Measurement-based adaptive Kalman filtering algorithm for GPS/INS integrated navigation system

To solve the problem that traditional adaptive filtering algorithms are inaccurate and unreliable in estimating the GPS measurement noises of GPS/INS integrated navigation system, a measurement-based adaptive Kalman filtering algorithm(MAKF) is put forward. Based on the complementary measuring characteristics of GPS/INS and the short-term high-precision features of INS, this algorithm can get the adaptive estimation of GPS measurement noises. The simulation results show that the MAKF, which can real-time track the GPS time-varying measurement noises and modify the observation error covariance matrix R, is superior to the improved sage-husa kalman filtering in filtering precision and stability. And it can effectively overcome the divergence of improved sage-husa adaptive filtering algorithm for the integrated navigation system, especially when with lower-precision gyro.