| Integrated navigation system refers to the two complementary systems on the vehicle,namely,satellite positioning system and strapdown inertial navigation system.By combining the fusion algorithm,the real-time output parameters(position,attitude and speed)with high reliability and high navigation accuracy overcome the fatal defects of a single navigation system.The technology has been applied to military and civil areas,and plays an important role in modern intelligent transportation system.Through the analysis and compensation of the system error of the third generation inertial sensor MEMS(Micro Electro-Mechanical System),this paper proposes a new algorithm design scheme for integrated navigation,and carries out the Simulink simulation verification.The results show that the positioning accuracy of the integrated navigation system has been significantly improved.The main work and research results are summarized as follows:(1)Error modeling and performance analysis of MEMS inertial sensor.In this paper,the measurement error of MEMS inertial sensor is analyzed and modeled.Because the precision of pure ins depends on the attitude precision to a great extent,and the gyro bias has a profound impact on the attitude precision,the MEMS sensor data processing is carried out in the experiment,and the bias estimation is carried out when the ins initializes the static base,so as to eliminate the bias;at the same time,when there is no GPS signal and the MEMS gyroscope has drift,the nonlinear dual vector complementary filtering attitude determination is adopted State fusion algorithm.In this paper,Simulink simulation is used to verify that the horizontal attitude angle accuracy obtained by this scheme is about 2 ° higher than that obtained by pure inertial navigation system,that is,the output of high-precision horizontal attitude angle is maintained within 0.4 °.This shows that the complementary filter attitude fusion algorithm can use the accelerometer measurement information to correct the bias caused by gyro bias in real time,and can significantly improve the horizontal attitude accuracy,and then improve the speed and position resolution accuracy.(2)GPS/SINS Integrated Navigation System based on 21 dimensional state estimation is established.In the local geographic coordinate system,the SINS algorithm based on attitude quaternion theory is established.Secondly,two kinds of GPS positioning methods are introduced,and the solution of related model navigation is obtained by iterative least square method.Finally,the error sources of GPS are analyzed in detail.Based on the algorithms and positioning error analysis of the above two navigation systems,this paper proposes a new scheme to extend the traditional 15 dimensional GPS / SINS integrated Kalman filter model to 21 dimensional GPS integrated navigation system,and the extended six dimensions are GPS three-axis velocity and position error.(3)The comparative simulation experiments of 15 and 21 dimensional state estimation of loosely integrated navigation system are carried out.Simulink simulation shows that when the GPS system can not obtain accurate measurements due to jump,the 21 dimensional Kalman filter algorithm can achieve higher navigation accuracy.The speed accuracy in three directions is improved by 0.04m/s,the latitude and longitude accuracy are improved by 0.2m and 0.02 M respectively,and the horizontal attitude angle accuracy is improved by 0.01. |
PDF Full Text Request