Adaptive Multisensor Fusion Using the First Two Moments of the Innovation Under Nonstationary Degradation
DOI:
https://doi.org/10.22517/23447214.26509Keywords:
adaptive Kalman filter, closed-loop control, innovation, multisensor fusion, sensor degradation, state estimationAbstract
Adaptive Kalman filters modify the confidence assigned to measurements when sensor quality changes. Many methods monitor innovation energy, although a slowly varying bias can shift its mean without producing large instantaneous innovations. This work proposes a channel-wise adaptation that combines bounded covariance inflation, a test on the mean of the whitened innovation, and a minimum-dwell-time gate. A differential-drive ground vehicle with GNSS, odometry, gyroscope, and magnetometer was simulated. Six conditions, five severity levels, and forty paired repetitions per cell were studied. A nominal EKF, a mistuned EKF, covariance matching, Sage-Husa, robust Huber weighting, and the proposed method were compared. Position RMSE predicted tracking RMSE with a mean Spearman correlation of 0.994, but the association fell to 0.491 for recovery time. Second-moment methods responded to variable noise and outliers but did not correct drifting bias. The first-moment test enabled action against this shift, although exclusion exhibited a bimodal distribution and improved 28 of 40 repetitions. Minimum dwell time reduced nominal switching. At maximum severity, test-triggered compensation improved upon the nominal EKF in all 40 repetitions, with a median improvement of 67.0% and a nominal cost of 32.2%. No action simultaneously dominated nominal performance, median bias performance, and dispersion. Adaptation should distinguish variance and mean changes and should be evaluated using average, transient, and nominal-cost metrics.
Downloads
References
Z. Chen, H. Biggie, N. Ahmed, S. Julier and C. Heckman, "Kalman Filter Auto-Tuning With Consistent and Robust Bayesian Optimization," IEEE Transactions on Aerospace and Electronic Systems, vol. 60, no. 2, pp. 2236–2250, 2024., doi:10.1109/TAES.2024.3350587
L. Zeng, Z. Fan, G. He, N. Li, H. Chang, X. Yu and G. Yuan, "JC-AKF: Joint calibration and NIS-guided adaptive Kalman filtering for a low-cost MEMS IMU array," Measurement, early access, Art. no. 122475, 2026., doi:10.1016/j.measurement.2026.122475
M. A. J. Rabby, R. Kannan and K. Bingi, "Adaptive NIS-Guided Extended Kalman Filter for Robust State-of-Charge Estimation of Lithium-Ion Batteries," Journal of The Electrochemical Society, vol. 173, no. 12, Art. no. 120507, 2026., doi:10.1149/1945-7111/ae73f7
B. Hang, W. Li, X. Li, R. Fan, H. Sun and F. Shen, "QA-MSCKF: A Statistically Adaptive Measurement Update Method for Visual–Inertial Odometry," IEEE Internet of Things Journal, vol. 13, no. 17, pp. 38934–38956, 2026., doi:10.1109/JIOT.2026.3704077
W. Hilal, N. Alsadi, S. A. Gadsden and M. A. AlShabi, "An adaptive SIF and KF estimation strategy for fault detection based on the NIS metric," in Proc. Sensors and Systems for Space Applications XVI, p. 26, 2023., doi:10.1117/12.2664054
X. Lai, G. Zhu and J. Chambers, "A fuzzy adaptive extended Kalman filter exploiting the Student’s t distribution for mobile robot tracking," Measurement Science and Technology, vol. 32, no. 10, Art. no. 105017, 2021., doi:10.1088/1361-6501/ac0ca9
Y. Yu, B. Liu, G. Xie, A. Han, X. Huang and H. Deng, "Adaptive Kalman Filter-Based Multi-Sensor Fusion for Robust Localization in GPS-Degraded Environments," in Proc. 2025 10th International Conference on Robotics and Automation Engineering, pp. 144–149, 2025., doi:10.1109/ICRAE67496.2025.00030
