Noise covariance estimation for Kalman filter tuning using Bayesian approach and Monte Carlo

2012 ◽  
Vol 27 (11) ◽  
pp. 957-973 ◽  
Author(s):  
Peter Matisko ◽  
Vladimír Havlena
Keyword(s):  
Monte Carlo ◽  
Kalman Filter ◽  
Bayesian Approach ◽  
Covariance Estimation ◽  
Filter Tuning ◽  
Noise Covariance

Kalman Filter With Recursive Covariance Estimation—Sequentially Estimating Process Noise Covariance

2014 ◽  
Vol 61 (11) ◽  
pp. 6253-6263 ◽  
Author(s):  
Bo Feng ◽  
Mengyin Fu ◽  
Hongbin Ma ◽  
Yuanqing Xia ◽  
Bo Wang
Keyword(s):  
Kalman Filter ◽  
Covariance Estimation ◽  
Process Noise ◽  
Noise Covariance

scholarly journals Maximum Likelihood-Based Measurement Noise Covariance Estimation Using Sequential Quadratic Programming for Cubature Kalman Filter Applied in INS/BDS Integration

2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Bingbing Gao ◽  
Gaoge Hu ◽  
Wenmin Li ◽  
Yan Zhao ◽  
Yongmin Zhong
Keyword(s):  
Kalman Filter ◽  
Maximum Likelihood ◽  
Sequential Quadratic Programming ◽  
Covariance Estimation ◽  
Measurement Noise ◽  
Likelihood Principle ◽  
Maximum Likelihood Principle ◽  
Measurement Noise Covariance ◽  
Cubature Kalman Filter ◽  
Noise Covariance

With the completion of the Beidou-3 system (BDS) in China, INS/BDS integration will become a promising navigation and positioning strategy. However, due to the nonlinear propagation characteristic of INS error and inevitable involvement of inaccurate measurement noise statistics, it is difficult to achieve the optimal solution through the INS/BDS integration. This paper proposes a method of cubature Kalman filter (CKF) with the measurement noise covariance estimation by using the maximum likelihood principle to solve the abovementioned problem. It establishes an estimation model for measurement noise covariance according to the maximum likelihood principle, and then, its estimation is calculated by utilizing the sequential quadratic programming. The estimated measurement noise covariance will be fed back to the procedure of CKF to improve its adaptability. Simulation and comparison analysis verify that the proposed method can accurately estimate measurement noise covariance to effectively restrain its influence on navigation solution, leading to improved navigation performance for the INS/BDS integration.

Sign in / Sign up

Export Citation Format

Share Document