Robot Motion Planning Under Uncertain Condition Using Deep Reinforcement Learning

DOI

10.2991/mecae-18.2018.22How to use a DOI?

Keywords

Motion planning, Reinforcement learning, Deep reinforcement learning, Uncertain condition.

Abstract

The motion planning of industrial robot plays an important role in today's production systems, such as Made in China 2025 and Industry 4.0. The motion planning under uncertain condition is an important research topic in autonomous robots. For promoting the ability of motion planning to adapt to the environment change, in this paper, we propose a deep reinforcement learning (DRL) method which combines reinforcement learning with deep learning for industrial robot motion planning. Our work shows that the DRL-agent is capable of learning how to control the robot to successfully reach robotic tasks without explicit prior Knowledge of kinematics. We conclude that DRL has great potential for industrial robots and production systems, especially in robot motion planning.

Copyright

© 2018, 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  - Zhuang Chen
AU  - Lin Zhou
AU  - Min Guo
PY  - 2018/03
DA  - 2018/03
TI  - Robot Motion Planning Under Uncertain Condition Using Deep Reinforcement Learning
BT  - Proceedings of the 2018 International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2018)
PB  - Atlantis Press
SP  - 122
EP  - 128
SN  - 2352-5401
UR  - https://doi.org/10.2991/mecae-18.2018.22
DO  - 10.2991/mecae-18.2018.22
ID  - Chen2018/03
ER  -