W. Huang, L. Xiang, R. Chen, S. Xu and Q. Wang, "Adaptive Multi-Sensor Fusion Localization with Eigenvalue-Based Degradation Detection for Mobile Robots," Sensors, vol. 26, no. 5, Art. no. 1653, 2026., doi:10.3390/s26051653
Y. Geng and J. Wang, "Adaptive estimation of multiple fading factors in Kalman filter for navigation applications," GPS Solutions, vol. 12, no. 4, pp. 273–279, 2008., doi:10.1007/s10291-007-0084-6
F. Deng, H. Yang and L. Wang, "Adaptive Unscented Kalman Filter Based Estimation and Filtering for Dynamic Positioning with Model Uncertainties," International Journal of Control, Automation and Systems, vol. 17, no. 3, pp. 667–678, 2019., doi:10.1007/s12555-018-9503-4
Z. Chen, C. Heckman, S. Julier and N. Ahmed, "Weak in the NEES?: Auto-Tuning Kalman Filters with Bayesian Optimization," in Proc. 2018 21st International Conference on Information Fusion, pp. 1072–1079, 2018., doi:10.23919/ICIF.2018.8454982
L. Cai, B. Boyacioglu, S. E. Webster, L. van Uffelen and K. Morgansen, "Towards Auto-tuning of Kalman Filters for Underwater Gliders based on Consistency Metrics," in Proc. OCEANS 2019 MTS/IEEE SEATTLE, 2019., doi:10.23919/OCEANS40490.2019.8962573
G. LaMountain, J. Vilà-Valls and P. Closas, "Measurement noise covariance estimation in Gaussian filters: an online Bayesian solution," EURASIP Journal on Advances in Signal Processing, vol. 2025, no. 1, 2025., doi:10.1186/s13634-025-01215-w
W. Li, X. Lin, S. Li, J. Ye, C. Yao and C. Chen, "Robust autocovariance least-squares noise covariance estimation algorithm," Measurement, vol. 187, Art. no. 110331, 2022., doi:10.1016/j.measurement.2021.110331
B. Sun, Z. Zhang, D. Qiao, X. Mu and X. Hu, "An Improved Innovation Adaptive Kalman Filter for Integrated INS/GPS Navigation," Sustainability, vol. 14, no. 18, Art. no. 11230, 2022., doi:10.3390/su141811230
Z. Li, H. Zhang, Q. Zhou and H. Che, "An Adaptive Low-Cost INS/GNSS Tightly-Coupled Integration Architecture Based on Redundant Measurement Noise Covariance Estimation," Sensors, vol. 17, no. 9, Art. no. 2032, 2017., doi:10.3390/s17092032
R. Xin, J. Lin, S. Shi, R. Zhang, J. Zhang and J. Zhu, "Research on dual-threshold detection based adaptive fault-tolerant kalman filtering algorithm for dynamic 6-DOF measurement," Mechanical Systems and Signal Processing, vol. 210, Art. no. 111190, 2024., doi:10.1016/j.ymssp.2024.111190
C. Hajiyev, "Adaptive Filtering Against Sensor/Actuator Faults," IFAC-PapersOnLine, vol. 58, no. 4, pp. 336–341, 2024., doi:10.1016/j.ifacol.2024.07.240
D. Wu, L. Xia and J. Geng, "Heading Estimation for Pedestrian Dead Reckoning Based on Robust Adaptive Kalman Filtering," Sensors, vol. 18, no. 6, Art. no. 1970, 2018., doi:10.3390/s18061970
S. Liu, S. Li, J. Zheng, Q. Fu and Y. Yuan, "C/N0 Estimator Based on the Adaptive Strong Tracking Kalman Filter for GNSS Vector Receivers," Sensors, vol. 20, no. 3, Art. no. 739, 2020., doi:10.3390/s20030739
Z. Wang, B. Li, Z. Dan, H. Wang and K. Fang, "3D LiDAR Aided GNSS/INS Integration Fault Detection, Localization and Integrity Assessment in Urban Canyons," Remote Sensing, vol. 14, no. 18, Art. no. 4641, 2022., doi:10.3390/rs14184641
W. Wang, W. Shangguan, J. Liu and J. Chen, "Enhanced Fault Detection for GNSS/INS Integration Using Maximum Correntropy Filter and Local Outlier Factor," IEEE Transactions on Intelligent Vehicles, vol. 9, no. 1, pp. 2077–2093, 2024., doi:10.1109/TIV.2023.3312654
