A Multi-Source Fusion Navigation System to Overcome GPS Interruption of Unmanned Ground Vehicles

Abstract

The Inertial Navigation System/Global Positioning System (INS/GPS) fusion navigation technology commonly used by unmanned ground vehicles is prone to GPS interruption in complex environments, resulting in the inability of the system to locate accurately. This paper designs an INS/GPS/Odometer/Vision/Magnetometer fusion navigation scheme in response to this problem. Add odometer, vision, and magnetometer to the INS/GPS integrated navigation system, use odometer dead reckoning to suppress the position offset caused by GPS interruption, reduce the cumulative error of the system through visual positioning, and correct the course of INS with the help of magnetometer orientation Angle, improve the positioning ability of the system in the GPS interruption environment. In order to further enhance the positioning accuracy of the system, the Federal Filter algorithm (FKF) was improved, and the Federal Robust Cubature Kalman Filter algorithm (FRCKF) was proposed. A Robust Cubature Kalman Filter algorithm is introduced into the framework of the federated filter algorithm, which improves the processing ability of the FKF algorithm for nonlinear systems and reduces the positioning error of the system. Finally, an experimental platform for unmanned ground vehicle navigation was developed, and the effectiveness of the multi-source fusion navigation system was verified through vehicle navigation experiments. The experiment results show that the multi-source fusion navigation system designed in this paper can effectively reduce the negative impact of GPS interruption on the INS/GPS integrated navigation system and improve the positioning accuracy of unmanned ground vehicles.

Keywords

• Unmanned ground vehicles
• GPS interruption
• INS/GPS/odometer/vision/magnetometer fusion navigation system
• federal robust cubature Kalman filter algorithm
• nonlinear systems