摘要
针对机器人路径规划的传统ACO(蚁群)算法存在迭代次数多、收敛速度慢、容易陷入局部最优和出现死锁状态等问题,提出一种改进的ACO算法。结合PSO(粒子群优化)算法对传统ACO算法进行改进,减少迭代次数,提高收敛速度;通过增加随机性来扩大搜索范围,避免局部最优;利用蚂蚁回退策略解决死锁状态问题。仿真实验结果表明:改进ACO算法可以快速准确地搜索到最优路径,具有很好的寻优与避障能力。
In view of the problems of traditional ACO(ant colony optimization) of robot path planning, such as multiple iterations, slow rate of convergence, proneness to local optimum and deadlock state, an improved ACO is proposed. Combined with PSO(particle swarm optimization), the traditional ACO is improved to reduce the iterations and improve the rate of convergence. By randomness increase, the search scope is expanded to avoid local optimum. The ant fallback strategy is used to solve the deadlock state problem. The simulation results show that the improved ACO can quickly and accurately obtain the optimal path, and has good performance in optimization and obstacle avoidance.
作者
刘胜
晏齐忠
张志鑫
张豪
申永鹏
LIU Sheng;YAN Qizhong;ZHANG Zhixin;ZHANG Hao;SHEN Yongpeng(School of Electrical Information Engineering,Zhengzhou University of Light Technology,Zhengzhou 450002,China)
出处
《浙江电力》
2021年第1期29-35,共7页
Zhejiang Electric Power
基金
国家自然科学基金青年项目(61803345)
河南省科技攻关项目(202102210303)。
