Metric State Space Reinforcement Learning for a Vision-Capable Mobile Robot

Details Metric State Space Reinforcement Learning for a Vision-Capable Mobile Robot Metric State Space Reinforcement Learning for a Vision-Capable Mobile Robot

2006-03-07

arxiv.org/abs/cs/0603023v1

Proc. 9th International Conf. on Intelligent Autonomous Systems (IAS 2006) pages 272-281

Computer Science

Robotics

14 pages, 8 figures

Scientific paper

We address the problem of autonomously learning controllers for vision-capable mobile robots. We extend McCallum's (1995) Nearest-Sequence Memory algorithm to allow for general metrics over state-action trajectories. We demonstrate the feasibility of our approach by successfully running our algorithm on a real mobile robot. The algorithm is novel and unique in that it (a) explores the environment and learns directly on a mobile robot without using a hand-made computer model as an intermediate step, (b) does not require manual discretization of the sensor input space, (c) works in piecewise continuous perceptual spaces, and (d) copes with partial observability. Together this allows learning from much less experience compared to previous methods.

Affiliated with

Also associated with

No associations

Rating

Metric State Space Reinforcement Learning for a Vision-Capable Mobile Robot does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Metric State Space Reinforcement Learning for a Vision-Capable Mobile Robot, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Metric State Space Reinforcement Learning for a Vision-Capable Mobile Robot will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-365990