Efficient Probabilistic Collision Detection for Non-Convex Shapes

TL;DR

This paper introduces efficient algorithms for probabilistic collision detection applicable to both convex and non-convex shapes, utilizing Gaussian representations and hierarchical models to improve speed and accuracy in complex scenarios.

Contribution

The paper develops novel algorithms for probabilistic collision detection that extend from convex to non-convex shapes using hierarchical representations, enhancing performance.

Findings

Algorithms perform well on synthetic benchmarks

Effective in trajectory planning for robotic arms

Applicable to general shape models

Abstract

We present new algorithms to perform fast probabilistic collision queries between convex as well as non-convex objects. Our approach is applicable to general shapes, where one or more objects are represented using Gaussian probability distributions. We present a fast new algorithm for a pair of convex objects, and extend the approach to non-convex models using hierarchical representations. We highlight the performance of our algorithms with various convex and non-convex shapes on complex synthetic benchmarks and trajectory planning benchmarks for a 7-DOF Fetch robot arm.