We
focus on parallelizing RRT on distributed-memory architectures, which requires the
use of the message passing interface (MPI). The idea of improving
motion-planning performance using parallel computation is not new. A survey of Star Student Project some early work
proposes a classification scheme to review various motion-planning approaches and
related parallel processing methods. A more recent trend is to exploit the
multi core technology available on many of today’s PCs, which allows having
multiple threads collaboratively solving a problem.