Approaches Of Position Increment Accuracy Assessment And Error State Transform Filtering Based On The Full Source GNSS/MIMU Tightly Coupled Navigation

In this thesis,the small UAV integrated navigation is taken as the research background,the full source GNSS/MIMU tightly coupled navigation is taken as the research object,aiming to meet the higher accuracy and stability of the full source GNSS/MIMU tightly coupled navigation,the research work is carried out around the error state transformation Kalman filter and the accuracy evaluation method of position increment.First of all,in order to meet the needs of higher accuracy and stability of the UAV integrated navigation,this paper studies two kinds of the full source GNSS/MIMU tightly coupled navigation systems based on traditional Kalman filter and error state transform Kalman filter,and completes the related design and modeling.This design includs the navigation state and error state,the differential and integral forms of the state and error state quantities.This modeling includes the continuous and discrete forms of the error state system equations,the observation equations of pseudo range single difference between satellites and Doppler single difference between satellites and carrier phase double difference between satellites and time.The information processing flow of the full source GNSS/MIMU tightly coupled navigation system.The simulation analysis result surface that the full source GNSS/MIMU tightly coupled navigation system based on the error state transformation filter is reliable.The on-board test analysis results surface that after convergence,the 3D position accuracy of the error state transform Kalman filter is stable within 6m,the three-dimensional speed accuracy is stable within 0.4m/s,the three-dimensional attitude accuracy is stable within 2 degrees.The on-board test analysis results also surface the accuracy and stability of the error state transform Kalman filter are better than the traditional Kalman filter.Secondly,according to the characteristics of MEMS inertial measurement unit and the shortcomings of previous papers,the system equation based on error state transformation Kalman filter is derived again in this paper,then a new EKF based on the full source GNSS/MIMU tightly coupled navigation and error state transform is put forward.The new Kalman filter can not only reduce the calculation error but also be used in the low-cost MEMS inertial devices.Finally,through the simulation and on-board test,it is verified that the new Kalman filter has advantages under the application background of tight integrated navigation using MEMS inertial measurement unit and full source satellite information of pseudo range,Doppler and carrier phases.Then,in view of the high precision position increment requirement of the cycle slip detection assisted by inertial navigation,this paper discusses the most effective method of calculating the position increment in real-time navigation for the first time,derives the general theoretical model for calculating the variance of the interval position increment,and puts forward an accuracy assessment method based on the system-prediction variance.By assessing the position increment of two filter models in different time intervals and different motion states with on-board observation data,it is proved that the accuracy assessment method of the position increment is useful.With the completion of BD-3 global satellites navigation system and the increasing demand of small UAV formation operation,the full source GNSS/MIMU tightly coupled navigation system based on error state transform filtering,and the position incremental accuracy evaluation method serving for the inertial navigation assisted cycle slip detection,will play an increasingly important role in the field of GNSS/MIMU coupled navigation.The research results of this paper can not only improve the position,speed and attitude accuracy of the full source GNSS/MIMU tightly coupled navigation system,but also improve the stability of the navigation system when the number of satellites is small,the motion changes between static and dynamic.At the same time,this paper can also provide theoretical support for the position increment’s accuracy assessment serviced for the cycle slip detection assisted by inertial navigation,provide theoretical support for the UAV GNSS/MIMU tightly coupled navigation system in the development of higher accuracy and integrity.