusions and Recommendations of the Conference on Master-Level Degrees, Helsinki, March 2003. 10 http://www.ehea.info/article-details.aspx?ArticleId=86 : Bologna process – European Higher Education Area: responses to Q5.1: Please provide the (approximate) percentages of first cycle study programmes across the following categories: 180ECTS, 240ECTS, Other.

http://www.ehea.info/article-details.aspx?ArticleId=86


46

D. Within all the above mentioned programmes have been traced specific elements that accounts for some responses to the key global technical challenges. The adopted methodology was to enlist a general form of such a programme and then to come up with different examples from different universities that provide a certain response to key challenges by means of certain modules or courses embedded within the programme.

E. Most of the modules/courses that claim to respond to key challenges are related to some research that is performed within the host University/ Faculty/ Department.

F. Despite the claim that some programmes may respond to key challenges that happens only to a limited degree while only some courses (sometime modules) contains specific elements that accounts for a response to key global technical challenges. Based on the survey findings it is to conclude that next step within the SALEIE work should account for designing EIE model programme and module curricula with a wider response to the current global technical challenge subjects.

A survey of existing EIE programmes in the key challenge areas has been released and its results have been discussed during the workshops. The industry feedback related to the required EIE technical and non-technical skills [6] has been fed within the project deliverables and finally two new proposed curricula on Renewable Energy (RE) and Information and Computer Technology (ICT) Security saw the daylight [4], [9].

Within the EU countries there is a diversity in defining and understanding terms as modules, courses, programme and therefore the WP3 experts group manage to come-up with the following definitions ([3], [6], [27]) in order to provide a clear understanding of the proposed curriculum:

Curriculum: The aggregate of modules of study given in a learning environment. The modules are arranged in a sequence;

Syllabus: Is an outline and summary of topics to be covered in an education or training programme;

Programme: A plan of modules to be covered to achieve a specific degree and/or qualification;

Module: Lectures, labs and other activities related to one topic. Given the focus on the Renewable Energies and ICT Securities technical challenges,

model curricula were designed for FCD and SCD programmes for each challenge area. The Bachelor structure has to embed some fundamental, widely adopted modules (e.g. fundamentals in Mathematics, Applied Physics, Circuit Theory), accounting for 180ECTS over six semesters. However, during the final studies the Bachelor’s graduates will already get an inside within, for example, Renewable Energy topics (e.g. Foundations on Renewable Energy, Fabrication Technologies, Transmission and distribution systems, Protections in power systems).

The proposed structure of the Renewable Energy curriculum for the Bachelor (180 ECTS) is shown in figure 4.44 and the one for the Master in figure 4.45 [9], [28].


47

Fig.4.44 The proposed structure of the Renewable Energy curriculum for the Bachelor (180 ECTS)

Fig.4.45 The proposed structure of the Renewable Energy curriculum for the Master (120 ECTS)

The model curriculum for the Bachelor level programme in ICT Securities is shown in

Figure 4.46. The model curriculum for the Master level programme in ICT Securities is shown in Figure 4.47 (see [4]).


48

Fig.4.46 The ICT Security Bachelor model curriculum [98]


49

Fig.4.46 The ICT Security Master model curriculum [99]

Each module description form contains the following points:

Module name;

Programme (e.g. Energy or ICT);

ECTS (European Credit Transfer and Accumulation System) number;

Type: Bachelor or Master;

Scope and form;

Duration (e.g. weeks, hours/week leading to a proper counting of hours of student workload);

Type of assessment;

Qualified prerequisites;

General module objectives;

Topics and short description;

Learning outcomes (embed knowledge, skills and competences);

Module recommended literature;

Other comments. A number of five modules were selected for evaluation (e.g. ICT Security Management

(Master and Bachelor), Renewable Energies Analysis and Simulation of Electrical Systems,


50

Renewable Energies Wind Energy Generation and Transmission, Renewable Energies Fuel Cells Energy). The main findings are positive and can be found on the SALEIE website11 of final report [4]. For example, “Renewable Energies Analysis and Simulation of Electrical Systems was found to be relevant to: Modern global technical challenges in Power Systems, New paradigms of the Electrical Power Systems, Renewable Energy and Uncertainties in new Power System. The content (including module title) is attractive to industry and complements European Union (EU) economic development and employability of the graduates. There is a good balance between the breadth and depth of content covered. The different technical aspects are suitably addressed and at the right academic level. The industry input and support for projects and internships has to be strengthened. This is the first module in the Power System area where the students study the system in a global way. The development of soft skills in the students is stimulated” [4].

5. Carrier development directions that require the Habilitation

During the past more than twenty years I have acquired both teaching and research experience in fields as robotics, systems theory and control, system identification, neuroprostheses control, brain-computer interfaces etc. Most of the work has been done as a team member aiming to share my expertise with the other partners. The habilitation would give me a new opportunity and frame to share all my knowledge, by supervising PhD students while involving them in new national and international projects and to explore new research fields (e.g. new submitted EU projects: NEURON9-07312; MoveAGAIN-ITN13). Main goals will target:

Development and experimental validation of new control methods in mobile robotics including new concepts as voice control and brain-computer interfaces;

Applications of robotic arms control to support hybrid FES-exoskeletons systems aiming to improve rehabilitation in disabled people;

New applications of brain-computer interfaces as integrated within a closed loop intention-action-perception in order to optimize rehabilitation.

Any new frontier application of mechatronics-neuroprostheses-BCI which ideas will emerge from the proposed research proposals.

11 http://www.saleie.co.uk/ [accessed on April 14th, 2016]. 12 NEURON9-073- ERA-NET RA-NET NEURON - European Research Projects on External Insults to the Nervous System, GRIP: Grasp Rehabilitation through Intention and Plasticity — Closed-loop aided neuromodulation to optimize corticospinal plasticity for rehabilitation after spinal cord injury – submitted on 14.03.2016; 13 MoveAGAIN-ITN, Marie-Curie RTN, submitted on Janury 12th, 2016.



