Kino-PAX+: Near-Optimal Massively Parallel Kinodynamic Sampling-based Motion Planner¶

Authors: Nicolas Perrault, Qi Heng Ho, Morteza Lahijanian

Published: 2026 (Preprint)

Source: arXiv

Algorithm: Kino-PAX+

arXiv: 2602.02846

Summary¶

Extends Kino-PAX with asymptotic near-optimal guarantees by focusing each GPU thread block's computation on the most promising nodes within local neighborhoods for propagation and refinement.

Abstract¶

Sampling-based motion planners (SBMPs) are widely used for robot motion planning with complex kinodynamic constraints in high-dimensional spaces, yet they struggle to achieve \emph{real-time} performance due to their serial computation design. Recent efforts to parallelize SBMPs have achieved significant speedups in finding feasible solutions; however, they provide no guarantees of optimizing an objective function. We introduce Kino-PAX^{+}, a massively parallel kinodynamic SBMP with asymptotic near-optimal guarantees. Kino-PAX^{+} builds a sparse tree of dynamically feasible trajectories by decomposing traditionally serial operations into three massively parallel subroutines. The algorithm focuses computation on the most promising nodes within local neighborhoods for propagation and refinement, enabling rapid improvement of solution cost. We prove that, while maintaining probabilistic \delta-robust completeness, this focus on promising nodes ensures asymptotic \delta-robust near-optimality. Our results show that Kino-PAX^{+} finds solutions up to three orders of magnitude faster than existing serial methods and achieves lower solution costs than a state-of-the-art GPU-based planner.

Links¶

Primary

Paper arxiv.org

Standard

arXiv Abstract 2602.02846 arXiv PDF 2602.02846 arXiv HTML 2602.02846