Multi-robot Cooperation Based on Continuous Reinforcement Learning with Two State Space Representations.
Toshiyuki YasudaKazuhiro OhkuraKazuaki YamadaPublished in: SMC (2013)
Keyphrases
- multi robot
- state space
- reinforcement learning
- action space
- continuous state spaces
- path planning
- reinforcement learning algorithms
- multi robot systems
- mobile robot
- multi agent
- markov decision processes
- cooperative
- multi robot exploration
- heuristic search
- optimal policy
- search and rescue
- uncertain environments
- function approximators
- continuous state
- motion planning
- potential field
- model free
- robot soccer
- formation control
- real robot
- markov decision process
- state variables
- dynamic programming
- multi robot cooperative
- multiple robots
- function approximation
- planning problems
- initially unknown
- machine learning
- robotic systems
- particle filter
- multi agent systems
- coalitional game theory
- policy iteration
- markov decision problems
- partially observable
- belief state
- reward function