改进蚁群算法在城市汽车导航中的应用

发布时间：2018-03-28 10:25

本文选题：汽车导航　切入点：蚁群算法　出处：《控制工程》2016年01期

【摘要】：针对城市汽车导航中的车辆路径规划问题,借助返回思想,提出有返回的改进蚁群算法,通过返回策略解决了搜索中的"死胡同"问题。细致研究了汽车导航中城市路网的基本特征,受几何学中"两点之间线段最短"的启发,提出动态有限区域搜索策略,减小了搜索范围,提高了搜索效率。鉴于A*算法搜索时间短的优势,将其与有返回的改进蚁群算法相结合,提出基于动态区域规划的分层蚁群算法(DHACO),利用A*算法和有返回的蚁群算法进行两次路径优化,提高了搜索效率和可行解的质量。在宣城市市区的网络交通图上对改进算法进行实例验证,与A*算法和有返回的改进蚁群算法相比,DHACO算法在更短的时间内搜索到了更短的路径,实验结果验证了其在工程实践中的可行性和有效性。
[Abstract]:In order to solve the problem of vehicle path planning in urban vehicle navigation, an improved ant colony algorithm with return is proposed with the help of the idea of return. The "dead end" problem in search is solved by return strategy. The basic characteristics of urban road network in automobile navigation are studied in detail. Inspired by the "shortest line segment between two points" in geometry, a dynamic finite region search strategy is proposed. It reduces the search range and improves the search efficiency. In view of the advantage of short search time of the A * algorithm, it is combined with the improved ant colony algorithm with return. A hierarchical ant colony algorithm (DHACOG) based on dynamic area planning is proposed. The algorithm A * and the returned ant colony algorithm are used to optimize the path twice. The search efficiency and the quality of feasible solution are improved. The improved algorithm is verified on the network traffic diagram of Xuancheng city. Compared with the A* algorithm and the returned improved ant colony algorithm, the DHACO algorithm can find a shorter path in a shorter time. The experimental results show the feasibility and effectiveness of the method in engineering practice.
【作者单位】：国网辽宁省电力有限公司;东北大学信息科学与工程学院;
【基金】：国家自然科学基金项目(60674021)
【分类号】：TP18

【相似文献】