摘要
针对标准樽海鞘群算法易陷入局部最优的问题,提出高斯分布策略的樽海鞘群算法。首先在保留领导者位置更新公式的基础上,将基于全局最优位置的高斯分布引入追随者位置更新公式中进行改进,通过优化种群,增强算法的局部开发能力,提高算法准确性。其次,与标准樽海鞘群算法、传统粒子群算法在CEC2005测试函数中的10个不同特征的标准测试函数上进行测试对比,并通过最优值、标准差、平均值、中值、最差值、收敛性分析和Wilcoxon检验评估改进算法的性能,理论分析证明:改进算法能显著提高寻优能力、收敛性能。此外,采用3种算法与三次样条插值相结合的方法,使路径规划更平滑,并在3种不同障碍物测试环境中进行对比仿真实验,验证了改进算法的适应性和有效性。
Aiming at the problem that the standard salp swarm algorithm is prone to getting stuck in local optima,a Gaussian distribution strategy salp swarm algorithm is proposed.Firstly,on the basis of retaining the leader position update formula,the Gaussian distribution based on the global optimal position is introduced into the follower position update formula for improvement.By optimizing the population,the local development ability of the algorithm is enhanced,and the accuracy of the algorithm is improved.Secondly,the improved algorithm is tested and compared with the standard salp swarm algorithm and traditional particle swarm algorithm on 10 different features of the CEC2005 test function.The performance of the improved algorithm is evaluated through optimal value,standard deviation,mean,median,worst value,convergence analysis,and Wilcoxon test.Theoretical analysis proves that the improved algorithm can significantly improve the optimization ability and convergence performance.In addition,a combination of three algorithms and cubic spline interpolation is used to make path planning smoother.Comparative simulation experiments are conducted in three different obstacle testing environments,and the adaptability and effectiveness of the improved algorithm are verified.
作者
王亚玲
马冬则
乔丹妮
卫娟
刘云
Wang Yaing;Ma Dongze;Qiao Danni;Wei Juan;Liu Yun(School of Information and Communication Engineering,Shanxi University of Electronic Science and Technology,Linfen,Shanxi 041000,China;School of New Energy and Materials Engineering,Shanxi University of Electronic Science and Technology,Linfen,Shanxi 041000,China;School of Microelectronics,Shanxi University of Electronic Science and Technology,Linfen,Shanxi 041000,China)
出处
《机电工程技术》
2025年第13期117-122,154,共7页
Mechanical & Electrical Engineering Technology
基金
山西省高等学校科技创新项目(2023L452)。
关键词
樽海鞘群算法
高斯分布
三次样条插值
机器人路径规划
salp swarm algorithm
Gaussian distribution
cubic spline interpolation
robot path planning
作者简介
王亚玲(1995-),女,山西平遥人,硕士研究生,助教,研究领域为智能信号处理。;马冬则(1994-),男,山西方山人,硕士研究生,助教,研究领域为气敏材料的制备及性能。;乔丹妮(1995-),女,山西霍州人,硕士研究生,助教,研究领域为多智能系统的有限时间一致性。;卫娟(1996-),女,山西襄汾人,硕士研究生,助教,研究领域为光子晶体光纤。;刘云(1967-),男,山西应县人,博士,教授,研究领域为铝合金用电磁泵铸造系统。
