Data processing for rail level dynamic inspection based on an adaptive Kalman filter

The inspection of the geometrical parameters of rail tracks is an important aspect in the daily maintenance and safe running of railways. The rail level (superelevation) is one of the important indicators susceptible to measurement noise. In this paper, the principle of the Kalman filter is studied, an adaptive Kalman filter algorithm is designed for level (superelevation) dynamic inspection, the selection principle for the filtering parameters is discussed and the performance of the algorithm is verified through simulation tests and pushing experiments using a rail inspection trolley. From analysis of the measurement data, it is concluded that the trolley speed is an important factor affecting level (superelevation) inspection and an improved algorithm including the trolley speed is proposed to further improve the filtering ability. The algorithm is easy to implement and can be extended to dynamic rail inspection.