Edge Nearest Neighbor in Sampling-Based Motion Planning
🤖 AI Summary
Sampling-based motion planners suffer from inefficient narrow-passage exploration due to neighborhood search being restricted to discrete vertices, neglecting the continuous geometric structure of tree edges.
Method: This work introduces, for the first time, continuous-edge nearest-neighbor queries in RRT/RRG frameworks—extending search beyond vertices to the entire edge manifolds. We propose a continuous-edge projection mechanism for nearest-neighbor computation, design an RRG variant tailored to this mechanism, and employ hierarchical spatial data structures with Euclidean distance metrics to accelerate queries.
Contribution/Results: We prove that the approach preserves probabilistic completeness and asymptotic optimality, with a strictly faster convergence rate than conventional vertex-only methods. Experiments demonstrate significantly improved narrow-passage traversal success rates and an average 23.6% reduction in total planning time. The core innovation lies in generalizing neighborhood search from discrete vertices to continuous edge spaces, establishing a new geometric modeling paradigm for sampling-based planning.
Application Category
📝 Abstract
Neighborhood finders and nearest neighbor queries are fundamental parts of sampling based motion planning algorithms. Using different distance metrics or otherwise changing the definition of a neighborhood produces different algorithms with unique empiric and theoretical properties. In cite{l-pa-06} LaValle suggests a neighborhood finder for the Rapidly-exploring Random Tree RRT algorithm cite{l-rrtnt-98} which finds the nearest neighbor of the sampled point on the swath of the tree, that is on the set of all of the points on the tree edges, using a hierarchical data structure. In this paper we implement such a neighborhood finder and show, theoretically and experimentally, that this results in more efficient algorithms, and suggest a variant of the Rapidly-exploring Random Graph RRG algorithm cite{f-isaom-10} that better exploits the exploration properties of the newly described subroutine for finding narrow passages.
Problem
Research questions and friction points this paper is trying to address.
Implementing efficient nearest neighbor finder for RRT algorithm
Exploring impact of distance metrics on motion planning algorithms
Enhancing RRG algorithm with improved narrow passage exploration
Innovation
Methods, ideas, or system contributions that make the work stand out.
Hierarchical data structure for nearest neighbor
Edge-based neighborhood finder in RRT
Enhanced RRG algorithm for narrow passages
🔎 Similar Papers
No similar papers found.
