An Actor-Critic Approach for Legible Robot Motion Planner.
Xuan ZhaoTingxiang FanDawei WangZhe HuTao HanJia PanPublished in: ICRA (2020)
Keyphrases
- robot motion
- actor critic
- optimal control
- reinforcement learning
- control strategy
- mobile robot
- temporal difference
- policy gradient
- approximate dynamic programming
- neuro fuzzy
- gradient method
- outdoor environments
- collision avoidance
- humanoid robot
- reinforcement learning algorithms
- function approximation
- heuristic search
- obstacle detection
- free space
- domain independent
- policy iteration
- real robot
- visual servoing
- mathematical model
- autonomous vehicles
- stereo vision
- path finding
- convergence rate
- state space
- dynamic environments
- dynamical systems
- control method
- average reward
- initial state
- motion planning
- evaluation function
- optimal policy
- vision system
- video sequences