Relay Hindsight Experience Replay: Continual Reinforcement Learning for Robot Manipulation Tasks with Sparse Rewards.
Yongle LuoYuxin WangKun DongQiang ZhangErkang ChengZhiyong SunBo SongPublished in: CoRR (2022)
Keyphrases
- manipulation tasks
- reinforcement learning
- real robot
- robot navigation
- human robot interaction
- motion planning
- humanoid robot
- function approximation
- markov decision processes
- service robots
- object manipulation
- robotic systems
- action selection
- state space
- model free
- end effector
- impedance control
- human activities
- learning algorithm
- robotic arm
- optimal policy
- dynamic programming
- reward function
- robot control
- machine learning
- mobile robot
- control policy
- degrees of freedom
- gesture recognition
- robot manipulators
- object detection
- path planning
- control strategies
- human motion
- autonomous robots