Research On The Information Fusion In GPS/SINS Integrated Navigation System

Using advanced integrated navigation data fusion algorithm, combine with data fusion of strapdown inertial navigation system, GPS and other navigation system outputs, can effectively improve the accuracy and reliability of the system. Aiming in Color noise problems at the integrated navigation system, a modified Kalman filtering algorithm was derived in this paper. This paper introduced color noise whitening to Kalman filter, gave the Kalman filter algorithm based on color noise whitening of concrete steps, and verified the correctness and rationality with the dynamic vehicle test. The main contents of the paper and innovations are as follows:1. Research on strapdown inertial navigation system works. Introduced the conversion relationship between navigation and coordinate systems; analysis of inertial navigation system works and specific force equation and calculated equation solving process of inertial navigation systems are described. Analyze the type of satellite navigation system error and causes.2. Research of Kalman filters works. Introduces the basic Kalman filter equations established a strapdown inertial navigation systems and satellite navigation system error model. Describes the strengths and combinations of integrated navigation system, applies Kalman filtering algorithm in integrated navigation system combined with the error models.4. Improved Kalman filter algorithm. The whitening colored noise is introduced into the Kalman filter, a new Kalman filter algorithm is designed based on the colored noise to solve problem that colored noise will be approximately seen as white noise in the traditional Kalman filter, which leads a larger positioning errors.5. PNS100 navigation system was introduced, and according to the data of the system protocol, the corresponding software on a PC is designed to receive navigation data sent from the integrated navigation system. Navigation data collected through dynamic test verifies the rationality of Kalman filtering algorithm based on color noise.