动态混沌蚁群系统及其在机器人路径规划中的应用

针对蚁群系统（ACS）解决机器人路径规划问题时种群多样性与收敛速度的不足，对蚁群系统引入动态混沌算子，从而平衡种群多样性和收敛速度之间的关系。动态混沌蚁群系统的核心是在传统蚁群系统引入Logistic混沌算子来增加种群多样性，从而提高解的质量。在迭代前期加入混沌算子，以调整路径中的全局信息素值，增加算法的种群多样性，从而避免算法陷入局域优化解；在后期则转为蚁群系统，来确保动态混沌蚁群系统的收敛速度。仿真结果表明，对于机器人路径规划问题，与蚁群系统相比，动态混沌蚁群系统具有更好的种群多样性、更高的解的质量和更快的收敛速度；与精英蚁群系统（EAS）和基于排序的蚂蚁系统（ASrank）相比，动态混沌蚁群系统能够平衡解的质量与收敛速度之间的关系，即使在复杂障碍物的环境下，动态混沌蚁群系统也能较好地找到最优解。动态混沌蚁群系统能够提升移动机器人路径规划中的效率。

Dynamic chaotic ant colony system and its application in robot path planning

To solve problems of population diversity and convergence speed when an Ant Colony System (ACS) is used to robot path planning, a dynamic chaos operator was introduced in the ACS. The dynamic chaotic ACS can balance population diversity and convergence speed. The core of dynamic chaotic ACS is that a Logistic chaotic operator was added to the traditional ACS to increase population diversity and improve the quality of the solutions. First, the chaotic operator was added to the pre-iteration to adjust the global pheromone value in the path to increase the population diversity of the algorithm, so as to avoid the algorithm to fall into the local optimal solution. Then, in the later stage, the ACS was used to ensure convergence speed of the dynamic chaotic ACS. The experimental results show that the dynamic chaotic ACS has better population diversity compared with the ACS for the robot path planning problem. The solution quality is higher and the convergence speed is faster. Compared with the Elitist Ant colony System (EAS) and the rank-based Ant System (ASrank), the dynamic chaotic ACS can balance the relationship between the quality of the solutions and the convergence speed. The dynamic chaotic ACS can find better optimal solutions even in the complex obstacle environment. The dynamic chaotic ACS can improve the efficiency of mobile robot path planning.