Keywords: Data driven methods, Design for reliability and safety, Fault-forecasting methods
Abstract: This paper develops a novel data-driven framework for predicting Remaining Useful Life (RUL) in nonlinear closed-loop dynamical systems subject to gradual control input modulated parametric degradation. A hybrid Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) architecture is developed that learns in a supervised manner, from standard control-loop signals such as setpoints, controller outputs, and system responses, eliminating the need for explicit degradation models. We demonstrate our approach on a thermo‑hydraulic MIMO simulation platform, analogous to a liquid‑propellant‑rocket subsystem. The hybrid CNN–LSTM not only outperforms standalone CNN and LSTM models, but also maintains accurate RUL estimates across diverse scenarios (abrupt reference changes, control saturation), highlighting its robustness to realistic disturbances and its suitability for closed‑loop prognostics.

Keywords: Health monitoring, Data driven methods, Automobile
Abstract: The health monitoring of key processes or subsystems in the manufacturing industry is of capital importance to improve maintenance policies. This work will look for degradation patterns in the aluminum die casting process used to generate hybrid engine blocks in the automotive industry. The die-casting process that generates each engine block is rather complex, and each die-casting machine can generate a new engine block every 90 seconds. Within this process, the aluminum injection stage is critical and it lasts only few seconds. The injection device monitoring system provides 2000 measurements of several physical variables involved in the process. This work has faced the challenges of providing an estimation of the degradation patterns of the piston head for one of these injection machines in a factory, using those time series obtained from the controller, and also to identify unusual malfunction patterns found during a three-month period. The problem was tackled using both traditional and deep machine-learning techniques for unsupervised learning. Results have shown that degradation patterns can be identified, but the presence of unexpected malfunctions will require additional techniques and/or data.

Keywords: Health monitoring, Data driven methods, Chemical processes
Abstract: Predictive maintenance focuses on predicting the end-of-life of components to optimize human intervention and parts usage, though it necessitates continuous monitoring. A key aspect of predictive maintenance is calculating the Remaining Useful Life (RUL), which estimates how much longer a system can function effectively. Traditionally, there are three main approaches to calculating RUL: model-based, data-driven, and hybrid methods. Data-driven approaches, whether statistical or neural network-based, typically require large datasets. This paper proposes a model-free approach to calculate the RUL of lithium-ion batteries. The method uses a polynomial to approximate capacity over cycles, with the coefficients determined through least mean squares optimization under linear constraints. With an average prediction horizon of 30 % remaining lifetime, this approach is both computationally efficient and minimizes data requirements.

Keywords: Health monitoring, Fault-forecasting methods, Data driven methods
Abstract: This paper presents a hybrid clustering-regression framework for predicting the Remaining Useful Life (RUL) of industrial Li-ion batteries. A health indicator derived from battery enthalpy is first extracted to track degradation. Battery life is then segmented into distinct phases using unsupervised clustering, comparing K-means and Fuzzy C-Means (FCM) methods. A phase-specific regression model is trained within each cluster to predict RUL more accurately. Experiments on real-world data show that the fuzzy clustering combined with regression reduces the Root Mean Squared Error (RMSE) by 31% and achieves a higher coefficient of determination (R2) compared to a global model. Beyond improved accuracy, the method provides interpretable insights into battery degradation dynamics, making it suitable for predictive maintenance applications.

Keywords: Health monitoring, Model-based methods, Power plants / energy transport
Abstract: Solid Oxide Fuel Cells are promising power generation technologies, especially for large-scale applications. As the marine industry is targeting a full de-carbonization by year 2050, increasing attention is being directed toward the implementation of these technologies. Solid Oxide Fuel Cells are complex systems where thermodynamics and electrochemical reactions are coupled, resulting in highly non-linear dynamics, tight operational constraints, and multiple distributed sensors. Those quantities that cannot be directly measured, need to be estimated. Among these, the so called Area Specific Resistance is an indicator of cell's health condition, related to the cell degradation. This paper proposes a Moving Horizon Estimator based on an extended state-space model of a methane-fueled Solid Oxide Fuel Cell, to estimate in real time the Area Specific Resistance of the cell. Using the estimated value, along with its maximum and average rates, a predictive framework is developed to estimate the Remaining Useful Life of the cell. Simulations are used to illustrate the application and the efficiency of the proposed method.

