| Unscented Kalman filter(UKF),as a stable and unbiased estimation method for nonlinear systems,is widely used in integrated navigation.However,in the case of bad noise conditions,it is difficult to establish the noise model accurately,which leads to serious performance degradation of UKF and cannot meet the requirements of integrated navigation.This thesis studies the innovation adaptive unscented Kalman filter method.For the integrated navigation system of global navigation satellite system(GNSS)and inertial navigation system(INS),it can enhance the adaptability to environmental uncertainty.The main work of this thesis is as follows:In view of the problem of the performance degradation of UKF in the noisy environment,a new innovative adaptive estimation unscented Kalman filter(IAE-UKF)is proposed.The Chi-square detection method is used to detect the operation of the integrated navigation system,judge whether the noise model is accurate,and adjust the noise model adaptively according to the detection results and filtering innovation.The traditional unscented Kalman filter,adaptive unscented Kalman filter based on correlation inference and innovation adaptive unscented Kalman filter are simulated and compared.The results show that the innovation adaptive unscented Kalman filter can improve the accuracy and stability of the system in the case of bad noise or inaccurate noise model.Aiming at the information fusion problem of integrated navigation system under the bad noise condition,based on the innovation adaptive unscented Kalman filter method,the GNSS / INS integrated navigation system is constructed,and the simulation experiment is carried out.The results show that this method can obviously alleviate the error divergence of the integrated navigation system with low-cost ins in the harsh noise environment,and improve the ability of the integrated navigation system with highperformance INS to resist the change of noise environment. |
PDF Full Text Request