Fernando Torres Medina Obstacles Using 2D LiDAR Sensor
Transcript of Fernando Torres Medina Obstacles Using 2D LiDAR Sensor
![Page 1: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/1.jpg)
Detection and Classification of Obstacles Using 2D LiDAR Sensor
Alejandro Olivas GonzálezFernando Torres Medina
Group of Automation, Robotics and Computer Vision (AUROVA), Universidad de Alicante
![Page 2: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/2.jpg)
● Student in the last year of the degree in robotic engineering in the University of Alicante.
● Final Degree Project: Segmentation and classification of the environment for the control and navigation of mobile robots.
● This work was funded by the Spanish Government’s Ministry of Science and Innovation through the research project RTI2018-094279-B-100.
2
![Page 3: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/3.jpg)
● Bot for Localization on Unstructured Environment (BLUE)
3
![Page 4: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/4.jpg)
Introduction● The use of the 2D LiDAR: horizontally or down-ward looking.
● Generate a 3D map from the data of the 2D LiDAR.
● Detect lines and classify them as ground, obstacle or pothole.
● The goal is to reduce the costs of mobile robots.
4
![Page 5: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/5.jpg)
Methodology● Data acquirement
● Line detection
● Map of lines
● Line classification
5
![Page 6: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/6.jpg)
Data acquirement● Data ⇾ Point cloud
● Point cloud ⇾ Local system
● Local system ⇾ Global system
6
![Page 7: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/7.jpg)
Line detection
7
![Page 8: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/8.jpg)
Line detection refinement
8
![Page 9: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/9.jpg)
Map of lines● The map of lines is divided into submaps.
● The lines are defined by two points.
● Search the submap where the first point is.
● Update the submap, searching the nearest line to the new one.
● Check if the line is in another submap.
9
![Page 10: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/10.jpg)
Line classification● The line are classified according to their mean height.
● Three classes:○ Ground○ Obstacle○ Pothole
10
![Page 11: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/11.jpg)
Experiments
11
![Page 12: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/12.jpg)
Experiments
12
![Page 13: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/13.jpg)
Results
13
Predicted class
Ground Obstacle Pothole
Trueclass
Ground 3633 0 90
Obstacle 0 29415 0
Pothole 0 0 107
![Page 14: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/14.jpg)
Conclusion● Correct classification in structural environments.
● Dynamic objects problem.
14
![Page 15: Fernando Torres Medina Obstacles Using 2D LiDAR Sensor](https://reader035.fdocuments.us/reader035/viewer/2022080217/62e8250f074dce7e2111ce70/html5/thumbnails/15.jpg)
Thank you
15