Keywords: Fault detection and isolation, Multi-agent systems, Model-based methods
Abstract: This paper deals with the problem of detecting,isolating and estimating of multiple sensor and actuator faults in stochastic linear discrete-time systems. A linear state filter is designed to generate minimum variance output residuals having directional properties in response to sensor and actuator fault structurally detectable at current time. The filter’s gain is obtained by minimizing the state prediction errors covariance matrix under decoupling constraints activated as late as possible from a feedback information on the real-time detectability conditions of sensor and actuator faults. During a finite-time structural transient, the degrees of freedom used to minimize the trace of the state prediction error covariance matrix are maximized. Two dual solutions for multiple sensor and actuator faults detection and isolation are derived, the first based on one fault isolation filter generating minimum size structured detection signals and the second based on a bank of fault isolation filter generating maximum size white structured detection signals. The convergence and stability conditions of the optimized fault isolation filter with structural transient are established from an upper bound of the state prediction errors covariance matrix obtained when the decoupling constraints are always activated.

Keywords: Fault detection and isolation, Statistical and signal processing, Manufacturing systems
Abstract: This paper introduces a sound fingerprint approach to detect mechanical faults in industrial machinery using only audio data from normal operations as reference dataset. Unlike traditional machine learning models that require large labeled datasets and frequent retraining, this method is data-efficient and fast, making it well-suited to industrial conditions. By comparing audio fingerprints of unlabeled machine sounds against a small database of known healthy samples, the method achieves high anomaly detection accuracy, even under varying speeds, background noise, and microphone types. Experiments on the ToyADMOS2 dataset show that the technique outperforms baselines in challenging conditions, using far fewer reference samples. This makes it ideal for dynamic industrial environments where faults are rare, noise is common, and quick deployment is essential. The result is a practical, scalable solution for real-time, explainable machine condition monitoring without the complexity of black-box models.

Keywords: Fault detection and isolation, Supervisory control, Chemical processes
Abstract: This paper presents the design of an interval observer (IO) for nonlinearities that are assumed to fulfill one-sided Lipschitz quadratically inner-bounded (OSL-QIB) conditions. This novel approach extends conventional design of IO for Lipschitz systems to OSL-QIB systems with the aim of improving observer design and convergence. Additionally, this new IO design is applied to fault detection, which is very important in control systems theory, since it can ensure the confidence of a process and detect difficulties during its operation. The study cases applied for fault detection and state estimation are a bioreactor for biomass production from substrate and a FHN system, respectively.

Keywords: Mechatronic and robotics, Fault detection and isolation, Model-based methods
Abstract: This paper presents a novel method for fault detection and isolation in linear time-varying systems. The approach extends the nullspace-based residual filter design technique originally developed for linear time-invariant systems and previously adapted for fault detection in LTV systems. Fault isolation is achieved by treating certain faults as disturbances, enabling a significant extension of the detection scheme to isolate specific faults through residual decoupling. The method leverages symbolic nullspace computations to account explicitly for the time-varying nature of the problem. The proposed fault detection and isolation framework is demonstrated on a robotic manipulator example, where time variations arise from motion along a predefined trajectory. Simulation results based on a nonlinear model validate the effectiveness of the approach.

Keywords: Fault detection and isolation, Fault-forecasting methods, Data driven methods
Abstract: The aim of this research is to detect faults in the air scavenge system of a marine dual-fuel engine. Scavenge air receiver non-return flaps are critical mechanical components that ensure proper dual-fuel marine engine functionality and safety. The aim is to detect whether these flaps are broken or damaged. Scavenge air temperature after cooler and scavenge air pressure signals were analyzed using a multivariate anomaly detection Generative Adversarial Network (GAN) method. Anomalies detected by the method showed alignment with ground truth data of faults described in a technical report. The approach demonstrated an ability to detect faults weeks before they were reported, showcasing its effectiveness. Further analysis is to be done upon the availability of more data.

