A Consumption-Based Approach for Fault Diagnosis in Multi-DOF Robots

Integrating AI with State Estimation for Fault Detection in Dynamic Systems: Methods, Challenges, and Opportunities

State estimation is a fundamental component of model-based Fault Detection and Diagnosis (FDD) in dynamic systems, underpinning real-time monitoring, predictive maintenance, and safety-critical operations across industries such as aerospace, power systems, robotics, and autonomous vehicles. Traditional estimators, including the Kalman Filter (KF) and its variants, provide physically interpretable residuals for fault detection but often fail to deliver reliable performance under nonlinear dynamics, modeling uncertainties, sensor faults, and non-Gaussian noise. This paper presents a comprehensive review of state estimation-based FDD approaches, with a particular focus on Artificial Intelligence (AI)-augmented Kalman filtering and hybrid frameworks that integrate Machine Learning (ML) models, including Neural Networks (NNs), Support Vector Machines (SVMs), and Gaussian Processes (GPs), with classical estimation theory. The review systematically evaluates model-based, data-driven, and hybrid methods, comparing their robustness, accuracy, computational efficiency, scalability, and interpretability in complex Cyber-Physical Systems (CPSs). Furthermore, emerging trends and open research challenges are identified, including online adaptation, fault-tolerant estimation, sensor fusion, explainable artificial intelligence (XAI), and deployment in Industry 4.0 and Internet of Things (IoT)-enabled environments. By bridging classical estimation theory with modern AI techniques, this review provides a roadmap for designing intelligent, adaptive, and resilient FDD systems capable of enhancing reliability, operational safety, and real-world applicability.

Robust reinforcement learning fault diagnosis for complex data-driven nonlinear systems

A Review of Fault Diagnosis of Model-based Mechatronic Systems

Fault diagnosis is a critical discipline within modern engineering that ensures the operational reliability, structural safety, and maximum availability of complex mechatronic systems. As contemporary automated processes become deeply integrated with mechanical, electrical, and computational sub-units, tracking internal health status transitions from an operational preference to a fundamental necessity. This paper provides an exhaustive, large-scale review of fault diagnosis methodologies, focusing explicitly on the architecture of model-based systems. Model-based fault diagnosis leverages high-fidelity mathematical and topological representations of physical processes to detect, isolate, and identify faults by evaluating analytical discrepancies, or residuals, between actual system responses and anticipated model states. This comprehensive review systematically categorizes these methodologies into qualitative approaches—including abstraction hierarchies, fault tree models, signed directed graphs, and fuzzy logic formulations—and quantitative mathematical techniques, encompassing analytical redundancy, parity spaces, Kalman filtering frameworks, parametric estimation algorithms, and diagnostic observer topologies. Furthermore, complementary architectural considerations, such as hardware redundancy and history-based data analytics, are investigated to build an integrated view of modern industrial developments. A detailed taxonomic synthesis highlights current system implementation hurdles, trade-offs between mathematical precision and computational complexity, and future diagnostic research paradigms for safety-critical systems.

Simulation-Generated Data for Fault Detection and Diagnosis in Mobile Robots

Fault detection and diagnosis is a crucial task for modern mobile robots, as it permits their correct functioning with a positive impact on availability, autonomy and safety. Despite several approaches of fault detection and diagnosis of sensor faults in mobile robots, there exists a lack of scientific studies addressing the low-cost LiDAR faults which are commonly adopted in indoor robots. This paper is a preliminary attempt to fill this research gap by proposing a data-driven approach for developing a diagnostic module for mobile robots, focusing on the specific case study of the Stretch robot.Several LiDAR faults are first modeled, then simulated in ROS and Gazebo environments to generate a high-quality dataset. The dataset is used for training a machine learning model, which can diagnose such faults as well as distinguish them from other typical faults, such as IMU pose drift.Our proposed model shows high detection and isolation accuracies across the injected fault scenarios, thus paving the way for the deployment on a real robotic system for further evaluation and analysis.

Data-driven edge intelligence two-axis servo system: multimodal end-to-end friction diagnosis

Industrial dual-axis servo systems serve as the core actuators for achieving precise positioning in high-end equipment such as intelligent robots and precision manipulation robots; therefore, their control accuracy directly affects the precision and stability of the robot’s end-effector operations. However, the inherent nonlinear friction in guide rails (the Striebeck effect) can cause single-axis tracking errors or dual-axis contour distortion, thereby affecting the performance of micro- and nanomanipulation as well as precision assembly. This paper proposes a physics-based, edge AI-driven multimodal fault diagnosis method. Specifically, low-frequency armature current and high-frequency vibration signals are effectively fused, and a heterogeneous multimodal representation is constructed using high-fidelity dynamic simulation methods at the system’s perception layer, thereby achieving exceptional noise resistance. Extensive robustness analysis demonstrates the framework’s resilience to signal noise, ensuring reliable deployment in dynamic industrial environments. Simulation results indicate that the proposed framework achieves a diagnostic accuracy of 98.73%, significantly outperforming single-modal baselines (40.39%/77.03%). The proposed end-to-end method avoids the need for complex manual feature engineering, features low computational complexity and high deployment convenience, and meets the stringent computational constraints of embedded robot controllers. Furthermore, with its extremely low computational complexity, this algorithm provides an excellent solution to fault diagnosis in precision motion control for intelligent robots.

Neural-Augmented Torque Control for Robotic Manipulators: Modeling, Learning, and Real-Time Performance

This thesis investigates advanced modeling and control strategies for robotic manipulators, focusing on the DLR-HIT II robotic hand and the KUKA LBR iiwa. It presents three core contributions that integrate simulation, model-based control, and data-driven methods to improve torque and position control under uncertainties and disturbances. First, a dual-platform simulation framework is developed using MATLAB Simscape Multibody and CoppeliaSim. The system accurately models the DLR-HIT II hand’s kinematics and dynamics, enabling both control validation and realistic interaction with virtual environments. The use of Unified Robot Description Format (URDF)-based modeling sup-ports reusability and modular analysis. Second, a physics-informed neural network (PINN) is proposed for direct torque and position control. This method uses only time and joint position inputs, internally computes derivatives, and generalizes well across various trajectory types. It eliminates the need for separate feedback controllers and shows strong robustness under disturbances, while maintaining low computational cost. Third, a Spike-Aware Hybrid Torque Control (SA-HTC) architecture is enhanced with a feedforward neural network (NN) trained offline. The network learns to improve Computed Torque Control (CTC) for unmodeled effects, friction, and external forces by refining the CTC torque output in real time. Simulation results across diverse trajectories and noise levels demonstrate that the SA-HTC method significantly improves tracking accuracy and robustness compared to classical CTC. To support deployment on position-controlled hardware, a physics-informed torque-to-position interface is introduced and compared with virtual stiffness and admittance wrappers, yielding lower steady-state bias, reduced phase lag, lower tracking error, and robust cross-trajectory generalization. Thus, these contributions advance the integration of learning-based and model-based control strategies in robotics. The results highlight scalable and efficient methods for accurate trajectory tracking and torque control, offering practical potential for robotic manipulation and automation applications.