Comparison of travel times estimation using intelligent infrastructure and floating car data

The aim of this paper is to compare two methods of measurement of travel times along road sections in the city. Comparison is made by using a traffic simulator that records travel times of simulated vehicles using two algorithms. Those algorithms generalize two approaches in measuring travelling time: by tracking selected vehicles and by identifying vehicles at the ends of road sections. The former method is a well-known Floating Car Data (FCD) technique, the latter refers to usage of sensors installed in road infrastructure like RFID readers or video cameras which are used in license plate recognition systems. It is shown that even small fraction of FCD-enabled vehicles gives reasonable estimations, but in case the technology used for vehicle identification at crossroads guarantees correctness greater than 50% then the second method gives much realiable results.