基于改进蚁群算法的AGV路径规划研究

摘要:

为解决蚁群算法在实现AGV路径规划时存在迭代速度慢、初期路径搜索盲目性大、路径拐点数量多、安全性较低等问题，提出一种改进的蚁群算法。该方法以栅格地图为AGV运行环境，在迭代初期引入势场力，将当前位置与目标点的势场力加入启发式信息中，解决路径初期搜索盲目性和算法迭代速度慢的问题；通过改进算法状态选择概率，提高获取优质解能力，避免算法陷入局部最优；提出一种基于路径长度、安全性和平滑性多目标约束的信息素更新规则，提高AGV行驶安全性；在此基础上，引入三次B样条路径平滑策略，使规划路径满足AGV需求，通过仿真实验可知，改进算法在收敛速度和稳定性方面表现效果较好，其收敛速度相较于传统算法提升8倍，路径长度较其他改进算法提升接近20%。

Abstract:

In order to solve the problems of slow convergence speed, large blindness in initial path search, multiple paths turning points, and low safety in the implementation of AGV path planning using ant colony algorithm, an improved ant colony algorithm is proposed. This method takes the grid map as the running environment for AGV, introduces the potential field force at the beginning of the iteration, and adds the potential field force between the current position and the target point to the heuristic information to solve the problems of blindness in initial path search and slow convergence speed of the algorithm. By improving the probability of algorithm state selection, the ability to obtain high-quality solutions is improved, and the algorithm is prevented from falling into local optimality. An information pheromone update rule based on multiple objective constraints such as path length, safety, and smoothness are proposed to improve the safety of AGV travel. On this basis, a cubic B-spline path smoothing strategy is introduced to make the planned path meet the requirements of AGV. Through simulation experiments, the improved algorithm has better performance in terms of convergence speed and stability. Its convergence speed is improved by 8 times compared with traditional algorithms, and the path length is improved by nearly 20% compared with other improved algorithms.