Research And Application Of GPS/INS Integrated Navigation And Nonlinear Filtering In Navigation System

There are several problems have been reasrched in order to improve the accuracy of innerial navigation system, including the research and implementation of the integrated navigation system using GPS and strapdown inertial navigation system(SINS),investigation on random drift model for MEMS gyroscope based on particle filtering and the research on initial alignment based on large misalignment angles.The main efforts done in this paper are listed as follows:1. The noise reduction method for micro-electro-mechanical systems (MEMS) gyroscopic random signal is discussed from the point of view on practical applications. Firstly, The random drift signal of a gyroscopic is preprocessed, and the time-varying ARMA model is built under the condition of unstable random process; Based on all the above-mentioned methods, particle filtering and the improved particle filtering to cope with nonlinear model are adopted to reduce noise; Besides, Allan variance analyzing method is applied to deal with the filtering results, at the same time to identify each error coefficient. The experiment results indicate that, this modeling and filtering method can effectively reduce the random drift error, depress the noise of a gyroscopic, what's more, estimate and compensate for the error. It is in some degree that the navigation precision is improved and it has been applied to, tested by practice.2. In order to highly improve the initial alignment accuracy of SINS, SINS nonlinear error model is deduced in detailed based on large misalignment angels using UKF filtering, the model is a nonlinear model. It widely used in a situation that both state and measurement processes noises are additive but complex noises, On basis of that, UKF algorithm is adopted. In order to reduce the computation and measurement model being linear, simplified UKF algorithm is further implemented. SINS initial alignment simulation tests are executed, and the results shows that initial alignment can be completed with the proposed SINS error model and the simplified UKF algorithm effectively and quickly, and the accuracy of the system is highly improved.3. The statement and measurement equations are built for the GPS/INS integrated navigation system using the technology of kalman filtering, at the same time, the data fusion has been achieved. On basis of that, adopting the above-mentioned research results, the soft system of the strapdown inertial navigation system is designed and implemented.