51

Annex 1 - References (selection) 1. Knitted product as orthesis, with knitted embedded electrodes, to rehabilitate the upper and

lower limbs in disabled people due to a CNS lesion ( Produs tricotat, tip orteză, cu electrozi incorporaţi prin tricotare, pentru recuperarea membrelor, la persoane cu handicap neuromotor), License patent request OSIM nr.A00673/21.09.2015; Targul International de Inventii si Idei Practice - InvestInvent 2015, Iasi, Romania, premiul II – Sectia de idei practice, (Poboroniuc et al., TUIASI, SC Magnum SRL, SC RO-GALU SRL; results of the NOVAFES 267/2014 UEFISCDI granted project).

2. Textile electrode with improved conductivity (Electrod textil cu conductivitate imbunatatita), Licence patent request OSIM-A00787/03.11.2015, (Poboroniuc et al., TUIASI, SC Magnum SRL, SC RO-GALU SRL; results of the NOVAFES 267/2014 UEFISCDI granted project).

3. Poboroniuc M., Ward A., et all., Report on a survey of existing European Higher Education Programmes orientated to the Renewable Energies and ICT Securities Technical Challenges, SALEIE book chapter, pp.1-67, published by The University of York, Department of Electronics, Heslington, York, YO10 5DD, UK, ISBN 978-1-85911-032-4, February 2016.

4. ***- SALEIE final report, SALEIE book chapter, pp.1-71, published by The University of York, Department of Electronics, Heslington, York, YO10 5DD, UK, ISBN 978-1-85911-030-0, February 2016.

5. Irimia D., Sabathiel N., Ortner R., Poboroniuc M., Allison B.Z., Coon W., Guger C., RecoveriX: A new BCI-based technology for persons with stroke, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, to be held in Orlando, Florida, USA, August 16-20,2016 (in press).

6. Poboroniuc M. et all., Report on existing programmes orientated to key challenge areas, SALEIE EU project, WP3 deliverables, http://www.saleie.co.uk/ [on-line reference, accessed on February 26th, 2016].

7. Curteza A., Cretu V., Macovei L., Poboroniuc M., The manufacturing of textile products with incorporated electrodes, AUTEX Research Journal, Volume 16, Issue 1, Pages 13–18, ISSN (Online) 2300-0929, DOI: 10.1515/aut-2015-0049, April 2016.

8. Serea F., Hartopanu S., Poboroniuc M.S., Irimia D.C., The development and tests of a new hybrid FES-Exoskeleton upper limb assisting device for disabled patients, Proc. of the 10th International Conference on Electromechanical and Power Systems, pp.152-157, 8-9 October 2015, Chisinau, Republic of Moldavia, ISBN 978-606-567-284-0.

9. Poboroniuc M.S., Gheorghe Livint, F. Maciel Barbosa, Wojciech Mysiński, Anna Friesel, Bahar Karaoglan, Yoana Ruseva, Dorin Popescu, Tomislav Kilic, Tony Ward, Noel Jackson, Ian Grout. Developing New Electrical and Information Engineering Related Curricula to Respond to the Actual Global Challenges: The Renewable Energy Curriculum. Proceedings of the EAEEIE2015 conference, 1-2 July 2015, Copenhagen, Denmark.

10. Florin Serea, Poboroniuc M. S., Sergiu Hartopanu, Danut Irimia, Towards Clinical Implementation of an FES&Exoskeleton to Rehabilitate the Upper Limb in Disabled Patients, 2015, International Conference on Control Systems and Computer Science (CSCS), May 2015, Bucharest, Romania, pp.827-832, DOI: 10.1109/CSCS.2015.114, indexed IEEExplore.

11. Hartopanu S., Poboroniuc M. S, Serea F., Irimia D.C., Livint G. New Issues on FES and Robotic Glove Device to Improve the Hand Rehabilitation in Stroke Patients, Proc. of 6th International Conference on Modern Power System 2015, 18-21 May 2015, Cluj Napoca, Romania, in Acta Electrotehnica, vol.56, No.3, pp.123-127, 2015, ISSN 1841-3323, ISSN 2344-5637, indexed Google Scholar.

12. Irimia D.C., Ignat B.E., Popescu D.C., Poboroniuc M. S, Engineering the BCI&FES systems to improve rehabilitation in disabled people due to the central nervous system disorders, WPA2015 International Congress “Primary Care Mental Health: Innovation and Transdisciplinarity”, IT Session II, June 24th-27th, 2015, Bucharest, Romania, in Romanian Journal of Psychiatry- Abstracts, vol. XVII, No.2, June2015, pp.161, 2015, p-ISSN 1454-7848, e-



52

ISSN 2068-7176, indexed CNCSIS B+, 18 European CME credits certificate, 10 EMC credits (address 4766/18.06.2015).

13. Poboroniuc S.M., Bulboaca A.C., Irimia D.C., Bulboaca A.E., Olaru R., The EXOSLIM hybrid mechatronic-neuroprosthesis system to rehabilitate the upper limb in people with lesion at the CNS level (Sistem hibrid mecatronic-neuroproteză EXOSLIM pentru recuperarea brațului la persoanele cu handicap neuromotor), OSIM Licence patent request no.693/15.09.2014.

14. Sergiu Hartopanu, Florin Serea, Danut Irimia, Poboroniuc M. S., Advanced devices based on FES, BCI and Exoskeletons to improve rehabilitation in hemiplegia, 2014 Catalogue of the Workshop BringITOn, pp.30-31, 14-15 November, Iasi, Romania, 2014, ISSN 2285-0929. Mentiune II.

15. Curteza A., Cretu V., Macovei L., Poboroniuc M. S., 2014, Designing functional clothes for persons with locomotor disabilities, AUTEX Research Journal, DOI: 10.2478/aut-2014-0028©AUTEX, ISSN 2300-0929 (On-line), http://www.autexrj.com, (impact factor 0.618 for 2013), Volume 14, Issue 4, pp.281-289, indexed Google Scholar.

16. Poboroniuc M. S., Curteza A., Cretu V., Macovei L., Designing wearable textile structures with embedded conductive yarns and testing their heating properties, in Proceedings of the 8th International Conference and Exposition on Electrical and Power Engineering, IEEE Catalog Number CFP-1447S-USB, Iasi, Romania, ISSN: 978-1-4799-5848-1, DOI: 10.1109/ICEPE.2014.6970016, pp. 778-783, 16-18 October 2014, indexed IEEExplore.

