Research On Real-time GNSS/INS Integrated Navigation System Based On Digital Signal Processor

With the development of today's society,the accuracy and reliability of the navigation system are required higher and higher.The single navigation system has been unable to meet the needs of practical applications.At present,the combination of the inertial navigation system and the satellite navigation system is the main component in the various kinds of integrated navigation systems,and the Kalman filtering method can effectively integrate the information of the two system to realize the complementary characteristics of the integrated navigation system.In this thesis,the GNSS/INS integrated navigation system is implemented and the algorithm is optimized for the real-time system.The real-time GNSS/INS integrated navigation system of the embedded platform and the client of the Windows platform are developed.And the validation of the system developed in this thesis was conducted in an open and urban environment.The main research contents are as follows:1?The basic knowledge of GNSS/INS integrated navigation is introduced in detail,including the common coordinate system and mutual conversion relationship,the mechanization of strapdown inertial navigation,the Kalman filter composition of GNSS/INS integrated navigation system and the additional constraints in the vehicle mode.2?This thesis studies the concrete realization method of real-time integrated navigation system,and discusses the real-time improvement of the system.The problems of the GNSS data delay and too long time consuming for Kalman updating exist in the real-time GNSS/INS integrated navigation system.In this thesis,an improved mathematical model which can transfer error state to the current time in one step after finishing the delayed observation update is proposed based on the Kalman filter theory.The improved model can reduce the effect of GNSS data delay and time-consuming observation update on the real time navigation output.So the model allows to finish the observation update in more sufficient time.It has the important significance for the integrated navigation algorithm's implementation in a low computing power processor and the application with highly real time requirement.Then,the design scheme of the integrated navigation hardware platform is described in detail,and the implementation method of the real-time GNSS/INS integrated navigation program in the embedded platform and the client program in the Windows platform are introduced.3?In order to verify the real-time GNSS/INS integrated navigation system in this thesis,many vehicle tests has been carried out in Wuhan,and in these tests we unplug the GNSS antenna to simulate the GNSS satellite signal interrupt.The navigation error statistics of these tests show that the standard deviation of postion and velocity is 0.05m,0.005m/s respectively,the standard deviation of roll and pitch is 0.004deg,and the standard deviation of heading is 0.03deg.The above results achieved the expected navigation accuracy.The drift level during the interruption of the GNSS satellite signal is in accordance with the IMU level and all the additional constraints in the vehicle mode work well.The time-consuming statistics in these tests show that,compared with the standard Kalman model,the improved model proposed in this thesis can reduce the time delay to about 1ms and is not affected by the GNSS observation data and the Kalman update calculation delay under the premise of ensuring the accuracy of the navigation system.So the improved model can make sure of the system's real time performance.