A Dual-Path Planning and Obstacle Avoidance Method for Multi-mobile Composite Robots Based on Zone-Based Picking

As intelligent warehousing logistics systems evolve toward higher automation and intelligence, the Zone-Based Task Allocation (ZTA) process enhances order processing efficiency and resource utilization through a ’zone-responsible, order-decomposed, parallel-coordinated’ mechanism. However, this approach also gives rise to two interrelated path-planning challenges: one is how to efficiently visit multiple task points within each sub-area, and the other is how to achieve collaborative obstacle avoidance among multiple mobile composite robots in the public passages outside the areas. Traditional single-path planning methods struggle to jointly optimize this hierarchical problem, often leading to system inefficiency, frequent path conflicts, or insufficient dynamic adaptability. To address this, this paper proposes a dual-path planning and obstacle avoidance framework with a hierarchical optimization strategy. Within zones, it integrates the A* global planning algorithm with the Dynamic Window Approach (DWA) to achieve local obstacle avoidance while approximating the global path. Outside zones, it combines the Ant Colony Optimization (ACO) for global search with DWA for dynamic adjustment, while introducing a conflict prediction model and spatiotemporal window synchronization mechanism to resolve multi-robot deadlocks and collisions. Experimental results demonstrate that the proposed method significantly outperforms traditional approaches in key metrics, including order completion time, robot utilization, and path conflict rate, validating its effectiveness and robustness in highly dynamic warehousing environments. This study provides an efficient and safe path planning solution for multi-composite robot systems under the ZTA process, offering theoretical value and engineering application prospects for intelligent warehousing systems.