摘要
针对蚁群算法(ACO)在变电站巡检机器人路径规划中存在的安全性不足、路径不平滑及多目标优化失衡等问题,提出一种改进的多目标路径规划方法。在启发式函数中引入安全距离约束与惩罚项,使机器人主动避开高压设备危险区域;结合转向代价因子,降低路径转向频率,增强运动学约束适应性。设计动态多样性控制的精英蚂蚁策略,平衡路径质量与多样性,避免陷入局部最优。构建多目标优化框架,综合路径长度、转向次数与安全性,采用时变权重平衡探索与开发。MATLAB仿真结果表明,该算法相较其他改进算法,路径长度缩短61%、搜索时间降低49%、转向次数减少71%,显著提升了规划效率与路径安全性,并在实际机器人应用中验证了其可行性。
To address the issues of insufficient safety,poor path smoothness,and imbalance in multi-objective optimiza-tion in the ACO algorithm for substation inspection robot path planning,an improved multi-objective path planning method is proposed.A safety distance constraint and penalty term are introduced into the heuristic function to enable the robot to actively avoid hazardous areas around high-voltage equipment.A turning cost factor is incorporated to reduce path turning frequency and enhance kinematic adaptability.A dynamically controlled elite ant strategy is de-signed to balance path quality and population diversity,preventing premature convergence.Furthermore,a multi-objec-tive optimization framework is constructed by integrating path length,turning number,and safety indicators,with a time-varying weight adjustment strategy to balance exploration and exploitation.MATLAB simulation results show that compared with other improved algorithms,the proposed method shortens the path length by 61%,reduces the search time by 49%,and decreases the number of turns by 71%,significantly improving planning efficiency and path safety.The algorithm's feasibility is further verified through practical robot applications.
作者
刘行俊
高焕兵
孙庚
冯子腾
LIU Xingjun;GAO Huanbing;SUN Geng;FENG Ziteng(School of Information and Electrical Engineering,Shandong Jianzhu University,Jinan 250101,China;Shandong Zhongjian Engineering Quality Inspection Co.,Ltd.,Jinan 250024,China)
出处
《自动化与仪表》
2025年第11期43-49,79,共8页
Automation & Instrumentation
基金
山东省科技型中小企业创新能力提升工程项目(2022TSGC1262)。
关键词
路径规划
改进蚁群算法
多目标优化
安全距离约束
动态权重调整
精英策略
path planning
improved ant colony algorithm
multi-objective optimization
safety distance constraint
dy-namic weight adjustment
elite strategy
