多约束条件下气象航线规划的智能混合算法  

Intelligent Hybrid Algorithms for Meteorological Route Planning under Multiple Constraints

作　　者：文婷 WEN Ting(Beijing Highlander Digital Technology Co.,Ltd.,Beijing 100095,China)

机构地区：[1]北京海兰信数据科技股份有限公司,北京100095

出　　处：《船舶设计通讯》2019年第2期69-74,共6页Journal of Ship Design

基　　金：智能船舶1.0研发专项——船舶辅助自动驾驶系统开发。

摘　　要：针对船舶航线规划面临气象条件、水文地理、航行需求、船舶物理特性等复杂条件约束的情况,为设计同时满足船舶航行安全性和经济性需求的最佳气象航线,提出一种结合传统A^*算法和遗传算法的智能混合算法。该算法使用A^*算法来提升遗传算法初始种群质量并加快搜索速度,根据航行需求和多约束条件设计目标函数和适应度函数,采用多种群技术和精英保留策略增加种群多样性并加快算法收敛速度。遗传算法在航线搜索的一个方向上是连续的,这与其他采用离散网格系统的算法相比拥有更高的搜索精度。实验结果表明该智能混合算法能高速、有效地完成船舶气象航线规划设计。In view of the complex constraints of meteorological conditions,hydrogeography,navigation requirements and physical characteristics of ships in the ship route planning,an intelligent hybrid algorithm combining traditional A^* algorithm and genetic algorithm was proposed to design the best meteorological route that could meet the requirements of both ship navigation safety and economy.The A^* algorithm was used to improve the initial population quality and to speed up the search speed of genetic algorithm.Objective function and fitness function were designed according to the navigation demand and multiple constraints.The multi-population technology and the elite retention strategy were used to increase the population diversity and to speed up the convergence speed of the algorithm.The genetic algorithm was continuous in one direction of the route search,and had higher search accuracy compared with other algorithms using the discrete grid system.The experimental results showed that the hybrid algorithm could accomplish the ship meteorological route planning and design efficiently and quickly.

关 键 词：最佳气象航线 遗传算法 智能混合 多约束 连续搜索 

分 类 号：U692.31[交通运输工程—港口、海岸及近海工程]