Reinforcement Learning with Prior Policy Guidance for Motion Planning of Dual-Arm Free-Floating Space Robot.
Yuxue CaoShengjie WangXiang ZhengWenke MaXinru XieLei LiuPublished in: CoRR (2022)
Keyphrases
- motion planning
- robotic arm
- humanoid robot
- reinforcement learning
- mobile robot
- belief space
- degrees of freedom
- joint space
- path planning
- configuration space
- trajectory planning
- robot arm
- multi robot
- action space
- optimal policy
- autonomous mobile robot
- obstacle avoidance
- robotic tasks
- manipulation tasks
- real robot
- collision free
- end effector
- control law
- inverse kinematics
- state space
- policy search
- mechanical systems
- action selection
- autonomous robots
- markov decision process
- climbing robot
- collision avoidance
- multi modal
- potential field
- human robot interaction
- dynamic environments
- vision system
- viewpoint