【摘要】 针对经典蚁群算法在复杂环境下的机器人路径规划问题中表现出的收敛速度慢,容易陷入局部最优等问题,提出一种改进算法.依据方向指导信息来优化初始信息素的分布,加快搜索速度,缩减搜索初期的时间消耗;通过优化信息素的挥发与更新规则,保留局部与全局优秀路径的优势信息,改善收敛速度慢的问题;基于区域安全因素对转移概率进行改进,从而避免陷入局部最优和死锁等问题.最后,通过栅格法对仿真环境建模,在不同复杂度与规模的多张地图上进行仿真实验,对比验证了该算法在复杂环境下路径规划问题上的有效性和对不同规模地图的适应性.更多还原

【Abstract】 For the problem such as slowconvergence and being easy to fall into local optimum of classical ant colony algorithm,an improved algorithm was proposed in this paper for robot path planning in complex environment.We optimize the distribution of the initial pheromone depending on the direction guidance information to accelerate the speed of search and reduce the time cost in initial search.For the optimizing of volatilization and updating rules of pheromone,we preserve the superiority of local excellent path and global excellent path to improve the slowconvergence.We ameliorate the computational method of transition probability based on the area safety factor to avoid the problems of falling into local optimum and deadlock.Finally,the grid method is used to establish simulation environment.Through simulation experiments on different complexity and scale maps,the algorithm is compared with classical algorithm.The experimentresults demonstratethe proposed algorithm shows more effectively and stronger in environmental adaptation in the complex environment.更多还原