Path Planning Research for Textile Warehouse AGV Based on Integrated Improved A* and DWA Algorithms

Abstract

To address the limitations of traditional A* algorithms in textile warehouse automated guided vehicle (AGV) navigation, including paths too close to obstacles, low search efficiency caused by redundant nodes, and frequent turning that reduces motion smoothness, this paper proposes an integrated path planning scheme combining an improved A* algorithm and an improved dynamic window approach (DWA). The method is designed for AGV navigation in apparel, silk, and fabric warehousing environments where dense storage layouts, dynamic obstacles, and electromagnetic or wireless sensing constraints require safe and stable motion planning. First, the evaluation function of the A* algorithm is improved by introducing an obstacle-density factor and kinematic constraints, enabling adaptive heuristic weighting, safer child-node screening, and smoother global reference paths. Second, the DWA evaluation function is modified by incorporating global path guidance, obstacle clearance, velocity, and smoothness-related weights, improving local obstacle avoidance decisions under dynamic conditions. Finally, a global-local coordination mechanism is developed so that the improved A* algorithm provides the optimized path skeleton and the improved DWA performs real-time local tracking and dynamic avoidance. Simulation experiments in typical indoor grid environments containing static and dynamic obstacles show that the proposed algorithm reduces planning time by approximately 17%, improves average motion smoothness by about 49%, reduces cumulative turning angles, and maintains safe obstacle clearance. The results demonstrate that the integrated method improves the efficiency, safety, and trajectory quality of textile warehouse AGV navigation.