摘要
短时、实时交通流预测是城市交通控制与诱导的基础,也是智能交通系统的重要功能之一。研究表明,没有一种预测模型和方法能够适用于各种环境和条件下的交通流预测。本文提出并建立了一种短时交通流的智能预测系统,按照智能预测和组合预测的思想,实现预测模型的智能选择,以及多个预测模型的预测结果的综合。仿真试验结果表明,系统具有较强的适应性,提高了预测的准确性。该系统采用多智能体技术进行设计和实现,文章给出了这种基于多智能体的智能交通预测支持系统的系统结构和工作原理。
The real time forecast for short-term traffic flow is the foundation of urban traffic control and guidance, which is also one of main functions of intelligent transportation system. It's proved by studies that none of forecasting models can dominate the others in all situations. A intelligent forecasting system for short-term traffic flow is proposed and constituted in this paper, in which the intelligent model choice and combination of forecasting results given by different models are implemented based on the thought of intelligent and combining forecast. It's shown by simulations that the system has good adaptability to the change of environment and can improve the forecast veracity. Applying the multi-agent technique, this forecast system is constructed and the architecture and principle of such an intelligent forecast supported system based on multi-agent is also mentioned in this paper.
出处
《预测》
CSSCI
2004年第2期28-34,共7页
Forecasting
基金
国家自然科学基金资助项目(70001003)
天津市自然科学基金资助项目(023607311)