17. Serea F., Hartopanu S., Poboroniuc M. S., New upper limb rehabilitation method in paralyzed people by means of FES and exoskeletons, in Proceedings of the 6th Edition of EUROINVENT - European Exhibition of Creativity and Innovation 2014, Editors: A.V. Sandu& I.Sandu, Alexandru Ioan Cuza Publishing House, ISBN: 978-606-714-037-8, RO54/pp.207, GOLD MEDAL award.

18. Irimia D.C., Poboroniuc M. S., Pasol I., Ortner R., Correlations Between Muscular Contraction Type and Muscle Electrical Activity, in Proceedings of the 8th International Conference and Exposition on Electrical and Power Engineering, IEEE Catalog Number CFP-1447S-USB, Iasi, Romania, ISSN: 978-1-4799-5848-1, 2014, DOI: 10.1109/ICEPE.2014.6969956 ,pp. 488-491, 16-18 October 2014, indexed IEEExplore.

19. Poboroniuc M. S., Popescu C.D., Irimia D.C., Ignat B.E., Bolbocean O.,2014, Enhancing motor recovery in stroke patients by means of FES&Exoskeleton systems, The 12th Congress of the Romanian Society of Neurology, Bucharest, Romania, 14-17 May, 2014, Romanian Journal of Neurology, Vol.XIII, Supl.1, pp. 17-18, ISSN: 1843-8148; ISSN online: 2069-6094, ISSN-L 1843-8148, Journal with CNCSIS B+ accreditation; journal listed on: EMBASE, EBSCO, Scirius, WAME.

20. Ignat B.E., Alexa D., Filos C., Bolbocean O., Poboroniuc M. S., Popescu C.D., ,2014, Impact of FES training on gait parameters of stroke survivors, The 12th Congress of the Romanian Society of Neurology, Bucharest, Romania, 14-17 May, 2014, Romanian Journal of Neurology, Vol.XIII, Supl.1, pp. 14, ISSN: 1843-8148; ISSN online: 2069-6094, ISSN-L 1843-8148, Journal with CNCSIS B+ accreditation; journal listed on: EMBASE, EBSCO, Scirius, WAME.

21. Irimia D.C., Popescu C.D., Poboroniuc M. S., Ignat B.E., Bolbocean O., 2014, Using a Motor Imagery based-BCI system for neuroprosthesis control, The 12th Congress of the Romanian Society of Neurology, Bucharest, Romania, 14-17 May, 2014, Romanian Journal of Neurology, Vol.XIII, Supl.1, pp. 15-16, ISSN: 1843-8148; ISSN online: 2069-6094, ISSN-L 1843-8148, Journal with CNCSIS B+ accreditation; journal listed on: EMBASE, EBSCO, Scirius, WAME.

22. Hartopanu S., Poboroniuc M. S., Serea F., Livint G., Towards human arm rehabilitation in stroke patients by means of a hybrid FES&robotic glove, in Proceedings of the 8th International Conference and Exposition on Electrical and Power Engineering, IEEE Catalog Number CFP-1447S-USB, Iasi, Romania, ISSN: 978-1-4799-5848-1, 2014, DOI: 10.1109/ICEPE.2014.6969886 , pp. 148-152, 16-18 October 2014, indexed IEEExplore.

23. Poboroniuc M. S., ONZOFF control method to support Functional Electrical Stimulation-based standing in paraplegia, in Proceedings of the 6th Edition of EUROINVENT - European


53

Exhibition of Creativity and Innovation 2014, Editors: A.V. Sandu& I.Sandu, Alexandru Ioan Cuza Publishing House, ISBN: 978-606-714-037-8, RO54/pp.206, GOLD MEDAL award & CYBERLIFE AWARD : Poboroniuc M. S., Irimia D.C., Serea F., Hartopanu S., Future Medical Devices controlled by means of Brain-Computer Interface. Euroinvent2014.

24. Poboroniuc M., ONZOFF method for controlling the motion of raising to upright position and sitting based on electrical functional stimulation applicable to patients with neuromotor impairment, Licence patent request RO-129704-A2 / 29.08.2014; International Patent Classification: A61N-005/08;(Web of Science& THOMSON REUTERS).

25. Hartopanu S., Serea F., Poboroniuc M. S., A new rehabilitation method based on a hybrid FES-mechatronic intelligent robotic glove, in Proceedings of the 6th Edition of EUROINVENT - European Exhibition of Creativity and Innovation 2014, Editors: A.V. Sandu& I.Sandu, Alexandru Ioan Cuza Publishing House, ISBN: 978-606-714-037-8, RO54/pp.204, GOLD MEDAL award.

26. Irimia D.C., Poboroniuc M. S., Facilitating cortical reorganization in stroke patients by means of a Brain-Computer Interface &FES hybrid system, in Proceedings of the 6th Edition of EUROINVENT - European Exhibition of Creativity and Innovation 2014, Editors: A.V. Sandu& I.Sandu, Alexandru Ioan Cuza Publishing House, ISBN: 978-606-714-037-8, RO54/pp.204, GOLD MEDAL award.

27. Poboroniuc M. S., Livint G., Pablo J.J.M., Friesel A., Grindei L., Ward A., SALEIE: An EU project aiming to propose new EIE curricula oriented to key global technical challenges, in Proceedings of the 8th International Conference and Exposition on Electrical and Power Engineering, IEEE Catalog Number CFP-1447S-USB, Iasi, Romania, ISSN: 978-1-4799-5848-1, 2014, DOI: 10.1109/ICEPE.2014.6969897, pp. 198-203, 16-18 October 2014, indexed IEEExplore.

28. Poboroniuc M. S., Livint G., Friesel A., Cojocaru D., Popescu D., Grindei L., Naaji A., Ward A., Trends and EIE higher education response to the current global technical challenges, in Proceedings of the 25th International Conference on European Association for Education in Electrical and Information Engineering, EAEEIE 2014, Izmir, Turkey, 30 May-1st June 2014, Article number 6879388, Pages 65-68, Category numberCFP1496D-ART; Code 107220, DOI: 10.1109/EAEEIE.2014.6879388, indexed Scopus & IEEE Xplore.

