基于人工鱼群算法的出租车调度优化研究

发布时间：2018-01-18 11:37

本文关键词：基于人工鱼群算法的出租车调度优化研究　出处：《西安科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：出租车作为城市公共交通工具的补充，为人们的交通出行提供了诸多方便，在城市交通运输中起着越来越重要的作用。然而由于出租车行业的粗放式管理模式以及落后的路边招手打车方式，致使出租车司机和乘客所掌握的信息不对称，进而导致了市民“打的难”而出租车空驶率却居高不下的问题，同时加重了交通拥堵，环境污染等诸多问题，人们的生活质量受到了严重的影响。而且目前所试行的电召及电话人工调度的出租车调度方式，叫车方式单一，调度方式效率低，成本高。因此，研究和实行新的高科技、高效率的出租车叫车系统显得尤为必要和迫切。本文充分研究了国内外出租车调度技术的发展和应用情况，针对出租车行业的管理特点和技术需求，提出了以北斗定位技术（BD）、第三代移动通信技术（3G）、Google MapsAPI技术为基础的出租车调度管理系统设计方案。本系统主要包含车载终端、通信平台、呼叫中心、短信平台、调度平台和管理平台六大部分。系统基于B/S模式设计，，采用Windows Server2003为操作系统、SQL Server2005为数据库，使用Microsoft VisualStudio2008集成开发环境，开发语言为ASP.NET+C#，并以Google Maps API作为开发地图模块的API。本文的工作重点是出租车调度系统中最短路径求解算法的研究与改进。主要针对基本人工鱼群算法因参数视野固定不变而导致算法后期收敛速度慢、运算量大、易陷入局部最优的缺陷，根据静态最短路径问题的特点，对人工鱼群算法进行了改进。该改进算法只对人工鱼的觅食行为的视野进行调整，使其随着迭代次数的变化而自适应地变化，并设置了视野值的下限，以防视野过小，算法又陷入局部最小。实验结果表明，改进型人工鱼群算法的收敛速度、计算量、寻优精度和准确性均优于基本人工鱼群算法及基本蚁群算法，而且道路越复杂，节点越多，这种优势越显著。
[Abstract]:As a supplement of urban public transportation, taxi provides a lot of convenience for people to travel. It plays a more and more important role in urban transportation. However, because of the extensive management mode of taxi industry and the backward wayside taxi hailing mode, the information of taxi drivers and passengers is not symmetrical. This has led to the problem of "hard to hit" and the high rate of empty taxi driving, which has aggravated many problems, such as traffic congestion, environmental pollution and so on. The quality of life of people has been seriously affected. And at present, the taxi dispatching mode of call and telephone manual dispatching is single, the efficiency of dispatching is low, and the cost is high. Research and implementation of new high-tech, high-efficiency taxi-hailing system is particularly necessary and urgent. This paper fully studies the development and application of taxi dispatching technology at home and abroad, aiming at the management characteristics and technical requirements of the taxi industry, put forward the Beidou positioning technology. The design scheme of taxi dispatching management system based on the third generation mobile communication technology: 3G / MapsAPI technology. This system mainly includes vehicle terminal, communication platform and call center. Short message platform, scheduling platform and management platform six parts. The system is based on B / S mode design, using Windows Server2003 as the operating system. SQL Server2005 is the database, Microsoft VisualStudio2008 is used to integrate the development environment, and the development language is ASP.NET C #. And using Google Maps API as the development map module API. This paper focuses on the research and improvement of the shortest path algorithm in taxi scheduling system, mainly aiming at the basic artificial fish swarm algorithm, which leads to the slow convergence rate of the algorithm because of the fixed parameter field of vision. The artificial fish swarm algorithm is improved according to the characteristics of the static shortest path problem. The improved algorithm only adjusts the field of vision of the artificial fish foraging behavior. It changes adaptively with the number of iterations and sets the lower bound of the visual field to prevent the field of vision from becoming too small and the algorithm fall into the local minimum. The experimental results show that the improved artificial fish swarm algorithm converges at the rate of convergence. The computational complexity, the accuracy and accuracy of the optimization are better than the basic artificial fish colony algorithm and the basic ant colony algorithm, and the more complex the road, the more nodes, this advantage is more obvious.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U492.22;TP18

【参考文献】