Collaborative Motion Planning for Multiple Tractor-Trailer Vehicles Based on Local Conflict Search and Priority Game Inheritance

Abstract

This paper introduces a novel approach for efficiently planning collision-free optimal trajectories for multiple tractor-trailer vehicles (TTVs) in environments with dense obstacles. The approach employs a hierarchical planning strategy to generate the initial homotopy path: at the higher level, it integrates body collision model and enhances conflict resolution through local priority strategy. At the lower level, the Tractor-Trailer Spatiotemporal Hybrid State A* (TT-SHA*) is employed, combining joint expansion with heuristic methods to efficiently generate paths for individual agents, while accounting for underactuated and nonholonomic constraints. To resolve priority deadlock issues, the concept of priority inheritance and backtracking is integrated into conflict search, which significantly improves the success rate and solving speed. Finally, it refines the path solution into smooth trajectories using nonlinear optimization based on corridor and relative safety distance constraints. Experimental results show that the proposed method outperforms existing TTV-improved Multi-Agent Motion Planning algorithms in scalability when handling large-scale TTV scenarios, providing a solution that is applicable for practical implementation.