Vehicle Positioning Method in GNSS-Denied Environment: INS/ODO/Mag Integrated Navigation System Algorithm

This paper presents a positioning method of INS/ODO/MAG integrated navigation system with GNSS rejection based on Extend Kalman filter (EKF). Firstly, the coordinate system used is explained and the correction matrix of odometer and magnetometer is established. Then, the output of magnetometer is converted into yaw angle as the correction of integrated navigation. Finally, the navigation and positioning effects is compared before and after correction. In addition, a dynamic on-board experiment was carried out. The data was collected by an unmanned ground vehicle (MUGV) loaded with integrated navigation equipment, and the magnetometer output data was obtained through semi-physical simulation. After data processing by the algorithm, before correction, the root-mean-square error (RMSE) of the INS/ODO integrated navigation system was 1.3125. After magnetometer correction, the RMSE of INS/ODO/MAG integrated navigation system was 0.1177, and the navigation accuracy was improved by 91.0363%. The results show that the positioning accuracy has been effectively improved and this method can provide a more accurate positioning method for the field of vehicle navigation and autonomous driving.