Research on Application of Two Kinds of Robust Kalman Filtering in Tunnel Subsidence Monitoring

Standard Kalman filtering model requires the statistical properties of system noise to meet the following conditions such as the zero mean as well as white noise,but in the practical engineering application,due to the complex enviroment,it is difficult to meet the requirements. Based on the background of actual railway tunnel subsidence monitoring,a comparative study on application of two different robust Kalman filtering algorithms based on adaptive factor and based on the robust estimation was made. The results show that,both robust Kalman filtering algorithms are able to eliminate the gross errors,but the robust Kalman filtering based on adaptive factor has a good practical effect.