OBPRM: Obstacle-Based PRM | Algorithms & Applications Group

Algorithms & Applications Group
OBPRM: Obstacle-Based PRM

supported by NSF, Dept. of Education, Texas Higher Education Coordinating Board
Nancy M. Amato
Project Alumni: Osman Burchan Bayazit, Lucia K. Dale, Chris Jones, Guang Song, Daniel Vallejo

Probabilistic RoadMap (PRM) motion planners use randomization to construct a graph (roadmap) in C-space in which nodes correspond to collision-free robot configurations and two nodes are connected by an edge if a 'local planner' can connect them. Queries are then processed by connecting the start and goal to the roadmap. In this project, we are developing an obstacle-based PRM (OBPRM). The main novelty in our approach is that roadmap nodes are sampled on or near C-obstacle surfaces (corresponding to contact configurations). We have shown that this strategy improves the quality of roadmaps for crowded environments.

Papers

Choosing Good Distance Metrics and Local Planners for Probabilistic Roadmap Methods, Nancy M. Amato, O. Burchan Bayazit, Lucia K. Dale, Christopher Jones, Daniel Vallejo, IEEE Transactions on Robotics and Automation, 16(4):442-447, Aug 2000. Also, In Proc. IEEE Int. Conf. Robot. Autom. (ICRA), pp. 630-637, Leuven, Belgium, May 1998. Also, Technical Report, TR98-010, Department of Computer Science, Texas A&M University, May 1998.
Proceedings(ps, abstract)

OBPRM: An Obstacle-Based PRM for 3D Workspaces, Nancy M. Amato, O. Burchan Bayazit, Lucia K. Dale, Christopher Jones, Daniel Vallejo, In Proc. Int. Wkshp. on Alg. Found. of Rob. (WAFR), pp. 155-168, Houston, TX, Mar 1998.
Proceedings(abstract)

A Randomized Roadmap Method for Path and Manipulation Planning, Nancy M. Amato, Yan Wu, In Proc. IEEE Int. Conf. Robot. Autom. (ICRA), pp. 113-120, Minneapolis, MN, Apr 1996.
Proceedings(abstract)

Department of Computer Science | Dwight Look College of Engineering | Texas A&M University
Privacy statement: Computer Science Engineering TAMU