Probabilistic Roadmaps Methods are Embarrassingly Parallel

Lucia K. Dale
Mathematics & Computer Science
University of the South
Sewanee, TN 37383-1000
ldale D0T sewanee D0T edu
Nancy M. Amato
Computer Science
Texas A&M University
College Station, TX 77843-3112
amato@cs.tamu.edu

Abstract

In this paper we report on our experience parallelizing probabilistic roadmap motion planning methods (PRMs). We show that significant, scalable speedups can be obtained with relatively little effort on the part of the developer. Our experience is not limited to PRMs, however. In particular, we outline general techniques for parallelizing types of computations commonly performed in motion planning algorithms, and identify potential difficulties that might be faced in other efforts to parallelize sequential motion planning methods.

In Proceedings of the 1999 IEEE International Conference on Robotics and Automation (ICRA'99), 1991. Full Paper (postscript)