Study On Key Technologies Of GNSS/INS Deep Coupled Based On Vector Tracking

With the increasing application of navigation technology in the military and civilian fields,users are increasingly demanding the accuracy and reliability of navigation systems.The Global Navigation Satellite System(GNSS)and the Inertial Navigation System(INS)are the two most common navigation systems.GNSS provides users with high-precision position,speed and time information in real time,with long-term positioning accuracy.However,the data update rate is low,the positioning signal is susceptible to interference,and the dynamic performance is poor.The INS can provide position,speed and attitude information to the user completely autonomously and continuously,with high precision and short-term accuracy.However,its positioning error will accumulate over time.The characteristics of GNSS and INS have strong complementary characteristics,and the organic integration of the two can effectively improve the performance of the navigation system.The GNSS/INS integrated navigation system has received more and more attention and research.The integration of GNSS and INS is usually divided into three modes according to the architecture: loose integration,tight integration and deep integration.Among them,the loose integration structure is the simplest,and the positioning parameters output by the two subsystems working independently are fused and filtered.The tight integration system uses the parameters such as the pseudorange and pseudorange rate of the GNSS output and the pseudorange and pseudorange rate information predicted according to the INS navigation parameters to perform fusion in the distance domain.Both the loose integration and the tight integration use the information of the GNSS receiver to suppress the error accumulation of the INS,and the system fails when the GNSS fails to work normally.Deep integration adopts the idea of vector tracking,and combines GNSS signal tracking and INS error divergence suppression into a combined algorithm.It penetrates into the GNSS receiver and completes the signal tracking layer fusion,which effectively improves the overall anti-interference and high dynamic performance of the system.Deep integrated navigation is one of the development directions of the future navigation system.To this end,this paper studies the key technologies of GNSS/INS deep integrated navigation system based on vector tracking.The main work includes:Firstly,the GNSS loop tracking technology and navigation solution technology based on Extended Kalman Filter(EKF)are studied.The EKF-based GNSS loop tracking technology is the basis of the vector tracking algorithm pre-processing filter.The EKF-based navigation solution technology is the basis of the vector tracking algorithm navigation filter.The separate implementation of the two can be applied to the traditional GNSS navigation receiver algorithm alone,replacing the original phase-locked loop-based signal tracking technology and single-point positioning algorithm,thus effectively improving the tracking and positioning performance of the traditional GNSS receiver..The Kalman filter system model and measurement equation are established.The algorithm is verified by C language software receiver and digital intermediate frequency signal acquisition platform using sports car data and satellite navigation simulator data.Second,the vector tracking technique based on the federated filter structure is studied.The federal filter structure was studied.The difference between vector tracking and traditional scalar tracking method is analyzed.The vector tracking structure is studied.The vector tracking system model and measurement equation are established.The vector tracking algorithm is verified by satellite navigation simulator signal and sports car data.Thirdly,the structure and model establishment of GNSS/INS deep integrated navigation system are studied.Firstly,the GNSS/INS loosely coupled and tightly coupled integrated navigation structure is researched and implemented,which lays a foundation for deep integration research.The structure and model of deep integrated navigation system are studied in detail,and the performance advantages of deep combined navigation structure are analyzed.The performance of the deep combined navigation structure was verified by using sports car experiments and simulation data.The C-based non-coherent cascaded deep integration navigation software receiver based on C language is designed and developed to lay the foundation for the subsequent engineering implementation.