Partially Observable Monte Carlo Planning with state variable constraints for mobile robot navigation.
Alberto CastelliniEnrico MarchesiniAlessandro FarinelliPublished in: Eng. Appl. Artif. Intell. (2021)
Keyphrases
- monte carlo
- partially observable
- state variables
- state space
- mobile robot navigation
- markov chain
- particle filter
- mobile robot
- reinforcement learning
- markov decision processes
- heuristic search
- reward function
- dynamical systems
- markov decision problems
- planning problems
- belief state
- decision problems
- dynamic systems
- unknown environments
- orders of magnitude
- optimal policy
- infinite horizon
- causal graph
- dynamic programming
- obstacle avoidance
- planning domains
- particle filtering
- reinforcement learning algorithms
- transition probabilities
- action space
- robotic systems
- image interpretation
- long run
- motion planning
- function approximation
- visual tracking
- path planning
- linear programming
- video sequences