Decentralized control of multi-robot partially observable Markov decision processes using belief space macro-actions.
Shayegan OmidshafieiAli-Akbar Agha-MohammadiChristopher AmatoShih-Yuan LiuJonathan P. HowJohn VianPublished in: Int. J. Robotics Res. (2017)
Keyphrases
- multi robot
- decentralized control
- belief space
- partially observable markov decision processes
- markov decision processes
- finite state
- motion planning
- planning under uncertainty
- optimal policy
- reinforcement learning
- state space
- mobile robot
- path planning
- partially observable
- continuous state
- dynamic programming
- multiagent systems
- decision theoretic planning
- reinforcement learning algorithms
- belief state
- average cost
- multi agent
- dynamic environments
- initial state
- infinite horizon
- robotic systems
- decision problems
- dynamical systems
- multi agent systems
- learning algorithm
- machine learning
- domain specific