Article Details

Title · Authors · Abstract · Keywords · DOI · References

Robust Residual-Driven Adaptive Kalman Filtering

FENG Li

(School of Information, Renmin University of China, Beijing 100872, China)

Section：Communication Engineering · Vol. 2026, Issue 10 · Pages：15-23 · 2096-4706（2026）10-0015-09 · CLC number：TP273+.4 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2026.10.003

Abstract

Accurately estimating dynamic states when the prior statistical characteristics of process and measurement noise are unknown, time-varying, and contain outliers has always been a challenge. To address this, this paper proposes a robust Residual-Driven Variational Bayesian Adaptive Kalman Filter (RDVBAKF) algorithm. This algorithm algebraically expands the residual of the Kalman filter, explicitly incorporating both process and measurement noise into a single linear observation equation, effectively decoupling the complex interactions between the noises. The paper adopts the inverse Wishart distribution as the conjugate prior for the noise covariance matrices and introduces the Gamma distribution as a hierarchical representation for auxiliary precision variables, forming a student-t distribution-based hierarchical model. This modeling approach not only ensures the analytical solvability of the posterior distribution but also enhances the algorithm's robustness to outliers through the heavy-tailed characteristics of the student-t distribution, allowing it to maintain stable filtering performance in non-Gaussian noise environments. Simulation experiments and real experiments verify the superior performance of the proposed method.