29. Cojocaru D., Popescu D., Poboroniuc M. S., Ward T., Educational Policies in European Engineering Higher Education System - Implementation of a survey, in Proceedings of the 2014 IEEE Global Engineering Education Conference EDUCON2014, Book Series: IEEE Global Engineering Education Conference, pp.229-234, ISBN:978-1-4799-3190-3, ISSN: 2165-9567, Istanbul, Turkey, 3-5 April 2014, DOI: 10.1109/EDUCON.2014.6826096, indexed ISI Web of Science (WOS:000343764100039)& IEEExplore.

30. Friesel A., Ward A., Mrozek Z., Poboroniuc M. S., Welzer T., Building a shared understanding of the skills and competences in order to respond to the current global technical challenges, in Proceedings of the 2014 IEEE Global Engineering Education Conference EDUCON2014, Book Series: IEEE Global Engineering Education Conference, pp.676-679, ISBN:978-1-4799-3190-3, ISSN: 2165-9567, Istanbul, Turkey, 3-5 April 2014, DOI: 10.1109/EDUCON.2014.6826166, indexed ISI Web of Science (WOS:000343764100109) & IEEExplore.

31. Curteza A., Cretu V., Macovei L., Poboroniuc M. S., 2014, Designing functional clothes for persons with locomotor disabilities, AUTEX Research Journal, DOI: 10.2478/aut-2014-0028©AUTEX, ISSN 2300-0929 (On-line), http://www.autexrj.com, (impact factor 0.618 for 2013), Volume 14, Issue 4, pp.281-289, indexed Google Scholar.

32. Poboroniuc M. S., Serea F., Hartopanu S., Development of Mechatronic Systems Aiming to Rehabilitate Upper Limb in CVA Patients, in Proceedings of the 9th International Conference on Electromechanical and Power systems SIELMEN2013, Chisinau, Republic of Moldavia, 17-18 October 2013, pp.142-146, ISBN 978-606-13-1560-4.


54

33. Irimia D.C., Poboroniuc M. S., Ortner R., Improved method to Perform FES&BCI Based rehabilitation, in Proceedings of the 4th IEEE International Conference on e-Health and Bioengineering EHB2013, Iasi, Romania, 21-23 November 2013, pp.1-4, ISBN 978-1-4799-2373-4, DOI: 10.1109/EHB.2013.6707384. indexed IEEE Xplore.

34. Poboroniuc M. S., Cojocaru D., Livint G., Ward T., Cachia E., Bencheva N., Preliminary Findings to Design EIE Curricula Harmonized to the Technical Global Challenges, in Proceedings of the 24th EAEEIE Annual Conference, SALEIE workshop, Chania, Greece, 30 – 31 May, 2013, pp. 198-203, CD:9789609988957, DOI:10.1109/EAEEIE.2013.6576529, IEEE Conference publications, indexed Web of Science Thomson Reuters & IEEE Xplore.

35. Serea F., Poboroniuc M. S., Irimia D.C., Hartopanu S., Olaru R., Preliminary Results on a Hybrid FES-Exoskeleton System Aiming To Rehabilitate Upper Limb in Disabled People, in Proceedings of the 17th International Conference on Systems Theory, Control and computing ICSTCC2013, Sinaia, Romania, 11-13 October 2013, pp.722-727, ISBN 978-1-4799-2228-4, ISBN 978-1-4799-2227-7, IEEE catalog Number CFP1336P-CDR, DOI: 10.1109/ICSTCC.2013.6689046 . indexed Web of Science Thomson Reuters &IEEE Xplore.

36. Hartopanu S., Poboroniuc M. S., Serea F., Irimia D.C., Livint G., Design of a Hybrid FES-Mechanical Intelligent Haptic Robotic Glove, in Proceedings of the 17th International Conference on Systems Theory, Control and computing ICSTCC2013, Sinaia, Romania, 11-13 October 2013, pp.687-692, ISBN 978-1-4799-2228-4, ISBN 978-1-4799-2227-7, IEEE catalog Number CFP1336P-CDR, DOI: 10.1109/ICSTCC.2013.6689040. indexed Web of Science Thomson Reuters &IEEE Xplore.

37. Poboroniuc M. S., Irimia D.C, Popescu N., Popescu Dorin, Engineered devices to support stroke rehabilitation, in Proceedings of the 19th International Conference on Control Systems and Computer Science CSCS19, 29-31 May 2013, Bucharest, Romania, pp. 289-295, DOI 10.1109/CSCS.2013.44, Published by the IEEE Computer Society, ISBN: 978-0-7695-4980-4, indexed Web of Science Thomson Reuters & IEEE Xplore.

38. Popescu N., Popescu Decebal, Ivanescu M., Popescu Dorin, Vladu C., Berceanu C., Poboroniuc M. S., Exoskeleton design of an intelligent haptic robotic glove, in Proceedings of the 19th International Conference on Control Systems and Computer Science CSCS19, 29-31 May 2013, Bucharest, Romania, pp. 196-202, DOI 10.1109/CSCS.2013.21, Published by the IEEE Computer Society, ISBN: 978-0-7695-4980-4, indexed Web of Science Thomson Reuters & IEEE Xplore.

39. Irimia D.C., Ortner R., Krausz G., Guger C., Poboroniuc M., 2012, BCI Application in Robotics Control, 14th IFAC Symposium on Information Control Problems in Manufacturing, Bucharest, Romania, 23-25 May, 2012, Proceedings Volumes by Elsevier Ltd on IFAC-PapersOnLine.net, Information Control Problems in Manufacturing, Vol.14, Part.1, pp. 1-6, ISSN: 1474-6670; ISBN: 978-3-902661-98-2, Digital Object Identifier: 10.3182/20120523-3-RO-2023.00432, , article – indexed SCOPUS.

