Fast reinforcement learning using stochastic shortest paths for a mobile robot.
Wooyoung KwonIl Hong SuhSanghoon LeeYoung-Jo ChoPublished in: IROS (2007)
Keyphrases
- shortest path
- mobile robot
- reinforcement learning
- direct policy search
- robot control
- real robot
- shortest path problem
- shortest path algorithm
- obstacle avoidance
- road network
- path planning
- path length
- markov decision processes
- geodesic distance
- optimal path
- dynamic environments
- minimum cost flow
- finding the shortest path
- travel time
- flow graph
- dynamic programming
- state space
- spatial networks
- simultaneous localization and mapping
- edge weights
- optimal policy
- random walk