Keywords

Kalman Filtering; Variational Bayesian; residual; non-Gaussian noise

References

  1. [1] LIANG S，XU G，QIN Z J. A Robust Kalman Filter for Heavy-Tailed Measurement Noise Based on Gamma Pearson VII Mixture Distribution [J].IEEE Access，2025，13：47680-47692.
  2. [2] JIA G，HUANG Y，BAI M B. A Novel Robust Kalman Filter With Non-Stationary Heavy-Tailed Measurement Noise [J].IEEE Transactions on Aerospace and Electronic Systems，2020，53（2）：368-373.
  3. [3] ZHU H，ZHANG G，LI Y，et al. An Adaptive Kalman Filter With Inaccurate Noise Covariances in the Presence of Outliers [J].IEEE Transactions on Automatic Control，2021，67（1）：374-381.
  4. [4] HUANG Y，ZHANG Y，WU Z，et al. A Novel Adaptive Kalman Filter With Inaccurate Process and Measurement Noise Covariance Matrices [J].IEEE Transactions on Automatic Control，2017，63（2）：594-601.
  5. [5] ZHAO J，WANG G D S P. Robust Cubature Kalman Filter Based on Generalized Hyperbolic Distribution for SLAM Under Colored Heavy-Tailed Measurement Noise [J/OL].Digital Signal Processing，2025：105390（2025-06-09）.https://doi.org/10.1016/j.dsp.2025.105390.
  6. [6] LI W，GUO L. Robust Particle Filtering With Time-Varying Model Uncertainty and Inaccurate Noise Covariance Matrix [J].IEEE Transactions on Systems，Man，and Cybernetics：Systems，2020，51（11）：7099-7108.
  7. [7] SÄRKKÄ S，NUMMENMAA A. Recursive noise adaptive Kalman filtering by variational Bayesian approximations [J].IEEE Transactions on Automatic Control，2009，54（3）：596-600.
  8. [8] Wang Guoqing, Yang Chunyu, Ma Lei, etc Nonlinear Kalman Filter Based on Gaussian Generalized Hyperbolic Mixture Distribution [J]. Journal of Automation, 2023, 49 (2): 448-460
  9. [9] GAO X H，MA Z Y，CHNENG Yet al. A Novel Residual-Based Bayesian Expectation–Maximization Adaptive Kalman Filter With Inaccurate and Time-Varying Noise Covariances [J/OL].Mechanical Systems and Signal Processing，2024，235：114937（2024-05-27）.https://doi.org/10.1016/j.measurement.2024.114937.
  10. [10] FU H，CHENG Y，HUANG W J M. A Novel Variational Robust Filter With Gaussian Mixture Model for Unknown Non-Gaussian Noises [J/OL].Mechanical Systems and Signal Processing，2023，221：113417（2023-08-17）.https://doi.org/10.1016/j.measurement.2023.113417.
  11. [11] HUANG W，FU H，LI Y，et al. A Gaussian-Multivariate Laplacian Mixture Distribution Based Robust Cubature Kalman Filter [J/OL].Measurement，2023，213：112732（2023-03-21）.https://doi.org/10.1016/j.measurement.2023.112732.
  12. [12] QIAO S，FAN Y，WANG G，et al. A Modified Federated Student's t-Based Variational Adaptive Kalman Filter for Multi-Sensor Information Fusion [J/OL].Measurement，2023，222：113577（2023-09-23）.https://doi.org/10.1016/j.measurement.2023.113577.
  13. [13] JIANG Z，ZHOU W，GUO Y，et al. A Novel Robust Multivariate Laplace Distribution-Based Distributed Consensus Information Fusion [J/OL].Measurement，2024，225：113986（2023-12-08）.https://doi.org/10.1016/j.measurement.2023.113986.
  14. [14] PAN C，GAO J，LI Z，et al. Multiple Fading Factors-Based Strong Tracking Variational Bayesian Adaptive Kalman Filter [J/OL].Measurement，2021，176：109139（2021-02-24）.https://doi.org/10.1016/j.measurement.2021.109139.
  15. [15] LIU X，XU H，ZHENG D，et al. A Novel Variational Bayesian Adaptive Kalman Filter With Mismatched Process Noise Covariance Matrix [J].IET Radar, Sonar & Navigation，2023，17（6）：967-977.
  16. [16] HUANG Y，ZHANG Y，LI N，et al. A Novel Robust Student's t-Based Kalman Filter [J].IEEE Transactions on Aerospace and Electronic Systems，2017，53（3）：1545-1554.
  17. [17] WANG G，YANG C，MA X. A Novel Robust Nonlinear Kalman Filter Based on Multivariate Laplace Distribution [J].IEEE Transactions on Circuits and Systems II: Express Briefs，2021，68（7）：2705-2709.
  18. [18] FU H，CHENG Y. A Computationally Efficient Robust Cubature Kalman Filter With Multivariate Laplace Distribution [J].IEEE Transactions on Instrumentation and Measurement，2023，72：1-11.
  19. [19] JÉGOU H，DOUZE M，SCHMID C. Product Quantization for Nearest Neighbor Search [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2011，33（1）：117-128.
  20. [20] HE K，ZHANG X，REN S，et al. Deep Residual Learning for Image Recognition [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas：IEEE，2016：770-778.