Keywords: Autonomous vehicles, Automobile, Networked control system
Abstract: This paper introduces a reinforcement learning (RL)-based framework for adaptively combining four different control strategies in response to varying operational conditions. Starting from a nominal model of the actual system, several feedback controllers are developed, each offering distinct performance benefits under different circumstances. The RL algorithm dynamically determines and mixes the outputs of the controller within specific operating ranges. Four control approaches are considered: Linear Parameter Varying (LPV), Ultra-local Model-based (ULM), Linear Quadratic Regulator (LQR), and a kinematic model-based method. The proposed solution is validated through different test scenarios using the high-fidelity vehicle simulation platform, CarMaker.

Keywords: Autonomous vehicles, Fault tolerant control / fault recovery
Abstract: This paper proposes a zonotopic filtering-based active Fault Tolerant Tracking Control (FTTC) scheme for autonomous vehicles, modelled as Linear Parameter-Varying (LPV) systems subject to disturbances, noises and actuator faults. A Zonotopic Kalman Filter (ZKF) is proposed to simul- taneously estimate both the unmeasurable states and actuator faults. Then, a fault tolerant tracking controller is developed, including a reference input, a state-feedback controller and a fault compensation mechanism. To minimize the impact of uncertainties on the estimates, the optimal filter gain is designed by minimizing the Frobenius norm of the state bounding zonotope. To reduce the tracking error and attenuate the disturbance effects, the controller design ensures the H∞ performance of the uncertain tracking error dynamics by constructing a zonotopic Lyapunov function. Finally, the proposed scheme is applied to vehicle lateral dynamics for reference path-tracking control in the presence of actuator faults, and its effectiveness is demonstrated through simulations using real collected data.

Keywords: Fault tolerant control / fault recovery, Maintenance policies, Automobile
Abstract: Automotive electric machines represent the core of the driving and supply line of the vehicle; therefore, it is crucial to ensure an acceptable and reliable performance all the time. This research paper presents a novel approach of fault-tolerant control based on model reference adaptive control for automotive electric machines. The approach is applied to a parameterized model of the alternator and takes advantage of the emerging concept of digital twin. It extends the work of previous research articles in the domain of fault detection and identification to include a mechanism to maintain the desired performance and ensure a safe and acceptable driving experience. The simulation results have shown powerful performance recovery provided by the adaptive control law.

Keywords: Model-based methods, Transportation systems
Abstract: This paper addresses the modeling and control of fuel consumption in diesel-powered cargo ships, focusing on ensuring robust performance under varying operating conditions. The objective is to regulate the ship's propulsion and engine dynamics by adjusting the engine power command, despite the presence of external disturbances such as waves and wind. The nonlinear behavior of the fuel consumption system is represented using a Takagi-Sugeno (T-S) fuzzy model derived from multiple operating scenarios. Based on this representation, a fuzzy static output feedback (SOF) controller is designed using the descriptor approach. The proposed control strategy guarantees stability and satisfactory fuel regulation performance through conditions formulated as linear matrix inequalities (LMIs). Simulation results demonstrate the effectiveness and robustness of the control scheme under realistic conditions.

Keywords: Multi-agent systems, Networked control system
Abstract: This paper deals with fault estimation in a network of agents with a linear, positive strictly Metzler structure, whose communication topology is defined by an undirected graph. For the estimation, an observer approach is used to estimate additive system faults based on the communication topology of the multi-agent system, using measurements of the relative outputs of the systems. The required gains of the estimator are obtained by solving an algebraic matrix equation and a set of linear matrix inequalities. The properties of the proposed approach are illustrated by an example.