40. Irimia D.C, Poboroniuc M. S., Motor Imagery Based BCI Approach to Control Neuroprostheses, in Proceedings of the International Workshop FIHS2012- “Fostering Innovation in Healthcare Services”, March 14-15, 2012, Brasov, Romania, FIHS Book chapter, pp. 113-118, Editors: Theodor Borangiu, Radu Dobrescu, Editura Universitara Carol Davila, ISBN: 978-973-708-659-4.

41. Popescu D., Selisteanu D., Poboroniuc M., Irimia D-C, 2011, Robotics application within bioengineering: Neuroprosthesis test bench and model based neural control for a robotic leg, Proceedings of the 3rd International Conference on Intelligent Decision Technologies (IDT´ 2011), INTELLIGENT DECISION TECHNOLOGIES (book chapter), J. Watada et al. (Eds.), Vol.10, Part 1, pp.283-294, ISSN 2190-3018, ISBN 978-3-642-22193-4, e-ISBN 978-3-642-22194-1, DOI 10.1007/978-3-642-22194-1, 2011, Edited by Springer-Verlag Berlin Heidelberg, article- indexed SPRINGER.


55

42. Curteza A., Macovei L., Cretu V., Poboroniuc M. S., Kalaoglu F., Karakas H., Gorgun B., Designing functional products for persons with neuromotor diseases, The XV-th International Conference of Inventics “INVENTICA 2011”, June 8-10, 2011, Iasi, Romania, INVENTICA2011 BOOK, Edited by Performantica Publishing House, Iasi, pp.548-555, ISSN: 1844-7880, Awarded with gold Medal, Diplom awarded by International Salon of Inventions and New Technologies “NEW TIME” from Sevastopol, Ukraine.

43. Ignat E.B., Bohotin V., Dobrin R., Chirita R., Poboroniuc M., Popescu C.D., 2011, The effect of functional electrical stimulation on cortical excitability and clinical parameters of gait in chronic stroke survivors, Buletin de Psihiatrie Integrativa, nr.2/2011 (49), 2011, ISBN: 973-9375-08-1 (article).

44. V. Bobot, S.D. Ionascu, Poboroniuc M. (joint author), J.K. Borup, M. Vettensaari, S. Wintgen, Editors: Silvia-Daniela Ionascu, Marian- Silviu Poboroniuc, Radu Ionascu, Cum se construieste un robot (Un ghid practic pas cu pas pentru elevii si profesorii din scoli gimnaziale si licee); Proiect ROKEY 2008-2010, 100 pp., 2011, Edited by IMPRIMIS, Str.Chimiei, nr.16, Iasi, Romania, ISBN 978-606-92400-1-4 (Faculty of Electrical Engineering of Iasi acts as associate partner).

45. Poboroniuc M. (joint author), Editors: V. Bobot, J.K. Borup, S.D. Ionascu, M. Vettensaari, S. Wintgen, How to build a robot (A step by step practical guide for students and teachers of secondary and high schools) (ROKEY project result – see www.rokey.eu web page); 99 pp., 2010, Edited by WEBPRINT s.r.o., Trencin, Slovac Republic, ISBN 978-80-970475-2-8 (Faculty of Electrical Engineering of Iasi acts as associate partner).

46. Poboroniuc M., Wood D.E., Riener R., Donaldson N.N., 2010, A New Controller for FES-Assisted Sitting Down in Paraplegia, Advances in Electrical and Computer Engineering (http://www.aece.ro/ ), Vol.10, No.4, November 2010, pp. 9-16, ISSN: 1582-7445, e-ISSN: 1844-7600, Digital Object Identifier: 10.4316/AECE.2010.04002, Edited by Stefan cel Mare University of Suceava, Romania, article - indexed ISI (ISI impact factor for 2010: 0.688) & SCOPUS.

47. Popescu D., Popescu N., Poboroniuc M., 2010, The FPGA implementation of a neurostimulator, Studies in Informatics and Control Journal (Useful Applications and Advanced Technology), Vol.19, Issue 1, March 2010, pp.85-92, ISSN 1220-1766, Edited by National Institute for R&D in Informatics ICI Bucharest, article- indexed ISI (since 2008, ISI impact factor for 2010: 0.671) & INSPEC database (since 1993) & SCOPUS.

48. Irimia D.C, Poboroniuc M. S., Stefan C.M., Livint Gh., Voice Controlled Neuroprosthesis System, in Proceedings of the 12th Mediterranean Conference on Medical and Biological Engineering and Computing MEDICON 2010, May 27-30, 2010, Chalkidiki, Greece, IFMBE Proceedings vol.29, p. 426-429, ISSN: 1680-0737, ISBN: 978-3-642-13038-0, DOI: 10.1007/978-3-642-13039-7_107, indexed in SCOPUS.

49. Popescu D., Ionete, C., Popescu, L, Poboroniuc M. S., Robotic leg control based on human motion analysis and neural control methods, in 2010 IEEE International Conference on Automation and Logistics (ICAL), August 216-20, 2010, Hong Kong and Macau, pp. 261 - 266, E-ISBN: 978-1-4244-8374-7, ISBN: 978-1-4244-8375-4, DOI: 10.1109/ICAL.2010.5585292, Indexed in IEEExplore & SCOPUS.

50. Poboroniuc M. S., Stefan C.M., Livint Gh., Irimia D.C, Chirila A.M., Voice recognition control in mobile robots and neuroprosthesis, in Proceedings of the 7th International Conference on Electromechanical and Power Systems SIELMEN 2009, October 8-9, 2009, Iasi, Romania, Volume 2, pp.II-141 – II-144, general ISBN: 978-606-520-618-2, ISBN vol.II: 978-606-520-623-6.

51. Poboroniuc M., Petrescu M., Stefan M-C, Livint G., 2009, Different robotic structures aiming to help in testing neuroprosthesis control strategies, RAAD2009 Program and Book of Abstracts, PRINTECH Publishing House, ISBN 978-606-521-315-9, pp. 99, Proc. of the 18th Int. Workshop on Robotics in Alpe-Adria-Danube Region, CD - ISSN 2066-4745, pp.1-6 and VENTIL


56

Journal (Revija za FLUIDNO TEHNIKO, AVTOMATIZACIJO IN MEHATRONIKO), edited by University of Ljubljana, Ventil 15/ 2009/6, pp.510-515, ISSN 1318-7279 (indexed in INSPEC database).

