ANFIS-Based Seamless Train Positioning Method

GNSS/INS integrated navigation system is the development trend of railway satellite positioning. However, there are many dense forests, mountains and tunnels in the railway environment, which will affect the satellite signals during the operation of the train, so that the positioning error of the INS rapidly diverges, and eventually the entire system fails. In order to solve this problem, this paper combines ANFIS with strong self-learning ability with GNSS/INS, and proposes a seamless train positioning method based on self-learning (GNSS/ANFIS/INS). The above method is applied to the Shuozhou-Huanghua railway, and the analysis of the test results shows that the method can effectively reduce the positioning error and velocity error in the case of GNSS failure. Compared with only INS, the RMSE of position and velocity are reduced by approximately 85% and 75%, respectively.