Train integrated positioning system based on GPS and inertial measurement unit

GPS receiver and inertial measurement unit are employed to form the train integrated positioning system, which includes the sensor input layer, fault detection and isolation layer, data fusion layer and the output layer. In this paper, the positioning calculation method of GPS and the inertial measurement unit are analyzed, the H∞ robust filter based method is designed for the data fusion, a wavelet based method is proposed to realize fault detection, and then the corresponding fault isolation and system reconfiguration strategies are determined. Validation with real train positioning data from field test and some simulation demonstrate that the train integrated positioning system could meet the requirement of precision and reliability in complex train operating environment and interference condition, earn higher stability than traditional Kalman filter, guarantee the system safety and continuity with effective fault detection and isolation, and keep better adaptive ability and practical value.