52. Poboroniuc M., Kamnik R., Bajd T., 2009, Research issues in neuroprostheses test and control of standing and walking in paraplegia (plenary lecture), RAAD2009 Program and Book of Abstracts, PRINTECH Publishing House, ISBN 978-606-521-315-9, pp. XXXI, Proc. of the 18th Int. Workshop on Robotics in Alpe-Adria-Danube Region, CD - ISSN 2066-4745, pp.1-7.

53. Popescu D., Poboroniuc M., Craciunescu P., Popescu L.C., 2009, Analysis and interpretation of the human body motion images for a robotic implementation, ICCGI2009, Proc. of the 4th International Multi-Conference on Computing in the Global Information Technology, Cannes/La Bocca, France, 23-29 august 2009, pp.253-258, indexed IEEE-Xplore & INSPEC database (DOI 10.1109.ICCGI.2009.45, ISBN 978-0-7695-3751-1/09).

54. Poboroniuc M. S., Stefan C.M., Livint Gh., Irimia D.C, Issues on mechatronic devices aiming to test neuroprosthesis, in Proceedings of the 2009 Advanced Technologies for Enhanced Quality of Life - AT-EQUAL 2009, July 22-26, 2009, Iasi, Romania, IEEE Computer Society Order Number P3753, pp.23-27, ISBN-13: 978-0-7695-3753-5, DOI 10.1109/AT-EQUAL.2009.16 - indexed IEEE-Xplore & INSPEC database.

55. Poboroniuc M., Stefan C., 2008, Method and mechatronic device to test a sitting-down FES-based control strategy in paraplegia, Bulletin of the Polytechnic Institute of Iasi, Tomul LIV (LVIII), Fasc. 5, 2008, Electrotehnică, Energetică, Electronică, Conference EPE2008, Iasi, pp.981-987, 3-5 oct.2008.

56. Chirila M., Irimia D., Poboroniuc M. (coordinator), 2009, Car binacle devices voice control emulated on a mobile robot - ROBVOICE, StudING Tg. Jiu, Volum de lucrari editat de Editura ACADEMICA BRANCUSI, pp.16-23, ISSN 1844-3249, Aprilie 27- Mai 3, 2009, Targu Jiu, Romania, 1st prize - StudING student competition.

57. Luiuz M., Roman C., Poboroniuc M. (coordinator), 2009, Robot mobil inteligent dotat cu echipament de recunoastere audio si video pentru urmarirea obiectelor si executarea unor comenzi, StudING Tg. Jiu, Volum de lucrari editat de Editura ACADEMICA BRANCUSI, pp.188-194, ISSN 1844-3249, Aprilie 27- Mai 3, 2009, Targu Jiu, Romania, 3rd prize - StudING student competition.

58. Poboroniuc M., 2009, An overview on neuroprosthesis control and test, Bulletin of the Polytechnic Institute of Iasi, Tomul LV (LIX), Fasc. 1, 2009, Sectia: Electrotehnică, Energetică, Electronică (Section: Electrical Engineering, Power Engineering and Electronics), pp.69-81.

59. Chirila M., Stan P., Poboroniuc M. (coordinator), 2008, Ghidarea wireless a unui robot hibrid si comanda inteligenta autonoma in preajma unor tinte-ROBSAFE, StudING Tg. Jiu, Volum de lucrari editat de Editura ACADEMICA BRANCUSI, pp.9-14, ISSN 1844-3249, Aprilie 18-20, 2008, Targu Jiu, Romania, 2nd prize - StudING competition.

60. Poboroniuc M., Stefan C., 2008, A method to test FES-based control strategies for neuroprostheses, WSEAS-ICAI2008, The 9th WSEAS International Conference on Automation and Information, WSEAS Press, pp.344-349, Section Automation&Information: Theory and Advanced Technology, June 24-26, 2008, Bucharest, Romania, ISSN 1790-5117, ISBN: 978-960-6766-77-0 ( ISI Web of Knowledge & Web of Science).

61. Poboroniuc M., Stefan C., EMULOBODY-a mechatronic device to test FES-based standing control strategies in paraplegia, in Proceedings of the 13th Annual Conference of the International Functional Electrical Stimulation Society, September 21-25, 2008, Freiburg, Germany, Published in Biomedizinische Technik/Biomedical Engineering, Vol. 53 (2008), Supplement 1, pp. 168-170, ISSN 0939-4990.

62. Poboroniuc M., Paicu G., Stefan C., Irimia D., 2008, A mechatronic device to teach neuroprosthesis control, COMEFIM'9: The 9th International Conference on Mechatronics and Precision Engineering, Bulletin of the Polytechnic Institute of Iasi, tom LIV (LVIII), Fasc.4, pp.1-6, Section Machine Manufacturing, June 12-14, 2008, Iasi, Romania, ISSN 1011-2855.


57

63. Poboroniuc M., Paicu G., Stefan C., Irimia D., A laboratory bench to test a neuroprosthesis control prior to using it on paraplegics, Proceedings of the 6th European Symposium on Biomedical Engineering, ESBME 2008, 19-21 iunie 2008, Chania, Greece, CD proceedings, Univ. of Patras.

64. Irimia Danut, Stefan C., Poboroniuc M., Luiuz M., Roman C., 2007, Mobile robot control in obstacles avoidance tasks by means of a Motorola-based hardware platform, 6th International Conference on Electromechanical and Power Systems, 4-6 October, Chisinau, Republic of Moldavia, in Annals of the University of Craiova, Electrical Engineering Series: No.31, vol.I, 2007, pp. 186-189, Universitaria Publishing House, 2007, ISSN: 1842-4805.

65. Poboroniuc M., Kamnik R., Stefan C., Livint Gh., Lucache D., Bajd T., New Experimental Results in Assessing and Rehabilitating the Upper Limb Function by Means of the Grip Force Tracking Method, Proceedings of the 11th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON2007, 26-30 iunie 2007, Ljubljana, Slovenia, IFMBE proceedings 2007, vol.16, nr.2, ISSN 1727-1983, ISBN 978-3-540-73043-9 Springer Berlin Heidelberg New York, ISSN 1680-0737 Springer Science& Business Media, Germany, pp.954-957 (in British library direct, http://demo.viidea.com/medicon07_poboroniuc_ner/, ISI Web of Knowledge & Web of Science).

