Solving Continuous-State POMDPs via Density Projection.
Enlu ZhouMichael C. FuSteven I. MarcusPublished in: IEEE Trans. Autom. Control. (2010)
Keyphrases
- continuous state
- partially observable markov decision processes
- reinforcement learning
- robot navigation
- finite state
- policy search
- control policies
- state dependent
- dec pomdps
- action space
- point based value iteration
- planning problems
- continuous action
- decision problems
- optimal policy
- markov decision processes
- dynamic programming
- heuristic search
- policy iteration
- state space
- hidden markov models
- continuous state spaces
- multi agent
- bayesian networks
- machine learning