计算机与现代化

• 人工智能 • 上一篇    下一篇

基于适应度的驾驶员路径选择模型及仿真

  

  1. 1.北方工业大学自动化系,北京 100144; 2.交通运输部公路科学研究院,北京 100088)
  • 收稿日期:2013-09-27 出版日期:2014-02-14 发布日期:2014-02-14
  • 作者简介:张金金(1987-),女,河南焦作人,北方工业大学自动化系硕士研究生,研究方向:交通流预测,智能交通控制; 苏婧(1976-),女,北京人,交通运输部公路科学研究院高级工程师,研究方向:交通流模型及仿真,高速公路机电系统; 刘小明(1974-),男,教授,博士,研究方向:智能交通控制,交通流理论。
  • 基金资助:
    国家科技支撑计划基金资助项目(2011BAH16B01-05); 北京市教委科技计划基金资助项目 (KM201210009001)

Model and Simulation for Driver Route Choice Based on Fitness Function

  1. (1. Department of Automation, North China University of Technology, Beijing 100144, China;
    2. Research Institute of Highway, Ministry of Transport, Beijing 100088, China)
  • Received:2013-09-27 Online:2014-02-14 Published:2014-02-14

摘要: 针对车辆的路径选择问题,首先描述路径选择单元的划分规则,然后从影响车辆路径选择的路径历史经验和路径最近经验入手,分别定义车辆对路径的初始适应度、历史适应度、最近适应度及路径选择行为概率,并对车辆记忆属性及其变化流程进行说明,最终给出车辆路径选择规则。在上述分析的基础上进行多次重复时间步仿真,仿真结果表明,本文给出的路径选择规则虽然相对较简单,但能够较好地模拟车辆路径选择行为,为交通仿真中车辆路径选择模型提供了一种新的思路。

关键词: 路径选择, 初始适应度, 历史适应度, 最近适应度, 行为概率

Abstract: In order to resolve the route choice problem in traffic simulation, with the route choice history experience and route choice recent experience of driver, the initial fitness, history fitness, recent fitness and route choice action probability are defined, then the memory property of driver and changing process are described, at last, the route choice rule is brought up. On the basis of above analysis time step simulation are carried out repeatedly, the simulation result proves that the rule is simple and effective, but provides a new way for the driver route choice model in traffic simulation.

Key words: route choice, initial fitness, history fitness, recent fitness, action probability

中图分类号: