An innovative algorithm for train detection

Train detection is a very important research issue affecting vehicles and line safety. Currently, the European Train Control System ETCS (a signalling, control and train protection system) Level 1 and 2 provide the train localization functionalities by using track circuits and/or axle counter systems: the problem of these solutions is represented by the high cost of track circuit and axle counter installation and of the related equipment management. This paper presents an innovative train detection algorithm, able to perform the train localization, by estimating its speed, crossing time instants and axle number. The aim of the proposed solution is to use the same processing approach to evaluate all these quantities, starting from the knowledge of the vertical loads on the sleepers directly measured on the track. The inputs are processed through cross-correlation operations to extract the required information in terms of speed, crossing time instants and axle counter. A suitable model of railway vehicle and track has been also developed to test the algorithm when experimental data are not available. The railway vehicle chosen as benchmark is the Manchester Wagon, implemented in the Adams VI-Rail environment. The physical model of the flexible track has been implemented in the Matlab and Comsol Multiphysics environments. A simulation campaign has been performed in order to verify the performance of the proposed algorithm, under different operative conditions. The research has been carried out in cooperation with Ansaldo STS and ECM.