A Reinforcement Learning Technique with an Adaptive Action Generator for a Multi-robot System.
Toshiyuki YasudaKazuhiro OhkuraPublished in: SAB (2008)
Keyphrases
- multi robot
- reinforcement learning
- action selection
- real robot
- path planning
- mobile robot
- robot soccer
- multi robot systems
- action space
- multi robot exploration
- search and rescue
- uncertain environments
- potential field
- state space
- robotic systems
- agent learns
- motion planning
- initially unknown
- dynamic task allocation
- human actions
- markov decision processes
- surveillance system
- reinforcement learning algorithms
- formation control
- robot teams
- multi robot coordination
- multi robot cooperative
- dynamic programming
- machine learning
- map building
- multi agent
- optimal policy
- coalitional game theory
- model free