Keywords: Data driven methods, Discrete event and hybrid systems, Fault-forecasting methods
Abstract: We present the implementation of a Cloud-Computing-based system for leak diagnosis in a water distribution network. Pressure head/flow rate measured in field is stored and processed in a virtual machine to provide a leak diagnosis in two stages: 1) first the leak is detected by comparing the current operating conditions and expected nominal operating conditions obtained from a previously adjusted simulation model, and 2) the leak exact location and its magnitude are determined using a meta-heuristic method. The performance of the proposed system is implemented for an experimental hydraulic system at a laboratory scale. Results demonstrate a good accuracy in the leak diagnosis metrics at a reduced economic and computational cost, demonstrating the potential results from the implementation of a similar system in a large-scale water distribution network.

Keywords: Water treatment, Fault detection and isolation, Data driven methods
Abstract: Water Distribution Networks (WDNs), critical to public well-being and economic stability, face challenges such as pipe blockages and background leakages, exacerbated by operational constraints such as data non-stationarity and limited labeled data. This paper proposes an unsupervised, online learning framework that aims to detect two types of faults in WDNs: pipe blockages, modeled as collective anomalies, and background leakages, modeled as concept drift. Our approach combines a Long Short-Term Memory Variational Autoencoder (LSTM-VAE) with a dual drift detection mechanism, enabling robust detection and adaptation under non-stationary conditions. Experiments on two realistic WDNs show that the proposed approach consistently outperforms strong baselines in detecting anomalies and adapting to recurrent drift, demonstrating its effectiveness in unsupervised event detection for dynamic WDN environments.

Keywords: Data driven methods
Abstract: Water utilities around the world typically use chlorine as the main disinfectant for ensuring high-quality drinking water. Usually, a few fixed sensors monitor water quality by detecting changes in parameters like chlorine residuals, guiding chlorination strategies. However, limited sensor coverage leaves most parts of the network unmonitored. Additionally, rapid urban growth and climate change complicate water quality dynamics, challenging conventional methods for sparse to dense state estimation. In this work, we propose a neural network based surrogate model for efficiently obtaining time-dependent approximations of the chlorine concentration dynamics in a water distribution system. We incorporate this surrogate model into an extended Kalman filter to estimate all chlorine concentration states on the basis of only a few sensors. We perform extensive empirical evaluations on popular benchmark water distribution systems from the literature.

Keywords: Data driven methods, Transportation systems, Statistical and signal processing
Abstract: This paper proposes a method to estimate both the friction factor and the roughness coefficient of the pipes in a water distribution network (WDN). Pressure head data is used for the training of a Long-Short Term Memory (LSTM) neural network to predict the pressure head at all of the other nodes in the network, as well as the flow rate through the pipes. When all the data is predicted, the estimated pressure head drops, and the flow rate for the entire water distribution network is used to compute both the friction factor and the flow rate. The proposed method is tested using the well-known Hanoi DMA (District Metered Area) as a case study. Satisfactory results are obtained when comparing the predicted friction factor and roughness coefficient against the expected values.

Keywords: Decision making, Multi-agent systems, Maintenance policies
Abstract: Efficient and sustainable water management is imperative due to the mounting pressure on global water supplies from over-exploitation, desertification, and pollution. Integrated Water Resource Management (IWRM) strategies have demonstrated efficacy in decision support; however, a more comprehensive integration of participatory and economic methodologies is required. The objective of this research is to enhance water resource management through collaborative, stakeholder-driven innovation by integrating experimental economics with Living Labs (LLs). Living Labs offer genuine environments for collaborative creation, enabling scientists and stakeholders to resolve water-related concerns such as supply, demand, and scarcity. These environments establish a connection between controlled experimental conditions and real-world applications, offering a comprehensive understanding of policy formulation and behavioral reactions. We use the Limassol Water Futures Living Lab (LWFLL) as a case study that is dedicated to the creation of a comprehensive, intelligent decision-making framework that will enable the effective management of water resources in the presence of unpredictable climate conditions. LLs can be strengthened and improved by economic methodologies, particularly in water valuation, through integrated frameworks that account for environmental externalities and opportunity costs. Real-time input is provided by technological innovations such as smart meters, desalination technologies or soil moisture sensors, which enables dynamic pricing models to accurately depict the economic and environmental costs associated with water consumption. Experimental economics' external validity is enhanced by the integration of behavioral insights and experimental approaches into LLs, whi

Keywords: Aeronautics / aerospace, Fault tolerant control / fault recovery, Fault detection and isolation
Abstract: This short communication aims to sketch some methodological pathways for addressing new safety challenges in future civil aviation operations. It integrates model-based methods inspired from robust control theories with data-driven learning-enabled techniques, incorporating human oversight. The goal is to formulate key enablers for robust, scalable, cyber-physical and human-centric innovative solutions. The purpose of this note is to offer general methodological directions rather than specific technical developments.

