Reinforcement learning with heuristic to solve POMDP problem in mobile robot path planning.
Widyawardana AdiprawitaAdang Suwandi AhmadJaka SembiringBambang R. TrilaksonoPublished in: ICEEI (2011)
Keyphrases
- reinforcement learning
- robot path planning
- dynamic programming
- state space
- markov decision processes
- continuous state
- partially observable
- path planning
- optimal policy
- model free
- international space station
- function approximation
- markov decision process
- hidden state
- policy evaluation
- optimal control
- learning algorithm
- partially observable markov decision processes
- markov decision problems
- least squares
- action selection
- belief state
- search algorithm
- route planning
- optimal solution
- multi agent