A Tube-Based Reinforcement Learning Approach for Optimal Motion Planning in Unknown Workspaces.
Panagiotis RousseasCharalampos P. BechlioulisKostas J. KyriakopoulosPublished in: ICRA (2024)
Keyphrases
- motion planning
- reinforcement learning
- degrees of freedom
- trajectory planning
- path planning
- humanoid robot
- mobile robot
- initially unknown
- dynamic programming
- robot arm
- belief space
- multi robot
- optimal control
- robotic arm
- autonomous mobile robot
- inverse kinematics
- robotic tasks
- mechanical systems
- configuration space
- manipulation tasks
- collision free
- control policy
- obstacle avoidance
- learning algorithm
- optimal policy
- state space
- optimal solution
- control law
- climbing robot