Parametric POMDPs for planning in continuous state spaces.
Alex BrooksAlexei MakarenkoStefan B. WilliamsHugh F. Durrant-WhytePublished in: Robotics Auton. Syst. (2006)
Keyphrases
- continuous state spaces
- partially observable markov decision processes
- continuous state
- state space
- planning problems
- reinforcement learning
- belief state
- finite state
- markov decision processes
- planning under uncertainty
- partially observable
- heuristic search
- dynamical systems
- optimal policy
- dynamic programming
- action space
- robot navigation
- decision problems
- control problems
- markov decision problems
- control policies
- multi agent
- motion planning
- planning domains
- approximate solutions
- computational complexity
- state dependent
- machine learning