基于公交车GPS数据的短时交通流预测研究

Research on Short-term Traffic Flow Forecasting Based on GPS Data of Bus

随着智慧化城市的提出,智能交通系统已经成为城市建设中至关重要的部分,而短时交通流预测是实现智能交通系统的核心研究内容之一[1]。本文对获取的公交车GPS数据进行了挖掘分析,提取公交车速度数据进行短时交通流预测算法研究。考虑到时序数据的时间相关性和交通流数据的准周期特性,本文设计长短期记忆人工神经网络(Long-Short Term Memory,LSTM)对交通流速度数据进行预测。结果表明,LSTM能够通过对历史速度数据的学习,找出时间序列之间的关系,利用LSTM的选择性记忆功能,能够对短时交通流速度进行更准确的预测。

With the introduction of the smart city,Intelligent transportation systems have become a crucial part of urban construction,and traffic flow prediction is a key issue in intelligent transportation system study.In this paper,we conducted data mining and analysis of bus GPS data,and take the bus speed data for short-term traffic flow prediction algorithm.Considering the temporal correlation of time series data and the quasi-periodic characteristics of traffic flow data,long Short-Term Memory neural network is designed to predict traffic flow velocity in this paper.The results show that LSTM neural network model can learn the past data very well,and identify the relationship between time series.Taking advantage of selective memory,LSTM can get a reliable prediction of Short-term traffic flow.