计算机应用 ›› 2005, Vol. 25 ›› Issue (04): 742-744.DOI: 10.3724/SP.J.1087.2005.0742

• 人工智能与仿真 • 上一篇    下一篇

基于遗传算法的动态网络中最短路径问题算法

邹亮,徐建闽   

  1. 华南理工大学交通学院
  • 出版日期:2005-04-01 发布日期:2005-04-01
  • 基金资助:

    广东省自然科学基金资助项目(E5320271)

Method of shortest paths problem on dynamic network based on genetic algorithm

ZOU Liang,XU Jian-min   

  1. College of Traffic and Communications,South China University of Technology
  • Online:2005-04-01 Published:2005-04-01

摘要:

提出了一种以随机Dijkstra最短路径算法为基础,运用遗传算法来求解动态路径诱导系统 中最短路径问题(ShortestPathproblemonDynamicRouteGuidanceSystem,SPDRGS)的算法。通过运用 该随机Dijkstra算法解决了将遗传算法应用与最短路径问题中初始种群的产生问题。考虑到目前动态 路径诱导系统(DynamicRouteGuidanceSystem,DRGS)对路径诱导算法的时间复杂度和网络约束条件 的要求,此算法不仅能够较快地求出较优的路径而且对网络没有任何的约束条件,同时对离散和连续的 动态网络模型有效,因此符合DRGS的要求。

关键词: 随机Dijkstra算法, 动态路径诱导系统, 最短路径, 遗传算法

Abstract:

An algorithm based on random Dijkstra algorithm and applying genetic algorithm to solve SPDRGS(Shortest Path problem on Dynamic Route Guidance System) was proposed. By applying random Dijkstra algorithm, the algorithm cleared out the biggest obstruction between the genetic algorithm and SPDRGS, which is how to get the initial generation of GA(Genetic algorithm). According to DRGS’s (Dynamic Route Guidance System) demand for time complexity and network constraint condition of route guidance algorithms, this algorithm can quickly find the excellent path and does not need any network constraint condition, which also can solve the problems on continuously and discrete dynamic networks. So the algorithm proposed can satisfy the demand of DRGS.

Key words: random Dijkstra algorithm, DRGS, shortest path, genetic algorithm

中图分类号: