Automatic Vehicle Classification System for Monitoring Highways

Automatic vehicle classification (AVC) systems provide information about classes of vehicle that can be used for many purposes. This paper describes a axle count and axle spacing using wireless accelerometers sensors and magnetometers sensors. The vehicle axles are detected by accelerometer sensors, and the vehicles arrivals, departures and speed are estimated by magnetometers. The prototype system is installed on Interstate 80 at Pinole, CA, USA, and tested under different traffic conditions. Commercial weighin-motion station provides video images and reports to analyze the performance of a system, including vehicle counts,axle spacing and classification. The results show that the prototype AVC system is reliable in classifying vehicles even under congested traffic with accuracy of 99%.