Improving motion-planning algorithms by efficient nearest-neighbor searching

Anna Yershova, Steven M. LaValle

Research output: Contribution to journalArticle

Abstract

The cost of nearest-neighbor (NN) calls is one of the bottlenecks in the performance of sampling-based motion-planning algorithms. Therefore, it is crucial to develop efficient techniques for NN searching in configuration spaces arising in motion planning. In this paper, we present and implement an algorithm for performing NN queries in Cartesian products of ℝ, S1, and ℝP3, the most common topological spaces in the context of motion planning. Our approach extends the algorithm based on kd-trees, called ANN, developed by Arya and Mount for Euclidean spaces. We prove the correctness of the algorithm and illustrate substantial performance improvement over the brute-force approach and several existing NN packages developed for general metric spaces. Our experimental results demonstrate a clear advantage of using the proposed method for both probabilistic roadmaps and rapidly exploring random trees.

Original languageEnglish (US)
Pages (from-to)151-157
Number of pages7
JournalIEEE Transactions on Robotics
Volume23
Issue number1
DOIs
StatePublished - Feb 1 2007

    Fingerprint

Keywords

  • Configuration space
  • Kd-trees
  • Nearest-neighbor (NN) searching
  • Probabilistic roadmaps (PRMs)
  • Rapidly exploring random trees (RRTs)
  • Sampling-based motion planning

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this