Journal of Frontiers of Computer Science and Technology ›› 2022, Vol. 16 ›› Issue (12): 2820-2831.DOI: 10.3778/j.issn.1673-9418.2107086

• Artificial Intelligence • Previous Articles     Next Articles

Single-Colony Adaptive Heterogeneous Ant Colony Algorithm for Mobile Robot Path Planning

ZHANG Songcan1,2, SUN Lifan1,+(), SI Yanna1, PU Jiexin1   

  1. 1. School of Information Engineering, Henan University of Science & Technology, Luoyang, Henan 471023, China
    2. School of Electrical Engineering, Henan University of Science & Technology, Luoyang, Henan 471023, China
  • Received:2021-07-22 Revised:2021-09-29 Online:2022-12-01 Published:2021-10-18
  • About author:ZHANG Songcan, born in 1973, Ph.D. candidate. His research interests include swarm intelligence algorithms and intelligent control of mobile robot.
    SUN Lifan, born in 1982, Ph.D., associate professor. His research interests include extended target tracking and information fusion.
    SI Yanna, born in 1990, Ph.D. candidate. Her research interests include reinforcement learning and intelligent control of robot.
    PU Jiexin, born in 1959, Ph.D., professor, Ph.D. supervisor. His research interests include intelligent information processing and intelligent control of robot.
  • Supported by:
    National Natural Science Foundation of China(U1504619);National “Thirteen-Five” Equipment Pre-Research Foundation of China(61403120207);National “Thirteen-Five” Equipment Pre-Research Foundation of China(61402100203);Aeronautical Science Foundation of China(20185142003);Program of Science and Technology Innovative Talents in Universities of Henan Province(21HASTIT030);Program of Young Backbone Teachers in Universities of Henan Province(2020GGJS073);Program of Leading Talents of Science and Technology Innovation in the Central Plains of China(194200510012)

单种群自适应异构蚁群算法的机器人路径规划

张松灿1,2, 孙力帆1,+(), 司彦娜1, 普杰信1   

  1. 1.河南科技大学 信息工程学院,河南 洛阳 471023
    2.河南科技大学 电气工程学院,河南 洛阳 471023
  • 通讯作者: +E-mail: lifan.sun@gmail.com
  • 作者简介:张松灿(1973—),男,河南郑州人,博士研究生,主要研究方向为群智能算法、移动机器人智能控制。
    孙力帆(1982—),男,河南洛阳人,博士,副教授,主要研究方向为扩展目标跟踪、信息融合。
    司彦娜(1990—),女,河南郑州人,博士研究生,主要研究方向为强化学习、机器人智能控制。
    普杰信(1959—),男,河南周口人,博士,教授,博士生导师,主要研究方向为智能信息处理、机器人智能控制。
  • 基金资助:
    国家自然科学基金(U1504619);国家“十三五”装备预研领域基金资助项目(61403120207);国家“十三五”装备预研领域基金资助项目(61402100203);航空科学基金资助项目(20185142003);河南省高校科技创新人才资助项目(21HASTIT030);河南省高等学校青年骨干教师资助项目(2020GGJS073);中原科技创新领军人才资助项目(194200510012)

Abstract:

Aiming at the existent problems of complex structure, slow optimization rate and inadequate adaptability of multi-colony ant colony algorithm, a single colony adaptive heterogeneous ant colony algorithm is proposed to solve mobile robot path planning. The proposed algorithm employs a single colony to avoid the problem of complex structure of multi-ant colony algorithm. Each ant in the colony has its own control parameters to realize the hete-rogeneous behavior and increase the diversity of the population. In the first iteration, only the heuristic factor is used to construct the candidate solution, which improves the quality of the initial population. The information exchange period is adaptively determined based on the change of population information entropy to enhance the adaptability of the algorithm. The information exchange strategy transfers the control parameters of the optimal ant to the worst ant to enhance the guiding role of the optimal ant. Parameter mutation operation is helpful for exploring better algo-rithm parameters in a larger parameter space and improving the ability to escape from local optimization. The results of simulation experiments and statistical tests verify the effectiveness, stability and superiority of the pro-posed algorithm.

Key words: colony algorithm, population information entropy, mobile robot, path planning, adaptive information exchange period, state transition rule, parameter mutation

摘要:

针对多种群蚁群算法存在结构复杂、优化速度慢及适应性不足等问题,提出一种单种群自适应异构蚁群算法,并用于机器人路径规划。该算法采用单种群结构,避免多蚁群算法结构复杂的问题;种群内每只蚂蚁都有自己的控制参数,实现蚂蚁的行为异构,增加种群的多样性;在首次迭代时,仅用启发因子构建候选解,提高了初始化种群的质量;根据蚁群优化过程中种群信息熵的变化,自适应确定信息交换周期;所设计的信息交换策略将最优蚂蚁的控制参数传递给最差蚂蚁,增强最优蚂蚁的引导作用;参数突变操作有助于在更大的参数空间探索更优的控制参数,提高算法逃离局部最优的能力。仿真实验与统计检验的结果验证了所提算法的有效性、稳定性和优越性。

关键词: 蚁群算法, 种群信息熵, 移动机器人, 路径规划, 自适应信息交换周期, 状态转移规则, 参数突变

CLC Number: