基于双种群蚁群算法的AGV路径规划研究
2023,31(5):193-199
摘要：针对蚁群算法存在的收敛速度慢、易陷入局部最优和容易死锁等问题,提出了一种用于自动引导车(Automated Guided Vehicle, AGV)路径规划的双种群蚁群算法。该算法引入差异化信息素初始值,修改启发函数并在信息素更新时对最优及最差路径进行奖惩；以改进策略为基础,引入自适应步长搜索策略,通过具有差异化步长的两个种群相互协作加强算法寻优能力和搜索效率；针对死锁问题,提出了将符合条件的单元格视为障碍物的“填充陷阱”策略。分别进行仿真实验和车间现场实验,结果表明,该算法可以为AGV规划出一条安全且综合性能较好的路径,为AGV路径规划提供了一种可行的方案。
关键词：AGV;路径规划;蚁群算法;自适应步长;双种群;死锁避免
Research on path planning of AGV based on dual population ant colony algorithm
Abstract：Aiming at the problems of Ant Colony Optimization, such as slow convergence, easy to fall into local optimum and easy to deadlock, a dual-population ant colony algorithm for path planning of Automatic Guided Vehicle (AGV) is proposed. The algorithm introduces different pheromone initial values, modifies the heuristic function and rewards and punishes the best and worst paths when pheromone is updated; Based on the improved strategy, the adaptive step-size search strategy is introduced, and the optimization ability and search efficiency of the algorithm are strengthened by the cooperation of two populations with different step sizes; To solve the deadlock problem, a strategy of "filling traps" is proposed, which regards qualified cells as obstacles. Simulation experiments and field experiments are carried out respectively. The results show that the algorithm can plan a safe and comprehensive path for AGV, and provides a feasible scheme for AGV path planning.
Key words：AGV; path planning; ant colony algorithm; adaptive step size; dual population; deadlock avoidance
收稿日期：2022-09-13
基金项目：
