Robust State and Fault Estimation in Mobile Robots under Dynamic Noise Environments using Hybrid LSTM-EKF with Adaptive Weighting

Document Type : Research Paper

Authors
Karim Khemiri 1
Mokhtar Ferhi 1
Noureddine Hidouri 2
Ridha Ennetta 3
Ridha Djebali 1
1 UR22ES12: Modelling, Optimization and Augmented Engineering, ISLAIB, University of Jendouba, Beja, 9000, Tunisia
2 Laboratoire de Recherche PEESE, Université de Gabès, Ecole Nationale d'Ingénieurs de Gabes, Gabes, Tunisie
3 Mechanical Modeling, Energy & Materials Laboratory, National School of Engineers, Gabes University, Zrig, 6029, Gabes, Tunisia
10.22055/jacm.2025.48393.5203
Abstract
This paper proposes an adaptive hybrid estimation framework combining a Long Short-Term Memory (LSTM) network and an Extended Kalman Filter (EKF) for simultaneous robot state estimation and actuator fault detection in mobile robots operating under uncertain conditions. The LSTM network learns temporal patterns and nonlinear relationships from sensor data, while the EKF provides model-based filtering with uncertainty quantification. A novel adaptive fusion mechanism dynamically balances these complementary approaches using a weighting factor derived from the estimation uncertainties of both components. The proposed method was extensively evaluated on a differential-drive robot model subject to various noise conditions and fault profiles, including progressive drift faults and abrupt jump faults. Simulation results demonstrate that our hybrid approach significantly outperforms standalone EKF and LSTM estimators, achieving up to 74.6% improvement in fault estimation accuracy and 34.5% reduction in position error under challenging high-noise conditions. The framework maintains consistent performance across diverse fault types, showing particular effectiveness in detecting gradual fault progression while remaining responsive to sudden fault events. These findings confirm that the adaptive LSTM-EKF fusion provides enhanced accuracy, robustness, and generalization capability compared to conventional approaches.
Keywords
Mobile robot localization
state and fault estimation
robustness in noisy environments
adaptive fusion factor
LSTM-EKF hybrid model
Subjects

Publisher’s Note Shahid Chamran University of Ahvaz remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

