Multi-Robot Motions in Milliseconds: Vector-Accelerated Primitives for Sampling-Based Planning

TL;DR

This paper extends the VAMP framework to multi-robot motion planning, introducing vector-accelerated primitives that significantly speed up validation and planning times through SIMD parallelism.

Contribution

It develops two new vector-accelerated primitives for multi-robot motion planning that leverage SIMD parallelism, achieving over 1100X speedup in validation and over 850X in planning.

Findings

Over 1100X speedup in motion validation time.

Multi-robot solutions found in milliseconds.

Planning time speedups of over 850X.

Abstract

In this paper, we extend the recent Vector-Accelerated Motion Planning (VAMP) framework to multi-robot motion planning (MRMP). We develop two vector-accelerated primitives, multi-robot MotionValidation (MotVal) and FindFirstConflict (FFC), which exploit SIMD parallelism within the multi-robot domain. On pure multi-robot motion validation tests, this achieves over 1100X speedup in validation time. Additionally, we modify a representative set of MRMP algorithms to use these new primitives. The relative speedup for each algorithm is studied on scenarios with manipulator, rigid body, and heterogeneous teams with some instances producing multi-robot solutions in the order of milliseconds and, in many cases, shows planning time speedups of over 850X.