66. Poboroniuc M., Petrescu M., Stefan C., Livint G., 2007, Learning neuroprostheses control by using remotely tutored simulation and experiments, SINTES13: The International Symposium on Systems Theory, Annals of the University of Craiova, Series: Automation, Computers, Electronics and Mechatronics, vol.4 (31), No.2 Universitaria Publishing House, 2007, pp. 115-120, ISSN 1841-0626.

67. Poboroniuc M., 2006, New experimental results on feedback control of FES-based standing in paraplegia, 2nd International Conference on Biomaterials and Medical Devices BiomMedD’2006, Conference proceedings - abstract volume, pp.114, November 09-11, 2006, Iasi, Romania, ISBN 973-718-566-8, AL.I.CUZA University Scientific Annals of Biophysics, Medical Physics and Environment Physics, tom.I, vol.III, pg.83-89, 2007.

68. Poboroniuc M., Stefan C. , Petrescu M., Livint Gh., 2006, FES-based control of standing in paraplegia by means of an improved knee extension controller, 4th International Conference on Electrical and Power Engineering EPE2006, Bulletin of the Polytechnic Institute of Iasi, tom LII (LIV), Fasc.5A, pp.517-522, October 12-13, 2006, Iasi, Romania, ISSN 1223-8139.

69. Petrescu M., L. Livint, Poboroniuc M., Romila E., 2005, Fuzzy logic control for yaw rate vehicle, Proceedings of the 5th International Conference on Electromechanical and Power Systems - SIELMEN 2005, vol.II, October 6-8, pp.729-732, 2005, Chisinau, Republic of Moldavia, ISBN 973-716-208-0.

70. Poboroniuc M., Stefan Ciprian, L. Livint, 2005, A comparison between FES-based control methods that aim to support standing-up and sitting-down for spinal cord injured persons, Proceedings of the 5th International Conference on Electromechanical and Power Systems - SIELMEN 2005, vol.II, October 6-8, pp.664-666, 2005, Chisinau, Republic of Moldavia, ISBN 973-716-208-0.

71. Poboroniuc M., Popescu C.D., Livint G., Lucache D., Petrescu M., 2004, New developments in assisting the physically impaired persons for training and assessment by means of FES, 3rd International Conference on Electrical and Power Engineering EPE’2004, Bulletin of the Polytechnic Institute of Iasi, tom L (LIV), Fasc.5A, pp.116-119, October 7-8, 2004, Iasi, Romania, ISSN 1223-8139.

72. Poboroniuc M., Wood, D., Donaldson, N., Riener, R., 2004, Stanmore Stimulator Application Programme to Sustain a Standing-Up, Standing and Sitting-Down Chained Motion in Paraplegia, 9th Annual Conference of the International FES Society September 6-9, pp.225-227, 2004, Bournemouth, UK, ISBN 1-85899-191-9.

73. Poboroniuc M., Popescu, C. D., Livint, G., Lucache, D., Petrescu, M., 2004, FES-based assistive rehabilitative devices for training of physically impaired persons, Proceedings of the 8th


58

International Symposium on Automatic Control and Computer Science - SACCS 2004, October 22-23, 4 pg., 2004, Iasi, Romania, ISBN 973-621-086-3.

74. Livint G., Petrescu M., Poboroniuc M., 2004, A fuzzy model of vehicles tire, 3rd International Conference on Electrical and Power Engineering EPE’2004, Bulletin of the Polytechnic Institute of Iasi, tom L (LIV), Fasc.5A, pp.69-74, October 7-8, 2004, Iasi, Romania, ISSN 1223-8139.

75. Petrescu M., Livint G., Poboroniuc M., 2004, Vehicles stability study, 3rd International Conference on Electrical and Power Engineering EPE’2004, Bulletin of the Polytechnic Institute of Iasi, tom L (LIV), Fasc.5A, pp.110-115, October 7-8, 2004, Iasi, Romania, ISSN 1223-8139.

76. Poboroniuc M., G. Livint, D. Lucache, 2003, From robots to human body: FES-based Control of Standing in Paraplegia, 12th International Symposium on Power Electronics – Ee 2003, Novi Sad, Serbia & Montenegro, November 5th -7th, Paper no. T4-1.8, pp.1-3, 2003.

77. Kamnik R., Poboroniuc M., Bajd T., Livint G., Lucache D., 2003, Robot and FES technology for augmenting standing-up and sitting down capabilities, IEEE International Conference on Industrial Technology, Maribor, Slovenia, December 10-12, 2003, pp.718-723, ISBN: 0-7803-7853-9, IEEE Catalog Number: 03TH8685C (DOI 10.1109/ICIT.2003.1290744 - indexed IEEE-Xplore & INSPEC database & SCOPUS).

78. M. Poboroniuc, Fuhr, T., Wood, D., Riener, R., Donaldson, N., A Fuzzy Controller to Model Shoulder Forces within FES Model-Based Simulation of a Paraplegic Patient, Proceedings of the 3rd Academic Biomedical Engineering Research Group (ABERG) Workshop 2002, Bournemouth, UK, September 17th, 2002, pp.11-16, ISBN:1-85899-178-1.

79. M. Poboroniuc, Fuhr, T., Wood, D., Riener, R., Donaldson, N., FES-Induced Standing-Up and Sitting-Down Control Strategies in Paraplegia. FESnet Conference 2002, September 2nd-3rd, Glasgow, UK, pp.1-3.

80. Fuhr T., Quintern J., Riener R., Schmidt G., Walk! - Experiments with a Cooperative Neuroprosthetic System for the Restoration of Gait, Proc. 6th Conf. of the IFESS, 1-3, 2001.

81. M. Poboroniuc, D. Popescu, L. Nita, Modeling and control of a mobile manipulator, 10th National Conference on Electric Drives CNAE’2000, Bulletin of the Polytechnic Institute of Iasi, tom XLVI (L), Fasc.5, 2000, 218-223, October 12-14, 2000, Iasi, Romania.

