Logo image
Efficient Learning of Dynamics Models using Terrain Classification
Technical documentation   Open access

Efficient Learning of Dynamics Models using Terrain Classification

Bethany Leffler, Christopher Mansley and Michael Littman
Rutgers University
2008
DOI:
https://doi.org/10.7282/T3KH0RRH

Abstract

Abstract. Terrain classification in robotics has heavily focused on determining a region for traversal, while also labeling obstacles. Our work attempts to expand this essentially binary viewpoint and to use terrain classifiers as an indicator for different system dynamics. By learning multiple models of the system dynamics, the robot is able to assess alternative paths based on traversal costs of different terrain types instead of strict distance metrics. We demonstrate a system that reliably learns an optimal control policy using this additional terrain information and contrast it with several systems based on more traditional methods that fail to reliably complete the same task.
pdf
Efficient Learning of Dynamics Models using Terrain4.12 MBDownloadView
Technical Documentation Open Access
url
Report an accessibility issueView
Please complete a content remediation request to report an accessibility issue with a library electronic resource, website, or service.

Metrics

143 File downloads
59 Record Views

Details

Logo image