Motions in Microseconds via Vectorized Sampling-Based Planning

TL;DR

This paper introduces a method to accelerate sampling-based motion planning algorithms by over 500 times, achieving microsecond-level planning speeds suitable for real-time robotic control without specialized hardware.

Contribution

The paper presents a general approach to exploit parallelism in sampling-based planners, significantly speeding up critical computations like collision checking and forward kinematics.

Findings

Achieved over 500x speedup in planning times

Demonstrated microsecond planning on complex robots

Validated on low-power hardware

Abstract

Modern sampling-based motion planning algorithms typically take between hundreds of milliseconds to dozens of seconds to find collision-free motions for high degree-of-freedom problems. This paper presents performance improvements of more than 500x over the state-of-the-art, bringing planning times into the range of microseconds and solution rates into the range of kilohertz, without specialized hardware. Our key insight is how to exploit fine-grained parallelism within sampling-based planners, providing generality-preserving algorithmic improvements to any such planner and significantly accelerating critical subroutines, such as forward kinematics and collision checking. We demonstrate our approach over a diverse set of challenging, realistic problems for complex robots ranging from 7 to 14 degrees-of-freedom. Moreover, we show that our approach does not require high-power hardware by also evaluating on a low-power single-board computer. The planning speeds demonstrated are fast enough to reside in the range of control frequencies and open up new avenues of motion planning research.