Keywords: Model-based methods, Aeronautics / aerospace, Design for reliability and safety
Abstract: This paper investigates the input-to-state stabilization of a class of uncertain nonlinear systems using a Parameter-Varying Persidskii (PVP) modeling approach. A key advantage over traditional Linear Parameter-Varying (LPV) methods, common in aeronautics, is that it does not require embedding all nonlinearities, leading to controllers with improved global stability and convergence properties. This advantage allows for a more direct treatment of nonlinearities and avoids limitations associated with local LPV approximations, such as issues in handling large parameter variations. Within this PVP framework, a novel stabilizing control design method is developed and applied to the longitudinal motion control of an aircraft. The effectiveness of the proposed approach is demonstrated and compared with LPV-based and classical gain-scheduling control strategies.

Keywords: Risk analysis, Fault tolerant control / fault recovery, Road infrastructure
Abstract: Ensuring worker safety in road construction zones is a critical challenge due to the inherently dynamic and unpredictable nature of these environments. This paper pro- poses the Adaptive-weighting and Fault-tolerant Drone-based Roadside Operator Protection System (AF-DROPS), a cyber- physical-human system that leverages aerial and ground data collection and elaboration to perform real-time dynamic risk management on road construction sites. The AF-DROPS framework is composed of four communicating subsystems: Aerial Detection (ADS), Ground Detection (GDS), Monitoring and Control (MCS), and Actuation (AS). In the ADS, a drone and a computer vision algorithm are devoted to monitoring the construction site and to measuring the distance of vehicles circulating near the site. A central master unit in the MCS continuously evaluates a dynamic Danger Index whose weighting factors are adapted in real-time. The AS receives control signals from the MCS and activates multimodal alerts (visual, auditory, and tactile) depending on the danger level. In addition, a manual distress signal function is included that ensures fast and reliable communication of workers’ alerts. AF-DROPS allows improving workers’ safety in road construction sites by the continuous implementation of the main risk management phases, i.e., Risk Analysis, Assessment, Treatment and Communication, and Monitoring and Review. In addition, fault-tolerant control is ensured through system redundancy: if the ADS fails, a smart cone in the GDS seamlessly ensures uninterrupted hazard monitoring and worker protection. Experimental tests show the AF-DROPS effectiveness and performance in realistic environment.

Keywords: Multi-agent systems, Model-based methods
Abstract: This paper addresses distributed event-triggered attitude estimation for multiple rigid bodies evolving on SO3. Based on relative attitude measurements under an undirected graph, the estimation errors asymptotically converge to a common orientation. A clock-based event-triggered condition is designed to reduce communication and computation loads. Rigorous analysis shows that the desired equilibrium is almost globally asymptotically stable. The proposed method is validated via a simulation involving multiple small satellites.

Keywords: Fault tolerant control / fault recovery, Networked control system, Model-based methods
Abstract: This paper considers the optimal fault-tolerant control (FTC) issue for network nonlinear systems developed by an inverse nested differential game (INDG), which is composed of zero-sum games between controllers and faults in subsystems and a graphical game between subsystems. The proposed INDG-based optimal FTC framework guides the design of meaningful performance indexes with respect to inner faults and coupling faults, such that the fault-tolerant controllers can be constructed to achieve the optimality and stability of the whole system. Sufficient criteria of INDG-based optimal FTC are proposed, and the stability of the network system is rigorously proved under the designed performance indexes and optimal FTC.

