An artificial neural network approach for spatially extending road traffic monitoring measures

Abstract

This paper focuses on road traffic monitoring and proposes a method based on artificial neural networks for extending data collected on some road links to others. The method may be used to reduce the costs of monitoring equipment since it can estimate the data to be monitored on road segments where there is no equipment installed. The approach is tested on a small network, assuming different neural network frameworks. The numerical results show that the approach is promising, being able in most cases to estimate traffic flows with acceptable errors.