摘要
针对传统蚁群优化算法的不足,结合区间概率和人类记忆的特征,提出一种具备记忆特征的区间蚁群算法。首先针对每条路径的信息素浓度都为固定值的局限性,将信息素推广到区间范围内,使蚂蚁路径的选择方式为区间概率,增大寻优过程中路径选择的多样性,其次对信息素的更新方式根据人类记忆特征,选择一定范围内的次优路径,分别利用长时记忆和短时记忆方法更新,提高信息素更新的多元化。对所提出的算法进行了收敛性分析,最后通过大量的仿真分析,并与其他相关算法对比分析,充分验证了算法良好的性能。
Aiming at the drawback of the traditional ACO,an interval probability ant colony algorithm inspired by the characteristic of human’s memory is proposed.The pheromone of the path in the ACO can be extend to the interval bound,and then the probability for the selection of the path can be set as the interval probability,which can expand the diversity of ACO in the path selection.The updating of pheromone can be performing according to the way of human's memory,and different path can be updated according to the long-term memory updating and short-term memory way within the sub-optimal path.The proposed algorithm can reach the satisfactory solution set through the convergence analysis.Lots of simulation results for path planning problem show that the proposed algorithm performs well than other algorithms.
作者
刘振
王亚蛟
LIU Zhen;WANG Yajiao(College of Coastal Defense Force,Naval Aviation University,Yantai 264001;No.92706 Troops of PLA,Ningbo 315813)
出处
《舰船电子工程》
2019年第9期27-31,44,共6页
Ship Electronic Engineering
基金
国家自然科学基金项目(编号:51605487,61174031)资助
关键词
蚁群优化算法
区间概率
人工记忆
收敛
ant colony optimization
interval probability
artificial memory
convergence
