Forecasting and control of interurban traffic networks using a state-space formulated traffic model

摘要

This paper first presents a statistical `data assimilation' approach to handling the large volume of real time measurements now available from instrumented traffic networks. The dynamic state-space modelling technique employed, herein called the Statistical Traffic Model (STM), transforms on-line roadside measurements of traffic flow into explicit assessments of the current and future state of an inter-urban road network, and provides transport management with a tool for monitoring, prediction and control. The paper then goes on to discuss the application of adaptive Proportional-Integral-Plus (PIP) control systems to a non-linear STM simulation of the Amsterdam ring road.