Multi-robot Box-pushing: Single-Agent Q-Learning vs. Team Q-Learning.
Ying WangClarence W. de SilvaPublished in: IROS (2006)
Keyphrases
- single agent
- multi robot
- multi agent
- potential field
- dynamic environments
- multi robot exploration
- path planning
- robot soccer
- multi robot coordination
- multi agent systems
- search and rescue
- multiple agents
- reinforcement learning
- decision problems
- path finding
- mobile robot
- multi robot systems
- robot teams
- learning agent
- multi agent coordination
- function approximation
- cooperative
- action space
- multiagent learning
- learning algorithm
- stochastic games
- optimal policy
- model free
- state space
- real robot
- multiple robots
- computational complexity
- dynamic programming