The Research Of Calibration And Compensation Methods For High Precision Strapdown Inertial Navigation System

The inertial navigation system(INS) which is inspired by inertial components, can produce vibration error, constant error and time accumulated error. If calibration technology is adopted in the system, it can take advantage of external information sources, it also can calculate and compensate the error which affect the accuracy of INS, and improve the performance of INS fundamentally. But, as a result of applying traditional calibration technology, it will be affected by the random gyro-drift, carrier movement and accuracy of external information. So develop error analysis of calibration scheme researching, and analyze error which affects the accuracy of calibration scheme, at last the key technology of calibration scheme is proposed.The calibration error analysis shows that the external reference sensor accuracy and reliability of the navigation inertial navigation system is a prerequisite and guarantee for the calibration. As Most fault detection method can’t detect the slowly varying soft fault sensitively, it also has great time delay and can’t determine the fault type. Proposed a navigation sensors fault detection method which based on the maximum principle of wavelet. The sensor observations are used to tell whether the sensor is fault, This paper shows principle of the fault detection based on wavelet singularity and the process of algorithm. The algorithm can not only have high sensitivity with soft errors of navigation sensor, but also determine the fault type by the the Lipschitz exponent of fault point, it can provide effective information for the fault sensors isolation and repairing. This algorithm can guarantee the reliability of external navigation information when INS carry out calibration scheme. Computer simulation shows that the proposed method is effective, and with high availability.In order to reduce effects caused by the gyro random gyro-drift, develop the design of real-time gyro filtering algorithm which based on wavelet technology. Propose the idea of considering the wave cycle during selecting SureShrink, which can optimize SureShrink threshold. Through real-time verification and the experimental results, This algorithm can meet conditions of real-time, and filter random gyro-drift noise effectively.Take the INS system used in underwater vehicle as research object, develop the muti-sensor calibration algorithm researching. Build combination of SINS/DVL/GPS calibration scheme based on the federated filter. Through Analysis of the principle of navigation sensors and deriving error equation, and complete the design of local filter The simulation results shows that the calibration scheme based on combination federal filter has well estimation capabilities, and improve positioning capability of INS.