Review of Data Collection Methods for Microscopic Traffic Simulation

Abstract

Research into developing more reliable microscopic traffic simulation models is seriously hampered by the availability of suitable microscopic data for calibration and validation purposes. This paper presents a review of several data collection technologies that the authors have been experimenting with. The methods include (i) video recording (ii) the use of differential GPS (DGPS) and (iii) an instrumented vehicle fitted with the S.M.S. Drive Recorder system. Examples of data collection with each method are briefly described, and some of the results presented to illustrate the capabilities of the methods. The last section of the paper summarises the advantages and limitations of each method and conclusions are drawn on their suitability for the required purpose. The experiments shows that each method can provide some useful results for microscopic simulation model development, but cannot provide continuous spacing-, speed- and acceleration profiles of large vehicle platoons moving through various traffic scenarios. Further research is required to develop more sophisticated data collection techniques and to investigate such aspects of driver behaviour as aggressiveness, and factors influencing lane selection.