摘要
针对传统蚁群算法(ACO)在塔式起重机路径规划中普遍存在的早熟、迭代速度慢、容易陷入局部最优和生成路径不平滑等问题,提出了一种改进的蚁群算法。根据塔机的工作环境,搭建出带有障碍物的二维栅格地图用以模拟不同的施工环境,将A*算法的评价函数融合到蚁群算法中,用以改进蚁群算法的启发函数,并在此基础上改进了信息素更新机制以及引入了新的路径平滑机制。通过Matlab将ACO、蚁群算法和本文改进的蚁群算法进行了对照仿真实验。仿真结果表明:无论是在简单环境还是复杂环境下,本文改进的算法在收敛性、最终生成路径的拐点、拐角个数以及长度等多方面的表现都要更加优异。
Traditional ant colony algorithm commonly faces challenges such as premature convergence,slow iteration speed,susceptibility to local optima,and generation of non-smooth paths in tower crane path planning.In response to these issues,an enhanced ant colony algorithm has been proposed.Taking into account the working environment of the tower crane,a 2D grid map with obstacles is constructed to simulate various construction scenarios.The evaluation function of A*algorithm is integrated into ant colony algorithm,thereby refining its heuristic function.Additionally,improvements have been made to the mechanism for updating pheromones,and a new path smoothing mechanism has been introduced.Through Matlab,comparative simulation experiments were conducted involving traditional ant colony algorithm,and enhanced ant colony algorithm proposed in this paper.The simulation results indicate that,whether in simple or complex environments,the improved algorithm in this study excels in terms of convergence,t he number of turning points and turning angles in the final generated path,and the overall path length.
作者
卢宁
金正南
董守峰
胡信凯
LU Ning;JIN Zhengnan;DONG Shoufeng;HU Xinkai(School of Mechanical-Electronic and Vehicle Engineering,Beijing University of Civil Engineering and Architecture,Beijing 102616,China;Beijing Engineering Research Center of Monitoring for Construction Safety,Beijing 100032,China)
出处
《中国工程机械学报》
北大核心
2025年第5期790-795,共6页
Chinese Journal of Construction Machinery
基金
北京建筑大学研究生创新项目资助(PG2024137)。
