A Continuous Off-Policy Reinforcement Learning Scheme for Optimal Motion Planning in Simply-Connected Workspaces.
Panagiotis RousseasCharalampos P. BechlioulisKostas J. KyriakopoulosPublished in: ICRA (2023)
Keyphrases
- learning scheme
- motion planning
- simply connected
- degrees of freedom
- path planning
- trajectory planning
- mobile robot
- robot arm
- learning algorithm
- autonomous mobile robot
- collision free
- robotic arm
- humanoid robot
- multi robot
- optimal path
- robotic tasks
- dynamic programming
- robot navigation
- machine learning
- mechanical systems
- manipulation tasks
- configuration space
- image sequences
- reinforcement learning
- rule learning
- dynamic environments