Planning Motion in Environments with Similar Obstacles
Jyh-Ming Lien and Yanyan Lu
Overview
In this work, we investigate solutions to the following question: Given two motion planning problems W1 and W2 with the same robot and similar obstacles, can we reuse the computation from W1 to solve W2 more efficiently? While the answer to this question can find many practical applications, all current motion planners ignore the correspondences between similar environments. Our study shows that by carefully storing and reusing the computation we can gain significant efficiency.Full Text
PDFDOWNLOAD from RSS website
Publications
Planning Motion in Similar Environments, Jyh-Ming Lien and Yanyan Lu, Proceedings of Robotics: Science and Systems V, June 2009
BibTeX
BibTeX
Example

- A tiny ob-map database
- transformed ob-maps
- composed global roadmaps
Related Work
- Point-based Minkowski Sum
- Gaussian Sampling for Probabilistic Roadmap Planners by Boor et al.
- Obstacle-Based PRM by Amato et al. at Parasol Lab
List of MASC Research Pages