Policy Rollout Action Selection with Knowledge Gradient for Sensor Path Planning.
Thore GerlachFolker HoffmannAlexander CharlishPublished in: FUSION (2021)
Keyphrases
- path planning
- action selection
- robot soccer
- path planner
- mobile robot
- collision avoidance
- reinforcement learning
- path planning algorithm
- basal ganglia
- dynamic environments
- temporal difference
- multi robot
- indoor environments
- obstacle avoidance
- decision making
- aerial vehicles
- robot path planning
- path finding
- optimal path
- multiple robots
- potential field
- degrees of freedom
- action space
- autonomous navigation
- motion planning
- optimal policy
- trajectory planning
- navigation tasks
- function approximation
- machine learning
- dynamic and uncertain environments