Configuration Space Decomposition for Scalable Proxy Collision Checking in Robot Planning and Control

TL;DR

This paper introduces D-Fastron, a space decomposition method using K-Means clustering and Fastron models to significantly accelerate collision checking in high-dimensional robot motion planning, achieving up to 29x speedup.

Contribution

The paper proposes a novel space decomposition approach combined with Fastron models to improve the speed and scalability of collision checking in complex robot environments.

Findings

Achieves 29x faster collision checks on a 7-DOF robot

Up to 9.8x faster motion planning compared to existing methods

Scales better to high-dimensional, complex environments

Abstract

Real-time robot motion planning in complex high-dimensional environments remains an open problem. Motion planning algorithms, and their underlying collision checkers, are crucial to any robot control stack. Collision checking takes up a large portion of the computational time in robot motion planning. Existing collision checkers make trade-offs between speed and accuracy and scale poorly to high-dimensional, complex environments. We present a novel space decomposition method using K-Means clustering in the Forward Kinematics space to accelerate proxy collision checking. We train individual configuration space models using Fastron, a kernel perceptron algorithm, on these decomposed subspaces, yielding compact yet highly accurate models that can be queried rapidly and scale better to more complex environments. We demonstrate this new method, called Decomposed Fast Perceptron (D-Fastron), on the 7-DOF Baxter robot producing on average 29x faster collision checks and up to 9.8x faster motion planning compared to state-of-the-art geometric collision checkers.