Model-based Sensing of Track Components for Location of Rail Vehicles

The characteristic arrangement of track components - such as rail clamps, switches, and the like - enables a train-borne location determination in railbound traffic. Sensors mounted on the vehicle measure signals of the rail track. These sensor signals are then compared with reference signals stored in a digital rail map by using correlation techniques. A model-based approach is presented to generate these reference signals. Moreover, the correlation methodology is extended by considering nonlinear distortions of the recorded signals, leading thus to a significantly higher robustness of the method. The performance of the proposed approach is demonstrated based on the detection of rail clamps as well as the identification of track sections and switches