Keywords: Decision making, Data driven methods
Abstract: This study examines the dynamics of citizens' policy attitudes for the allocation of public resources for natural disaster prevention and response, with a focus on the role of experience with extreme environmental events and perceived probability of future events. Through a nationally representative survey currently underway in three US states (California, New York, and Texas), we investigate the influence of geographic and emotional proximity to extreme events in shaping relevant preferences. The results presented are from the first wave of the study (Wave 1), with subsequent waves already planned to be incorporated into the final version of the study. The preliminary analysis suggests that individuals prioritise resource allocation towards recently experienced shocks, and that this prioritization is not driven by subjective risk assessment alone. The final phase of the research, through the collection of data from subsequent waves, will allow us to investigate the temporal duration and dynamics of the impact of external shocks on citizens' political attitudes.

Keywords: Decision making, Design for reliability and safety
Abstract: Water distribution systems (WDS) are critical urban infrastructures reliably delivering safe drinking water and supporting firefighting activities. When developing a system master plan, water utilities may choose—or be compelled by their environment—to prioritize different reliability perspectives, such as hydraulic, mechanical, or firefighting. Mathematical optimization can aid in system intervention planning, but accurate reliability assessments require many computationally expensive simulations. To reduce this burden, previous studies have proposed reliability surrogate metrics and explored their correlation with the actual definitions. Instead, this work examines how adopting different reliability perspectives in WDS optimization affects system design and the potential cost of misrepresenting a utility’s planning perspective. This question is particularly relevant because of the long planning horizons, as deep uncertainties can affect utility priorities over time. Different bi-objective optimizations of the Anytown benchmark system are performed, each reflecting a distinct perspective that a water utility might adopt. The cost objective—always present due to tight budget constraints—is paired with another objective expressing the utility’s most pressing concern: operational efficiency, reliability under pipe failures, or fire response capacity. An evolutionary algorithm seeks Pareto-optimal solutions for these WDS design optimization problems. Results show that each perspective prioritizes different network components at comparable budget levels. WDS design should prioritize hydraulic reliability and daily operational efficiency, treating other goals as constraints or secondary objectives to broaden design exploration.

Keywords: Fault detection and isolation, Quality monitoring, Civil engineering
Abstract: Contamination diagnosis in water distribution systems (WDS) is critical for ensuring public health and maintaining water quality. However, exact localization of contamination sources is often highly challenging due to sparse sensor deployment, uncertain water flow dynamics, and the ambiguity of contamination signatures. To address this challenge, we propose a novel end-to-end deep learning framework for ranked contamination diagnosis in WDS. Instead of relying on a single-source prediction, the deep learning model outputs a probabilistic ranking of likely contamination source nodes, enabling top-k diagnosis strategies under uncertainty. A unified classification layer jointly performs contamination detection and source localization, allowing the model to operate in an end-to-end fashion. Extensive experiments on a benchmark WDS demonstrate that the proposed framework achieves strong top-k diagnosis accuracy and remains robust across varying sensor configurations. This work offers a practical and scalable approach to contamination diagnosis in WDS, supporting more informed and effective decision-making in real-world operations. The code is available at: https://github.com/Xiaohan-Chen/ST-GAT.

Keywords: Quality monitoring, Civil engineering, Maintenance policies
Abstract: Biofilms in drinking water distribution systems (DWDS) pose a critical challenge to water quality. If left unchecked, they can compromise the biological stability of delivered water and ultimately public health. Existing biofilm sensing techniques primarily focus on metabolic or genetic indicators of activity, often using local and destructive methods. While rich in information, such data are difficult to apply in developing practical biofilm growth models. Biofilm thickness, however, is a more representative and scalable metric for this purpose. Yet, limited research exists on non-invasive thickness sensing in DWDS. This study introduces two non-destructive methods for measuring biofilm thickness by leveraging changes in heat resistance and residence time. Heat resistance was evaluated using ambient and water temperature measurements, while residence time was assessed with a conservative tracer. Both techniques were tested in the Slimer experimental setup (50 m long, 13.2 mm diameter PVCp pipe) under realistic hydraulic conditions. Results showed a strong correlation between biofilm thickness and residence time drift, indicating flow disturbance as a reliable indicator of biofouling. In contrast, heat resistance sensing exhibited considerable natural variability, limiting its analytic value. The findings highlight residence time analysis as a promising, non-invasive approach for estimating biofilm thickness. This method offers continuous, non-destructive monitoring, enabling early detection of biofilm-related anomalies and providing valuable input for both laboratory and field applications aimed at enhancing DWDS resilience.

