MeshA*: Efficient Path Planning With Motion Primitives

TL;DR

MeshA* is a novel path planning algorithm that efficiently searches over motion primitives by integrating them into grid cell searches, significantly reducing runtime while maintaining optimality guarantees.

Contribution

The paper introduces MeshA*, a new algorithm that improves lattice-based path planning efficiency by combining primitive sequences with grid cell searches, preserving completeness and optimality.

Findings

MeshA* reduces planning runtime by 1.5 to 2 times compared to traditional methods.

The algorithm maintains guarantees of completeness and optimality.

Experimental results demonstrate significant efficiency improvements.

Abstract

We study a path planning problem where the possible move actions are represented as a finite set of motion primitives aligned with the grid representation of the environment. That is, each primitive corresponds to a short kinodynamically-feasible motion of an agent and is represented as a sequence of the swept cells of a grid. Typically, heuristic search, i.e. A*, is conducted over the lattice induced by these primitives (lattice-based planning) to find a path. However, due to the large branching factor, such search may be inefficient in practice. To this end, we suggest a novel technique rooted in the idea of searching over the grid cells (as in vanilla A*) simultaneously fitting the possible sequences of the motion primitives into these cells. The resultant algorithm, MeshA*, provably preserves the guarantees on completeness and optimality, on the one hand, and is shown to notably outperform conventional lattice-based planning (x1.5-x2 decrease in the runtime), on the other hand.