  21. [21] CHEDED L，DORAISWAMI R. Tracking the Trajectory of an Object in a Noisy Environment With Unknown Statistics: A Novel Robust Kalman Filter Residue-Based Approach [J].IEEE Transactions on Instrumentation and Measurement，2023，72：1-19.
  22. [22] ZHANG X，DUAN L，GONG Q，et al. State of Charge Estimation for Lithium-Ion Battery Based on Adaptive Extended Kalman Filter With Improved Residual Covariance Matrix Estimator [J/OL].Journal of Power Sources，2024，589：233758（2023-10-22）.https://doi.org/10.1016/j.jpowsour.2023.233758.
  23. [23] GONG M，ZHANG K，SCHOELKOPF B，et al. Discovering Temporal Causal Relations From Subsampled Data [C]//International Conference on Machine Learning. Lille：PMLR，2015：1898-1906.
  24. [24] IMPRAIMAKIS M，SMYTH A W J M S. A New Residual-Based Kalman Filter for Real Time Input–Parameter–State Estimation Using Limited Output Information [J/OL].Mechanical Systems and Signal Processing，2022，178：109284（2022-05-17）.https://doi.org/10.1016/j.ymssp.2022.109284.
  25. [25] FENG W，LI Q，LU Q. Force Localization and Reconstruction Based on a Novel Sparse Kalman Filter [J/OL].Mechanical Systems and Signal Processing，2020，144：106890（2020-04-24）.https://doi.org/10.1016/j.ymssp.2020.106890.
  26. [26] GILLIJNS S，DE MOOR B J A. Unbiased Minimum-Variance Input and State Estimation for Linear Discrete-Time Systems [J].Automatica，2007，43（1）：111-116.
  27. [27] ZHANG Y，SONG X，LIU Z. Maximum Correntropy Unbiased Minimum-Variance Filter [J/OL].Signal Processing，2024，219：109405（2024-01-26）.https://doi.org/10.1016/j.sigpro.2024.109405.
  28. [28] ZHU Z，ZHU S，WANG Y W，et al. Structural Dynamic Response Reconstruction With Multi-Type Sensors，Unknown Input，and Rank Deficient Feedthrough Matrix [J/OL].Mechanical Systems and Signal Processing，2023，187：109935（2022-12-01）.https://doi.org/10.1016/j.ymssp.2022.109935.
  29. [29] XU B，HUANG X C，ZHANG J H，et al. A Fault Detection Method for a Practical Electro-Hydraulic Variable-Displacement Pump With Unknown Swashplate Moment [J].IEEE Transactions
  30. on Instrumentation and Measurement，2023，72：1-11.
  31. [30] LI Y，TIAN W，XIA Y，et al. Principal Force Pattern and Impulse Response Mode for Structural Equivalent Force Estimation and Full-Field Response Reconstruction [J/OL].Mechanical Systems and Signal Processing，2023，200：110660（2023-08-06）.https://doi.org/10.1016/j.ymssp.2023.110660.
  32. [31] LIU Y，WANG Z，ZOU L，et al. Unbiased State and Fault Estimation for Discrete-Time Complex Networks With Time Delays [J].International Journal of Control，2023，97（8）：1819-1829.
  33. [32] ALMAGBILE A，WANG J，DING W. Evaluating the Performances of Adaptive Kalman Filter Methods in GPS/INS Integration [J].Journal of Global Positioning Systems，2010，9（1）：33-40.
  34. [33] HARTIKAINEN S. Variational Bayesian Adaptation of Noise Covariances in Non-Linear Kalman Filtering [J/OL].arXiv:1302.0681 [stat.ME].（2013-02-04）.https://arxiv.org/abs/1302.0681.
  35. [34] STUART A，ORD J K，ARNOLD S. Kendall's Advanced Theory of Statistics. Vol. 2A: Classical Inference and the Linear Model [M].6th ed. London：Hodder Arnold，1999.
  36. [35] HUANG Y，ZHANG Y，ZHAO Y，et al. A Novel Robust Gaussian–Student's t Mixture Distribution Based Kalman Filter [J].IEEE Transactions on Signal Processing，2019，67（13）：3606-3620.
  37. [36] BAI M，HUANG Y，CHEN B，et al. A Novel Robust Kalman Filtering Framework Based on Normal-Skew Mixture Distribution [J].IEEE Transactions on Systems, Man, and Cybernetics: Systems，2021，52（11）：6789-6805.
  38. [37] BAI M，HUANG Y，ZHANG Y，et al. Statistical Similarity Measure-Based Adaptive Outlier-Robust State Estimator With Applications [J].IEEE Transactions on Automatic Control，2022，67（8）：4354-4361.
  39. [38] HUANG Y，ZHANG Y，ZHAO Y，et al. A Novel Outlier-Robust Kalman Filtering Framework Based on Statistical Similarity Measure [J].IEEE Transactions on Automatic Control，2020，66（6）：2677-2692.
  40. [39] MAJDIK A L，TILL C，SCARAMUZZA D. The Zurich urban micro aerial vehicle dataset [J].International Journal of Robotics Research，2017，36（3）：269-273.

About the authors

Feng Li (1998-), male, Han ethnicity, from Heze, Shandong, is currently pursuing a master's degree in computer application technology.