Formulation and Validation of Model for Calculating the Number of Probe Cars Considering Traffic Information Update Frequency

Abstract

Probe cars are very effective tools in traffic information gathering. Probe cars can collect the traffic condition data without using roadside sensors but with only using their own equipped sensors. By using probe cars, it is expected that the accuracy of traffic information will be improved and the area coverage of the traffic information service will be increased, and many experimental probe car data collection projects have been implemented. However, the number of probe cars required for adequate information update frequency has not been theoretically discussed. This paper formulates the probabilistic relationships among the traffic volume of probe cars, the area coverage, and the update frequency of traffic information provision. This formulation calculates the adequate number of probe cars to realize traffic information services in advance considering two major service factors: area coverage and update frequency. This paper also validates the theoretical formulation with experimental probe car data collected in a taxi monitoring system.