摘要
单一的蚁群算法在路径搜索前期具有盲目性,导致收敛速度较慢,容易陷入局部最优;而传统的人工势场法在凹形障碍物时避障困难。将人工势场法与蚁群算法结合,提出一种由势场力引导的蚁群算法。算法将人工势场法中的力因素转化为一种蚁群算法里的信息素权值,能够在开始时加快蚁群算法的收敛速度,并随着迭代次数增加逐渐降低力因素对蚂蚁的影响,在陷入局部最优时通过回退降低了蚁群算法陷入局部最优的可能性。又增加了路径曲线距障碍物距离的设置,使其更贴近现实应用。最后使用该算法与其它类似势场蚁群算法进行了仿真比较,在条件一致的情况下,文中算法的收敛速度更快,路径距离更短。
The single ant colony algorithm has blindness in the early stage of path search,which leads to slower convergence and easily fall into local optimum.However,the traditional artificial potential field method is difficult to avoid obstacles in concave obstacles.Combining artificial potential field method with ant colony algorithm,an ant colony algorithm guided by potential field force is proposed.The algorithm converts the force factor in the artificial potential field method into a pheromone weight in an ant colony algorithm,which can accelerate the convergence speed of the ant colony algorithm at the beginning,and gradually reduce the influence of the force factor on the ant as the number of iterations increases.The fall back into local optimism reduces the possibility of the ant colony algorithm falling into local optima.It also increases the distance between the path curve and obstacles,making it more practical.Finally,the proposed algorithm is compared with other similar potential field ant colony algorithm.Under the same conditions,the proposed algorithm has the faster convergence speed and the shorter path length.
作者
郭彤颖
刘雍
李宁宁
富楚涵
GUO Tong-ying;LIU Yong;LI Ning-ning;FU Chu-han(School of Information and Control Engineering,Shenyang Jianzhu University,Shenyang 110168,China)
出处
《组合机床与自动化加工技术》
北大核心
2020年第6期18-20,26,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
中国国家自然科学基金面上基金项目(61873174)
辽宁省自然科学基金项目(2019-ZD-0681)
中国建设教育协会教育教学科研项目(2019161)。
关键词
移动机器人
人工势场法
蚁群算法
路径规划
mobile robot
artificial potential field method
ant colony algorithm
path planning
