Fused attitude estimation method based on Mahony filtering and Error-State Kalman filtering
In order to solve the problem of the low accuracy of Inertial Measurement Unit(IMU)and the large error of traditional attitude estimation algorithm,an attitude estimation method based on IMU combining Mahony filter and Error State Kalman Filter(ESKF)was proposed.In order to reduce the error caused by non-gravity acceleration and magnetic interference,two adaptive PI controllers are used to improve the Mahony algorithm,and the IMU measurement data are solved by the improved Mahony algorithm as the ESKF measurement value.ESKF defines the true state as the combination of the nominal state and the error state,and indirectly completes the estimation of the system state.Because for small amount of error,error state linearization error is smaller,the calcula-tion of Jacobi matrix is simpler.Experimental results show that compared with the traditional attitude calculation algorithm,the fusion algorithm can effectively reduce the error caused by high-frequency noise and data drift,and improve the accuracy of attitude calcula-tion.
万方数据
维普