L. Qian, F. Qin, A. Li, K. Li and J. Zhu, "An INS/DVL integrated navigation filtering method against complex underwater environment," Ocean Engineering, vol. 278, Art. no. 114398, 2023., doi:10.1016/j.oceaneng.2023.114398
X. Wu, Z. Su, L. Li and Z. Bai, "Improved Adaptive Federated Kalman Filtering for INS/GNSS/VNS Integrated Navigation Algorithm," Applied Sciences, vol. 13, no. 9, Art. no. 5790, 2023., doi:10.3390/app13095790
Z. Mahmoudi, K. Nørgaard, N. K. Poulsen, H. Madsen and J. B. Jørgensen, "Fault and meal detection by redundant continuous glucose monitors and the unscented Kalman filter," Biomedical Signal Processing and Control, vol. 38, pp. 86–99, 2017., doi:10.1016/j.bspc.2017.05.004
G. Gao, B. Gao, S. Gao, G. Hu and Y. Zhong, "A Hypothesis Test-Constrained Robust Kalman Filter for INS/GNSS Integration With Abnormal Measurement," IEEE Transactions on Vehicular Technology, vol. 72, no. 2, pp. 1662–1673, 2023., doi:10.1109/TVT.2022.3209091
Y. Cheng, S. Zhang, X. Wang, H. Wang and H. Yang, "Kalman Filter with Adaptive Covariance Estimation for Carrier Tracking under Weak Signals and Dynamic Conditions," Electronics, vol. 13, no. 7, Art. no. 1288, 2024., doi:10.3390/electronics13071288
H. Kim, A. Bienkowski and K. R. Pattipati, "A Single-pass Noise Covariance Estimation Algorithm in Multiple-model Adaptive Kalman Filtering," in Proc. 2023 IEEE Aerospace Conference, pp. 1–9, 2023., doi:10.1109/AERO55745.2023.10115725
D. C. Lowry, W. H. Woodall, C. W. Champ and S. E. Rigdon, "A Multivariate Exponentially Weighted Moving Average Control Chart," Technometrics, vol. 34, no. 1, pp. 46–53, 1992., doi:10.1080/00401706.1992.10485232
E. S. Page, "CONTINUOUS INSPECTION SCHEMES," Biometrika, vol. 41, no. 1 and 2, pp. 100–115, 1954., doi:10.1093/biomet/41.1-2.100
R. K. Mehra, "On the identification of variances and adaptive Kalman filtering," IEEE Transactions on Automatic Control, vol. 15, no. 2, pp. 175–184, 1970., doi:10.1109/TAC.1970.1099422
P. J. Huber, "Robust Estimation of a Location Parameter," Annals of Mathematical Statistics, vol. 35, no. 1, pp. 73–101, 1964., doi:10.1214/aoms/1177703732
Downloads
-
Vistas(Views): 55
- PDF (Español (España)) Descargas(Downloads): 44
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Francisco Javier Burgos Florez
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
The undersigned authors declare that the article submitted to the journal Scientia et Technica is an original work and that all its content is free of third-party copyright restrictions or has the corresponding authorizations. Consequently, the authors assume responsibility for any litigation or claim related to intellectual property rights, releasing the Technological University of Pereira and the journal Scientia et Technica from any liability.
If the submitted work is accepted for publication, the authors retain copyright to the article and grant the journal Scientia et Technica the right of first publication, as well as a non-exclusive, perpetual license to reproduce, edit, distribute, display, and publicly communicate the article in any medium or format, including print, electronic, databases, repositories, the Internet, or other scientific dissemination systems. The authors agree that the article will be published in open access and distributed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).
The journal Scientia will respect in all cases the moral rights of the authors, in accordance with the provisions of article 30 of Law 23 of 1982 of the Republic of Colombia, recognizing the authorship of the work, the right to integrity and the right of disclosure, which are inalienable and non-waivable.