Answers are grounded in this paper and cite their sources.
Warehouse logistics environments contain dense obstacles and narrow passages and impose strict real-time requirements on mobile-robot navigation. This study proposes a BIRRT*-MPPI hierarchical path-planning method to address the slow convergence and redundant paths of conventional RRT* global planning and the limited local adaptability of MPPI in cluttered environments. At the global level, bidirectional RRT* is combined with an environment-perception model, a density-aware dynamic goal-bias strategy, and an on-the-fly inter-tree connection mechanism. The resulting reference path is refined using line-of-sight pruning and spline smoothing. At the local level, an MPPI controller uses an obstacle distance field and a weighted multi-objective cost function under kinematic and control constraints. In the complex maze environment, compared with RRT*, RRT*-APF, and RRT*-DWA, the proposed method reduced the average runtime by 77.8%, 65.7%, and 63.8%, respectively; reduced the average iteration count by 75.2%, 58.5%, and 71.9%, respectively; and shortened the average path length by 18.7%, 2.4%, and 2.1%, respectively. Across four static simulation environments, the proposed method generally achieved shorter planning times, fewer iterations, and shorter paths than the comparison methods. These results demonstrate the potential of the hierarchical planner for wheeled mobile-robot navigation in structured warehouse environments.
No takes yet. Share an insight, caveat, or question.
Yali Xi (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: