Multi-Sensor Fusion Framework Based on GPS State Detection

In this paper, a GPS Fusion Framework Based on Graph optimization is proposed. Integrating visual inertial navigation odometer, a robust and high-precision UAV positioning algorithm for indoor and outdoor complex environments is completed. By using the proposed method, UAV platform can accurately obtain its own pose in indoor and outdoor, tunnel and other different environments. Compared with the pure visual positioning algorithm, the proposed multi-sensor fusion algorithm based on GPS state detection improves the positioning accuracy significantly. Finally, the algorithm is verified on the KITTI dataset and the quadrotor UAV platform. It is proven that the proposed algorithm has high precision and strong robustness.