Abstract: In flexible manufacturing system, Automated Guided Vehicles(AGV)could not operate in planned routes of time window under dynamic uncertain environment. Aiming at this problem, the updating AGV operating pass was proposed by changing the priority of AGV passing through the nodes and adjusting the passing sequences of corresponding nodes. Meanwhile, time-window algorithm was improved to realize multiple AGV route planning in dynamic environment. In simulation experiment, rain line system with eight AGV was used to compare the unimproved time-window algorithm with the improved time-window algorithm, and the realization process of the algorithms was illustrated. Through the simulation, the proposed algorithm was proved to have better robustness and flexibility, and could reduce the number of conflicts, realize conflict-free effectively and improve system efficiency.

Key words: time window, automated guided vehicle, real-time conflict-free, path planning

摘要： 针对柔性制造系统中的自动导引车在动态不稳定环境下不能按照时间窗规划好的既定路径运行的问题，提出通过实时改变自动导引车通过节点的优先级，调整相应节点的自动导引车的通过顺序来更新自动导引车运行路径，对时间窗算法进行改进，从而实现多自动导引车动态环境下的路径规划。以含有8台自动导引车的淋雨线系统为应用案例进行仿真实验，对改进前的时间窗算法和改进后的时间窗算法进行对比，说明了算法的实现过程。通过仿真证明了该算法能减少冲突数目，有效实现多自动导引车避碰，具有更好的鲁棒性和柔性，同时能够提高系统效率。

关键词: 时间窗, 自动导引车, 实时避碰, 路径规划