Application of adaptive Kalman filtering algorithm in IMU/GPS integrated navigation system

Abstract

The IMU(inertial measurement unit) error equations in the earth fixed coordinates are introduced firstly. A fading Kalman filtering is simply introduced and its shortcomings are analyzed, then an adaptive filtering is applied in IMU/GPS integrated navigation system, in which the adaptive factor is replaced by the fading factor. A practical example is given. The results prove that the adaptive filter combined with the fading factor is valid and reliable when applied in IMU/GPS integrated navigation system.

Static alignment of inertial navigation systems using an adaptive multiple fading factors Kalman filter

“…By considering the obtained results of the above equation and according to Equation ( ) 15 , the following conclusion is achieved:…”

“…A fast fading occurs when the system model is not accurate and slow fading is used for accurate models. Few years later, an optimal fading factors Kalman filtering algorithm had been offered, in which exponential weight changing approach was used to balance the model errors and unknown drifts [14], [15], [16], [17]. [18], shows the use of adaptive filtering techniques to develop the A c c e p t e d M a n u s c r i p t 5 speed of the dynamic alignment of a micro-electro-mechanical system inertial measurement unit (MEMS IMU) with real-time kinematic global positioning system (RTK GPS) for a nautical function.…”

Windowing-based random weighting fitting of systematic model errors for dynamic vehicle navigation

“…However, this method requires that residual vectors or predicted residual vectors at each time point be in the identical type, spatial dimension and distribution, which is difficult to achieve in a highly dynamic environment [18,30]. The adaptive fading filter incorporates the suboptimal fading factor as a multiplier in the filtering process to improve the influence of predicted residual information for the enhanced filtering performance [11,25,31,39]. Although this filter has a simple structure, it achieves the filtering convergence with the trade-off of the filtering precision [6].…”

Analysis on the kalman filter performance in GPS/INS integration at different noise levels, sampling periods and curvatures

“…In studies like [2,3] by Zheng and Chow and [4] by Ojha, Klingenberg and Chow, it is important to understand the relationship between sampling rate, road curvature and the system performance in order to build an efficient bandwidth allocation scheme for intelligent transportation management. Several studies have been done on improving the performance of GPS signals and integrating the GPS data with the INS [5][6][7][8]. Wei and Schwarz use two Kalman filters separately to filter the GPS signals and then use the filtered GPS signal as updates to the estimates obtained from the inertial data in [5].…”

“…al. [7] and Hide [8] use adaptive Kalman filtering in order to compensate for the unpredictable amount of measurement noise in GPS. While these studies focus on introducing different types of Kalman filters as a tool for better estimation, some other studies have looked into the issues that come with using these filters.…”