Design of a Virtual Robotic Arm based on the EMG...
Transcript of Design of a Virtual Robotic Arm based on the EMG...
Design of a Virtual Robotic Arm based on the EMG
variation
Ho-Sun Shin1, Asilbek Ganiev 2
, Kang-Hee Lee2
1Department of Cultural Contents, Soongsil University,
369 Sangdo-Ro, Dongjak-Gu, Seoul, 06978, Republic of Korea
2Department of Digital Media, Soongsil University,
369 Sangdo-Ro, Dongjak-Gu, Seoul, 06978, Republic of Korea
Corresponding Author: [email protected]
Abstract. This paper studies design of a virtual robotic arm based on
electromyography (EMG), gyroscope and accelerometer. Data from a Myo
armband which detects various living body signals is delivered through
Bluetooth to a virtual robotic arm which is built in Unity 3D. EMG is electrical
signal generated by tension and relaxation of muscles and a virtual robotic arm
based on EMG is available to control minute movements of forearm muscles
without dynamic motions. Thus, a virtual robotic arm based on EMG is more
suitable to develop applications for hand amputee than a virtual robotic arm
based on gyroscope and accelerometer.
Keywords: Myo armband, Electromyography, Controlling virtual robotic arm,
Unity 3D.
1 Introduction
The purpose of this paper is to design a virtual robotic arm controlled by EMG of
different motions. The reason to basing EMG is to show manipulating virtual robotic
arms in minimum muscle’s movements.
The device which measures EMG, gyroscope and accelerometer, Myo armband
detects electrical activity in forearm muscles [1]. Human forearm has different kinds
of muscles, each of them has different appointment, and these muscles control our
wrist’s movements. And this device connects with other devices or computer via
Bluetooth and it is very comfortable to measure living body signal.
The remaining structure of this paper is outlined as follows: The next section
briefly introduces tools used to make a virtual robotic arm. Section 3 shows
This work was supported by the National Research Foundation of Korea Grant funded by the
Korean Government(NRF-2013S1A5A8020988) And this work was also supported by the
National Research Foundation of Korea Grant funded by the Korean Government(NRF-
2014R1A1A1A05008028)
Advanced Science and Technology Letters Vol.113 (Art, Culture, Game, Graphics, Broadcasting and Digital Contents 2015), pp.38-43
http://dx.doi.org/10.14257/astl.2015.113.09
ISSN: 2287-1233 ASTL Copyright © 2015 SERSC
architecture design and section 4 describes experimental results of virtual robotic
arms. These procedures are also carried out to demonstrate their features.
2 Motivation and Related Research
The proposed system gets data from electromyography, gyroscope and accelerometer
and analyzes them to understand the wrist gesture by measuring which part of
forearms muscles electro activated.
2.1 EMG and Myo sensor
Most gesture-control systems still detect movements with cameras, which can be
thrown off by poor lighting conditions, distance, and simple obstructions. By drawing
gesture information directly from your arm muscles instead of a camera, Myo
circumvents all these problems and also works with devices that don’t have a camera
in the first place [2].
EMG is an electro-diagnostic technique for evaluating and recording the electrical
activity produced by skeletal muscles. Electromyography detects the electrical
potential generated by muscle cells when these cells are electrically or neurologically
activated. The signals can be analyzed to detect medical abnormalities, activation
level, or recruitment order or to analyze the biomechanics of human movement [3].
Fig.1. Myo armband to control devices and games.
Advanced Science and Technology Letters Vol.113 (Art, Culture, Game, Graphics, Broadcasting and Digital Contents 2015)
Copyright © 2015 SERSC 39
3 Architecture design
This section shows Myo armband which has eight different blocks and each of them
contains a medical-grade EMG sensor. The armband also has a three-axis gyroscope
and three-axis accelerometer.
Section 3.1 shows the architectural design for controlling a virtual robotic arm.
Fig. 2. Five hand gestures and their functions.
Fig. 3. Myo armband and the number of sensors.
3.1 Design of virtual robotic arm
Figure 4 shows the whole process of controlling a virtual robotic arm in Unity 3D.
Advanced Science and Technology Letters Vol.113 (Art, Culture, Game, Graphics, Broadcasting and Digital Contents 2015)
40 Copyright © 2015 SERSC
Fig. 4. Architecture of the proposed virtual robotic arm.
4 Experimental results
(a) Wave left (b) Fingers spread
Fig. 5. Controlling robotic arm with gestures using EMG sensor.
Figure 5 is an example of controlling virtual robotic arm by EMG sensors. In this
case, five gestures: wave left for turning left robotic arm, wave right for turning right,
fingers spread for raising up, fist puts down and double tap with fingers are for
Advanced Science and Technology Letters Vol.113 (Art, Culture, Game, Graphics, Broadcasting and Digital Contents 2015)
Copyright © 2015 SERSC 41
automatically finding balloons and blow them. It is because EMG can detect minute
activities generated tension and relaxation of forearm muscles [4].
(a) Rising up (b) Putting down
Fig. 6. Control of robotic arm by gyroscope and accelerometers data in X, Y, Z directions.
Figure 6 represents the control of virtual robotic arm by using gyroscope and
accelerometer. In this case, robotic arm moves in X, Y and Z directions.
5 Conclusion
Nowadays, the fields of electromyography (EMG) and Electroencephalogram (EEG)
are still actively studied. Also, frequency of practical use for rehabilitation and
nursing becomes high. The study mostly is used for purposes of curing and
rehabilitation. Among living body signals like EEG, the field of EMG which has high
accessibility to measure and analysis is more vigorous than other fields.
This paper showed how to control a virtual robotic arm that prototype was built in
Unity 3D by using electromyography, gyroscope and accelerometer sensors. Because
of EMG sensors, we could get clear and important data and we used them to control a
virtual robotic arm. By comparing virtual robotic arms based on EMG, and gyroscope
and accelerometer, we judged which type of virtual robotic arm is suitable for hand
amputee. Experimental results show the control a virtual robotic arm based on
gyroscope and accelerometer. It should go with dynamic movement of whole arm. On
the other hand, a virtual robotic arm based on EMG needs only specific muscles
activities [5]. Therefore, a virtual robotic arm based on EMG is more helpful to use in
case of hand amputee.
References
1. MYO, http://www.myo.com.
2. Myo reads your muscles for the snappiest gesture tracking ever devised,
http://www.digitaltrends.com/pc-accessory-reviews/
Advanced Science and Technology Letters Vol.113 (Art, Culture, Game, Graphics, Broadcasting and Digital Contents 2015)
42 Copyright © 2015 SERSC
3. Stella Y.Botelho.: Comparison of simultaneously recorded electrical and mechanical
activity in myasthenia gravis patients and in partially curarized normal humans. The
American Journal of Medicine. 19, 693--696 (1995).
4. M.B.I. Raez, M.S. Hussain, F.Mohd-Yasin.: Techniques of EMG signal analysis :
detection, processing, classification and applications, Biological Procedeures Online,
BioMed Central. 8, 11—35 (2006)
5. Mondelli, Mauro, Aretini, Alessandro, Greco, Giuseppe.: Knowledge of
electromyography(EMG) in patients undergoing EMG examinations, Functional
Neurology, CIC Edizioni internationali. 29, 195—200 (2014)
Advanced Science and Technology Letters Vol.113 (Art, Culture, Game, Graphics, Broadcasting and Digital Contents 2015)
Copyright © 2015 SERSC 43