Space Teleoperation Visual Recognition System Using a Novel Pattern Recognition Technique

DOI

10.2991/amee-17.2017.46How to use a DOI?

Keywords

recognition; teleoperation; hand-eye calibration; camera calibration

Abstract

A Space Teleoperation Visual Recognition System is developed. It is used to recognize an object and measure the 3D pose. A highly efficient no iterative approach is used to calculator the external parameters in camera calibration. It divides all reference points into 3-point subsets in order to achieve a series of fourth order polynomials. The optimal solution is determined by solving the roots of the derivative of the cost function. Laser tracker is used to measure accurately the coordinate system of the robot, the end of the robot arm and the 3D target, then the hand-eye relation matrix is calculated based on these measurement results and external camera parameters. Then, we propose a novel pattern recognition technique to show how to reduce the dimensionality by utilizing group and convexity model. The performance of this visual system is tested in a space robot prototype.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Mingming Wang
PY  - 2017/09
DA  - 2017/09
TI  - Space Teleoperation Visual Recognition System Using a Novel Pattern Recognition Technique
BT  - Proceedings of the 2017 2nd International Conference on Automation, Mechanical and Electrical Engineering (AMEE 2017)
PB  - Atlantis Press
SP  - 224
EP  - 227
SN  - 2352-5401
UR  - https://doi.org/10.2991/amee-17.2017.46
DO  - 10.2991/amee-17.2017.46
ID  - Wang2017/09
ER  -