Method for Accelerometer and Magnetometer Based Vehicle Classification

With the rapid development of smart cities and intelligent transportation systems (ITS), there is an increasing demand for efficient and accurate vehicle classification methods. Existing classification systems often rely on limited sensor inputs or complex infrastructure, which pose drawbacks such as high costs, limited scalability, and reduced accuracy. In this paper, we propose a novel method for vehicle classification using data from accelerometer and magnetometer sensors installed in the pavement. Calculation of vehicle speed and length, axle count, and axle spacing using sensor fusion is presented. Different methods for vehicle classification into defined classes are assessed.