基于改进遗传算法的多AGV两阶段调度方法

Multi-AGV two-stage scheduling method based on improved genetic algorithm

为了提高多AGV无碰撞路径规划方案的有效性及运算效率,提出一种基于改进遗传算法的两阶段调度方法。采用蚁群算法得到的精英种群作为遗传算法的初始种群,提升算法精度和效率,将改进的遗传算法与时间窗模型结合,提前预判多AGV路径上节点占用情况,根据生产任务需求和AGV优先级,对冲突做出快速反应,有效避免多AGV运行时的碰撞冲突。将研究结果应用到J公司电驱动系统装配车间多AGV线边物流运输,论文所提两阶段调度方法有效、模型运算效率显著提高。

With the development of automated logistics technology,AGV has been widely used in industrial production and logistics field.The collision-free path planning of multiple AGVs is a research hotspot and difficulty.In order to improve the effectiveness and computing efficiency of the collision-free path planning scheme of multiple AGVs,a two-stage scheduling method based on improved genetic algorithm is proposed.The elite population obtained by ant colony algorithm is used as the initial population of the genetic algorithm to improve the accuracy and efficiency of the algorithm.The improved genetic algorithm is combined with the time window model to predict the node occupancy on the path of multiple AGVs in advance.According to the production task requirements and AGV priority,the conflict is quickly responded to,and the collision problem of multiple AGVs is effectively avoided.The research results are applied to the multi-AGV line-side logistics transportation of the electric drive system assembly workshop of Company J.The results show that the two-stage scheduling method proposed in this paper is effective and the computing efficiency of the model is significantly improved.