Multirobot Navigation Using Partially Observable Markov Decision Processes with Belief-Based Rewards.
Alexandros E. TzikasDerek KnowlesGrace Xingxin GaoMykel J. KochenderferPublished in: J. Aerosp. Inf. Syst. (2023)
Keyphrases
- multi robot
- partially observable markov decision processes
- belief space
- belief state
- fully observable
- motion planning
- reinforcement learning
- markov decision processes
- point based value iteration
- path planning
- partial observability
- planning under uncertainty
- mobile robot
- state space
- finite state
- multi robot exploration
- dynamic environments
- optimal policy
- partially observable
- partially observable stochastic games
- belief revision
- dynamical systems
- planning problems
- dynamic programming
- decision problems
- total reward
- multi agent
- leader follower
- robotic systems
- robot navigation
- learning algorithm
- surveillance system
- infinite horizon
- heuristic search
- multi agent systems
- multi robot coordination
- image sequences