摘要
针对传统人工势场法易陷入局部最小值的问题,依据破坏智能体在非目标点时的平衡状态的思路,提出斥力衰减权重的人工势场改进算法。该算法为智能体在探测范围内遇到的障碍物分配权重,并将此权重应用到障碍物对智能体的斥力计算中,权重随着智能体的运动而衰减,确保智能体在目标点之前避免产生合力相等的情况,从而走出局部极小值点。仿真实验结果表明,与传统人工势场法及其他解决局部最小问题的随机扰动法和边缘行走法相比,所提算法解决了局部最小值问题,在收敛速度、稳定性和能耗方面均有显著改善。
Since the traditional artificial potential field method is prone to falling into local minima,an improved artificial potential field algorithm which introduces repulsion attenuation weight is proposed according to the idea of destroying the equilibrium state of agents at non-targets.This algorithm assigns weights to obstacles encountered within the agent's detection range and applies these weights to the calculation of repulsive forces exerted by the obstacles.The weights decay as the agent moves,ensuring that the net force acting on the agent does not balance out before reaching the target,thereby allowing the agent to escape local minima.The results of the simulation experiments demonstrate that,in comparison with the traditional artifi-cial potential field method and other methods avoiding local minima such as the random disturbance method and the edge-fol-lowing method,the proposed algorithm avoids the local minima effectively and shows significant improvements in convergence speed,stability and energy consumption.
作者
何兴
柏艳红
HE Xing;BAI Yanhong(School of Electronic Information Engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China;College of Intelligent Manufacturing Industry,Shanxi University of Electronic Science and Technology,Linfen 041000,China)
出处
《现代电子技术》
北大核心
2025年第17期181-186,共6页
Modern Electronics Technique
基金
山西电子科技学院人才引进科研启动基金资助项目(2023RKJ018)。
关键词
人工势场法
多智能体编队
局部最小值
衰减权重
斥力计算
权重分配
artificial potential field method
multi-agent formation
local minima
attenuation weight
repulsive force calcula-tion
weight allocation
