ACKF filtering algorithm based on exponential weighting
Abstract
In order to optimally track the tracking and exploration targets, it is based on the performance of the traditional volume Kalman filter (CKF) [1]–[2] in practical applications and the characteristics of uncertain and unknown time-varying system noise statistical characteristics. This paper proposes an ACKF algorithm that combines maximum a posteriori estimation and exponential weighting with CKF filtering. After using Sage-Husa's suboptimal unbiased maximum posterior estimator to estimate the unknown time-varying or inaccurate noise statistics in real time [9] , and combining it with the CKF algorithm to obtain the ACKF recursive formula, and then use the progressive. The combination of memory index weighting and limited memory index weighting continuously corrects the noise statistics, and strengthens the robustness of the filter by reducing the state estimation error.
Related Papers
- → A Comparison of Complementary and Kalman Filtering(1975)293 cited
- → New Internet search volume-based weighting method for integrating various environmental impacts(2015)24 cited
- → A study of weighting factors of the quadratic performance index(1969)5 cited
- → MMSE optimal detector via kalman filter(2005)
- → Optimal estimation of distances between holes precision by Kalman filter theory(2002)