Integrated reinforcement learning of automated guided vehicles dynamic path planning for smart logistics and operations
提出一种强化学习方法，用于智能仓库中自动导引车的实时动态路径规划，通过集成物联网和人工智能优化多车协调与电池管理，仿真验证了其在复杂动态场景下的鲁棒性和效率。
• We propose a reinforcement learning approach for dynamic AGV path planning. • The RL model enhances decision-making in smart logistics environments. • Our framework integrates AGV path planning with IoT and AI for smart warehouses. • Simulations demonstrate robustness and efficiency under complex dynamic scenarios. • The proposed system optimizes multi-AGV coordination and battery management. Automated guided vehicles (AGV) play a critical role in fostering a smarter logistics and operations environment. Conventional path planning for AGVs enables the load-in-load-out of the items, but existing approaches rarely consider dynamic integrations with smart warehouses and factory systems. Therefore, this study presents a reinforcement learning (RL) approach for real-time path planning in automated guided vehicles within smart warehouses or smart factories. Unlike conventional path planning methods, which struggle to adapt to dynamic operational changes, the proposed algorithm integrates real-time information to enable responsive and flexible routing decisions. The novelty of this study lies in integrating AGV path planning and RL within a dynamic environment, such as a smart warehouse containing various workstations, charging stations, and storage locations. Through various scenarios in smart factory settings, this research demonstrates the algorithm’s effectiveness in handling complex logistics and operations environments. This research advances AGV technology by providing a scalable solution for dynamic path planning, enhancing efficiency in modern industrial systems.