[1] Shafiezadeh, A., Bhatt, N.P., Hashemi, E., LiDAR-Based navigation using normal distributions transform filter, 2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC), 2024, 4046–4051.
[2] Chakraborty, S., Verma, A., Hartman, A., Evaluation of Visual Inertial Navigation System for Autonomous Robot Tours on Campus, 2024 IEEE 3rd International Conference on Computing and Machine Intelligence (ICMI), 2024, 1–7.
[3] Fekrmandi, H., Frye, A.J., Tamjidi, A., Rakoczy, J., Hoover, R.C., Autonomous multi-agent systems using SVGS camera sensor for lunar surface mobility applications, IEEE Aerospace Conference, 2021, 1–10.
[4] Ceccarelli, N., Di Marco, M., Garulli, A., Giannitrapani, A., Vicino, A., Set membership localization and map building for mobile robots, Birkhäuser Boston eBooks, 2006, 289–308.
[5] Li, Y., Xu, X., The application of EKF and UKF to the SINS/GPS Integrated Navigation systems, 2010 2nd International Conference on Information Engineering and Computer Science, Wuhan, China, 2010, 1–5.
[6] Ding, L., Wen, C., High-Order Extended Kalman filter for state estimation of nonlinear systems, Symmetry, 16, 2024, 617.
[7] Eichstädt, S., Makarava, N., Elster, C., On the evaluation of uncertainties for state estimation with the Kalman filter, Measurement Science and Technology, 27, 2016, 125009.
[8] Alsaggaf, A.U., Saberi, M., Berry, T., Ebeigbe, D., Nonlinear kalman filtering in the absence of direct functional relationships between measurement and state, IEEE Control Systems Letters, 2024, 1.
[9] Wang, Y., Truccolo, W., Borton, D.A., Decoding hindlimb kinematics from primate motor cortex using long short-term memory recurrent neural networks, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, USA, 2018, 1944–1947.
[10] Schneider, J-N., Gorißen, L., Kaster, T., Walderich, P., Hinke, C., LSTM-based Inverse Dynamics Learning for Franka Emika Robot, 2024 International Conference on Control, Automation and Diagnosis (ICCAD), Paris, France, 2024, 1–6.
[11] Kumar, S.D., Dyanesh, S., Madhushree, K., Madhuram, M., Advanced Condition Monitoring and Fault Detection in AC Motors using Machine Learning Techniques, 2025 7th International Conference on Intelligent Sustainable Systems (ICISS), India, 2025, 748–754.
[12] Davari, N., Veloso, B., De Assis Costa, G., Pereira, P.M., Ribeiro, R.P., Gama, J., A Survey on Data-Driven Predictive Maintenance for the Railway industry, Sensors, 21, 2021, 5739.
[13] Cohen, N., Klein, I., Inertial Navigation meets Deep Learning: a survey of current trends and future directions, Results in Engineering, 24, 2024, 103565.
[14] Singh, A., Kalaichelvi, V., Karthikeyan, R., Machine learning-based multi-sensor fusion for warehouse robot in GPS-denied environment, Multimedia Tools and Applications, 83, 2023, 56229–56246.
[15] Song, F., Li, Y., Cheng, W., Dong, L., Learning to Track Multiple Radar Targets with Long Short-Term Memory Networks, Wireless Communications and Mobile Computing, 2023, 2023, 1–9.
[16] Saied, M., Mishi, A., Francis, C., Noun, Z., A deep learning approach for Fault-Tolerant data fusion applied to UAV position and orientation estimation, Electronics, 13, 2024, 3342.
[17] Yu, B., Wang, G., Zhu, E., Yao, S., Zhou, Y., Predicting lithium-ion battery state of charge with long short-term memory network enhanced extended Kalman filter, Journal of Energy Storage, 132, 2025, 117849.
[18] Liu, X., Hu, Y., Konstantinou, C., Jin, Y., CHIMERA: A Hybrid Estimation Approach to Limit the Effects of False Data Injection Attacks, 2021 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), Aachen, Germany, 2021, 95–101.
[19] Zhang, M., Peng, C., Li, G., Bu, X., Wang, Y., Research on Indoor Localization Algorithm for Emergency Rescue Personnel Based on INS/UWB, 2025 6th International Conference on Electrical, Electronic Information and Communication Engineering (EEICE 2025), Shenzhen, China, 2025, 700–705.
[20] Zhou, Z-L., Cheng, Y-R., Mao, G-T., Peng, S-F., Neural network-based error correction algorithm for inertial navigation in deep-sea mining vehicles, Marine Systems & Ocean Technology, 20, 2025, 43.
[21] Xu, H., Zhao, J., Zhang, H., Jiang, J., Chen, L., Target tracking method based on LSTM-EKF, Lecture Notes in Electrical Engineering, 2024, 60–67.
[22] Wagstaff, B., Kelly, J., LSTM-Based Zero-Velocity Detection for Robust Inertial Navigation, 2018 International Conference on Indoor Positioning and Indoor Navigation (IPIN), Nantes, France, 2018, 1–8.
[23] Zheng, K., Wei, H., Li, P., Gong, B., A Strong Tracking Unscented Kalman Filter Algorithm, 2024 3rd International Conference on Electronics and Information Technology (EIT), Chengdu, China, 2024, 841–846.
[24] Russell, R.L., Reale, C., Multivariate uncertainty in deep learning, IEEE Transactions on Neural Networks and Learning Systems, 33, 2021, 7937–7943.
[25] Yin, S., Li, P., Gu, X., Yang, X., Yu, L., Adaptive Kalman filter with LSTM network assistance for abnormal measurements, Measurement Science and Technology, 35, 2024, 075113.
[26] Lin, X., Chao, S., Yan, D., Guo, L., Liu, Y., & Li, L. Multi-Sensor Data Fusion Method Based on Self-Attention Mechanism, Applied Sciences, 13(21), 2023, 11992.
[27] Ma, Y., Qin, Y., Zhang, H., Jiang, K., Research on Multimodal Data Fusion Based on Optimal Adaptive Coding, 2024 5th International Conference on Artificial Intelligence and Computer Engineering (ICAICE), Wuhu, China, 2024, 364–369.
[28] Yu, X., Wang, J., Zhang, K., Chen, Z., Tong, M., Sun, S., Shen, J., Zhang, L., Wang, C., Research on missing data estimation Method for UPFC submodules based on Bayesian multiple Imputation and Support vector machines, Energies, 18, 2025, 2535.
[29] Yue, J., Lang, J., Feng, R., An adaptive feature fusion strategy using dual-layer attention and multi-modal deep reinforcement learning for all-media similarity search, Discover Artificial Intelligence, 5, 2025, 71.
[30] Xu, C., Zhao, H., Xie, H., Gao, B., Multi-sensor decision-level fusion network based on attention mechanism for object detection, IEEE Sensors Journal, 24, 2024, 31466–31480.
[31] Hou, D., Cao, M., A hybrid deep learning model approach for performance index prediction of mechanical equipment, Measurement Science and Technology, 33, 2022, 105108.
[32] Ma, X., Yan, T., Wang, B., Feng, Y., Integrating Wavelet Reconstruction with Hybrid TCN-LSTM Model for Enhanced Fault Diagnosis in Distributed Photovoltaic System, 2021 China Automation Congress (CAC), 2024, 5987–5992.
[33] Najdi, B., Benbrahim, M., Kabbaj, M.N., Bearing Fault Diagnosis with a Hybrid CWT-ResNet-LSTM Model, Lecture Notes in Networks and Systems, 2024, 454–463.
[34] Hage, J.A., Mafrica, S., Najjar, M.E.B.E., Ruffier, F., Informational framework for Minimalistic visual odometry on Outdoor robot, IEEE Transactions on Instrumentation and Measurement, 68, 2018, 2988–2995.
[35] Oonk, S., Maldonado, F.J., Li, Z., Reichard, K., Pentzer, J., Extended kalman filter for improved navigation with fault awareness, IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2014, 2681–2686.
[36] Sadeghzadeh-Nokhodberiz, N., Poshtan, J., Distributed interacting multiple filters for fault diagnosis of navigation sensors in a robotic system, IEEE Transactions on Systems Man and Cybernetics Systems, 47, 2016, 1383–1393.
[37] Geng, K., Chulin, N.A., Wang, Z., Fault-Tolerant Model Predictive Control Algorithm for path tracking of autonomous vehicle, Sensors, 20, 2020, 4245.
[38] Aghili, F., Su, C-Y., Robust relative navigation by integration of ICP and adaptive Kalman filter using laser scanner and IMU, IEEE/ASME Transactions on Mechatronics, 21, 2016, 2015–2026.
Journal of Applied and Computational Mechanics
Volume 12, Issue 2
Spring 2026
Pages 726-738