改进粒子群算法在水上飞机与多目标船避碰决策中的应用  

Application of Improved Particle Swarm Optimization Algorithm in Collision Avoidance Decision-making of Seaplane and Multi-target Ship

作　　者：翁建军[1,2] 胡冲 李龙浩[2] 严庆新[1,2] 王群朋[3] 郑帅祥 WENG Jianjun;HU Chong;LI Longhao;YAN Qingxin;WANG Qunpeng;ZHENG Shuaixiang(School of Navigation, Wuhan University of Technology, Wuhan 430063, China;Hubei Inland Shipping Technology Key Laboratory, Wuhan 430063, China;School of Navigation, Guangzhou Maritime University, Guangzhou 510725, China)

机构地区：[1]武汉理工大学航运学院,武汉430063 [2]武汉理工大学航运技术湖北重点实验室,武汉430063 [3]广州航海学院海运学院,广州510725

出　　处：《武汉理工大学学报（交通科学与工程版）》2021年第4期799-804,共6页Journal of Wuhan University of Technology（Transportation Science & Engineering）

基　　金：国家自然科学基金国际合作重点项目(51920105014);国家自然科学基金青年项目(51809206)。

摘　　要：为优化水上飞机与多船共存的港口水域交通环境,降低碰撞风险,提高通航效率,优化避碰决策,从水上飞机与船舶操纵特点出发,构建水上飞机与目标船会遇态势的碰撞危险判断模型,基于其计算结果,通过构建危险度隶属函数评价水上飞机与船舶碰撞危险程度,并将粒子群算法应用于水上飞机转向时机和转向幅度的研究中.在综合考虑避碰的多种因素,建立适应度模型,引入免疫算法中的免疫算子避免陷入局部极值陷阱,添加扰动因子解决后期搜索能力不足的问题,从而得到改进后的免疫粒子群算法避碰决策.仿真结果表明:改进后的粒子群算法可为水上飞机与多船会遇局面的避碰决策研究提供一种可行的求解途径.In order to optimize the traffic environment in port waters where seaplanes coexist with multiple ships,reduce the collision risk,improve navigation efficiency and optimize collision avoidance decision-making,a collision risk judgment model for the encounter situation between seaplanes and target ships was established based on the operating characteristics of seaplanes and ships.Based on the calculation results,the risk degree of collision between seaplane and ship was evaluated by constructing the risk degree membership function,and the particle swarm optimization algorithm was applied to the research of seaplane steering timing and steering amplitude.Considering various factors of collision avoidance,the fitness model was established.The immune operator in immune algorithm was introduced to avoid falling into the trap of local extremum,and disturbance factor was added to solve the problem of insufficient searching ability in the later period,so that the improved immune particle swarm optimization algorithm can avoid collision.The simulation results show that the improved particle swarm optimization algorithm can provide a feasible solution for the decision-making research of collision avoidance when seaplanes and multiple ships encounter.

关 键 词：水上飞机 多船避碰 粒子群算法 免疫算法 避碰决策 

分 类 号：U698.5[交通运输工程—港口、海岸及近海工程]