82. Petrescu M., Poboroniuc M., L. Livint, E-learning issue on remote control for mobile robots by means of an Easy Java program, Proceedings of the 17th EAEEIE Annual International Innovation in Education for Electrical and Information Engineering, June 1st-3rd, 2006, Craiova, Romania, pg.54-59, ISBN 973-742-350-X (ISBN 978-973-742-350-4).

83. M. Poboroniuc, G. Livint, Smooth feedback control for path tracking on a three wheels mobile robot, Scientific Bulletin of “Politehnica” University of Timisoara, vol.45(59), No.1, 87-92, Fourth International Conference on Technical Informatics CONTI’2000, October 12-13, 2000, Timisoara, Romania.

84. M. Poboroniuc, D. Popescu, Gh. Livint, A. El Hajjaji, An improved fuzzy control technique for the path tracking problem on mobile robots, Proceedings of the 7th International Conference on Optimization of Electrical and Electronic Equipment OPTIM'2000, 2000, 655-658, May 11-12, Brasov, Romania.

85. M. Poboroniuc, D.Popescu, A. El Hajjaji, Gh. Livint, X. Dovifaaz, Improved fuzzy and neural control on mobile robots path tracking, 3rd International Conference on Applied Mathematics and Engineering Sciences CIMASI’2000, October 23-25, 2000, Casablanca, Morocco.

86. M. Poboroniuc, D.Popescu, Gh. Livint, A. El Hajjaji, An improved fuzzy control technique for the path tracking problem on mobile robots, Proceedings of the 7th International Conference on Optimization of Electrical and Electronic Equipment OPTIM'2000, 2000, 655-658, May 11-12, Brasov, Romania.

87. Poboroniuc M., D. Popescu, A. El Hajjaji, A. Rachid, X. Dovifaaz, Fuzzy and neural control technics for mobile robots, Workshop "Mobile robots", Technical Science Academy, Craiova University,1999, October 29-30, Romania.


59

88. M. Poboroniuc, L. Caramelle, Gh. Livint, M. Zasadzinski, H-infinity controller reduction technique applied to an asynchronous motor on vehicle and mobile robots, Bulletin of the Polytechnic Institute of Iasi, tom XLV (IL), Fasc.5A, 148-151, 1999, ISSN 0258-9109, Iasi, Romania.

89. M. Poboroniuc, A. El Hajjaji, A. Rachid, Path tracking fuzzy controller for mobile robots, Bulletin of the Polytechnic Institute of Iasi, tom XLV (IL), Fasc.5A, 144-147, 1999, ISSN 0258-9109, Iasi, Romania.

90. Wood D. E., Donaldson N. N., Perkins T. A., Apparatus to measure simultaneously 14 isometric leg joint moments. Part 2: Multi-moment chair system. Med. Biol. Eng. Comput. 37, 2, 148-154 (1999).

91. Riener, R., Fuhr, T. 1998. Patient-Driven Control of FES-Supported Standing Up: a Simulation Study. IEEE Trans. Rehab. Eng.; 6: pp.113-124.

92. Wood D.E., Harper V.J., Barr F.M.D., et al., Experience in using knee angles as part of a closed-loop algorithm to control FES-assisted paraplegic standing, 6th Vienna Int. Wokshop on FES: Basics, Technology and Application, 137-140, 1998.

93. M. Poboroniuc, Gh. Livint, Parallel robot trajectory modeling with neural networks, Proceedings of fourth Groningen International Information Technology Conference for Students, 1997, 53-56, Groningen , Olanda.

94. Donaldson N. d. N., and Yu C. H., FES standing control by handle reactions of leg muscle stimulation (CHRELMS). IEEE Trans. Rehabil. Eng., 4, 280-284, (1996).

95. Mulder A. J., Veltink P. H., Boom H. B. K., On/off control in FES-induced standing up: A model study and experiments, Med. BioEng., Vol. 30, 205-212, 1992.

96. *** “Un profesor din Iasi a creat prima mana robotica pentru persoanele paralizate.” – ProTV journal broadcast – on-line reference published on April 10th, 2016: http://stirileprotv.ro/article/ilikeit/smart-things/un-profesor-iesean-a-creat-prima-mana-robotica-pentru-persoanele-paralizate-ce-s-a-intamplat-cu-cei-care-au-testat-o.html [accessed on April 18th, 2016].

97. *** “Invenție revoluționară la Iași. Mănușa mecatronică pentru bolnavii de paralizie” – Digi24 TV journal broadcast - on-line reference published on March 16th, 2016: http://www.digi24.ro/Stiri/Digi24/Actualitate/Stiri/Inventie+Iasi+Manusa+mecatronica [accessed on April 18th, 2016].

98. *** SALEIE WP3 deliverables – on-line reference –accessed on April 19th, 2016. http://www.saleie.co.uk/WP3TechChallenges/docs/D3_4_SALEIE_ICT_Bachelor.pdf

99. *** SALEIE WP3 deliverables – on-line reference –accessed on April 19th, 2016. file:///E:/Date_diskE/Pobo/_2014_Abilitare/Teze%20abilitare/D3_4_SALEIE_ICT_Master.pdf

http://stirileprotv.ro/article/ilikeit/smart-things/un-profesor-iesean-a-creat-prima-mana-robotica-pentru-persoanele-paralizate-ce-s-a-intamplat-cu-cei-care-au-testat-o.html

http://stirileprotv.ro/article/ilikeit/smart-things/un-profesor-iesean-a-creat-prima-mana-robotica-pentru-persoanele-paralizate-ce-s-a-intamplat-cu-cei-care-au-testat-o.html

http://www.digi24.ro/Stiri/Digi24/Actualitate/Stiri/Inventie+Iasi+Manusa+mecatronica

http://www.saleie.co.uk/WP3TechChallenges/docs/D3_4_SALEIE_ICT_Bachelor.pdf

DOMAIN: ELECTRICAL ENGINEERING Prof. univ. dr. ing. Marian ...

Documents

Transcript of DOMAIN: ELECTRICAL ENGINEERING Prof. univ. dr. ing. Marian ...