公交到站时间预测研究现状与发展趋势

实时的公交到站时间预测是智能公交的重要组成部分,准确的预测有利于帮助居民进行出行规划和减少等待时间.通过研究公交到站时间预测的原理和方法,系统总结了基于GPS、APC等数据的统计学方法和分析模型.对历史平均法、神经网络、卡尔曼滤波、支持向量机和基于概率的预测模型等几种典型方法的预测原理进行了介绍,从预测精度、实时性、计算复杂性等几个方面对模型的优缺点进行了比较.分析了公交到站时间预测发展趋势,提出了该领域需要进一步研究的问题.

Current Study and Development Trend of Bus Arrival Time Prediction

Bus arrival time prediction is one of the most important parts in intelligent public transportation system.Accurate prediction could help residents to make travel plan and reduce waiting time at bus stop.This review summarizes the basic theories and methods of bus arrival time prediction based on GPS and APC data in recent years.First,this paper discusses the main prediction models,such as historic average model,neural network model,Kalman filter model and support vector machine model.Then,the paper compares the advantages and disadvantages of these four models related to the precision,dynamics,model complexity,etc.Finally,the paper analyzes the development of bus arrival time prediction in the future.Some new issues related to the prediction models are put forward at the end of this paper.