Keywords: Risk analysis
Abstract: This study investigates the relationship between Environmental, Social, and Governance (ESG) momentum and Sustainable Development Goals (SDG) integration within international equity markets. Leveraging a robust dataset spanning 2002–2023, we identify pronounced ESG momentum effects in stock returns across 63 global markets. Our ESG momentum factor, derived through monthly rebalancing, demonstrates an impressive, annualized Sharpe ratio of 0.7, underscoring its financial viability. Beyond returns, the study highlights the pivotal role of ESG controversies in shaping short-term financial performance. We advanced the discourse by integrating ESG principles with the SDG framework, proposing a novel model to calculate the SDG footprint of financial portfolios. This alignment between ESG momentum and SDG implementation emerges as a significant tool for investors and policymakers, particularly considering regulatory advancements like the Corporate Sustainability Reporting Directive (CSRD).

Keywords: Aeronautics / aerospace, Fault tolerant control / fault recovery
Abstract: This paper presents a fault-tolerant controller for a bio-inspired miniature UAV. The UAV is equipped with a centrally mounted propeller and utilises control surfaces for attitude control. However, the UAV omits the vertical tail fin commonly found on typical fixed-wing aircraft. The main wings are designed as symmetrically folding structures that allow the wings to change dihedral angle, which can be manipulated collectively for control. The proposed controller is synthesised using LPV-based Sliding Mode Control Allocation to handle the morphing of the wings, variations in the flight conditions and actuator faults and failures. Simulations based on the nonlinear model show the efficacy of the proposed scheme.

Keywords: Design for reliability and safety, Fault tolerant control / fault recovery, Aeronautics / aerospace
Abstract: Ensuring both safety and fault tolerance in reinforcement learning (RL)-based control systems is critical for their deployment in aerospace systems. In this context, the aim of this work is twofold. First, a new zonotopic reachability algorithm is proposed to address the case of bounded inputs with bounded variations. It is obtained from a reformulation based on bounded delayed exogenous inputs. Second, this algorithm is used for offline and training-independent safety verification of control loops hybridizing a baseline model-based controller with a constrained reinforcement learning (CRL) agent. The agent is then trained not only to improve given performance metrics under unmodeled environmental dynamics, but also to achieve incipient fault accommodation. A numerical example based on aircraft flight dynamics illustrates the main results and shows the efficiency of the proposed scheme.

Keywords: Fault detection and isolation, Aeronautics / aerospace
Abstract: This paper introduces a novel methodology for designing linear time-varying (LTV) fault detection filters and solve the LTV fault detection problem for LTV systems. The proposed approach builds upon the nullspace-based design technique formulated for linear time-invariant (LTI) systems. The general definition of the LTV fault detection problem sets the basis for the extension of the nullspace-based method to LTV systems. In contrast to the numeric solutions for LTI fault detention problems, symbolic nullspace computations are proposed herein in order to explicitly take the time-dependent nature of the design problem into account. The proposed methodology is demonstrated using the launch trajectory of a space rocket with its time-varying dynamics. The fault detection performance and robustness against false alarms of the proposed LTV detection filter are verified through nonlinear simulations of the rocket along the launch trajectory.

Keywords: Fault detection and isolation, Aeronautics / aerospace
Abstract: Thrust Vector Control (TVC) systems are critical components of space launch vehicles, as actuator faults can significantly compromise mission success. With Electro-Mechanical Actuators (EMAs) increasingly adopted in launcher TVC systems due to their cost-effectiveness and reduced integration complexity, this paper introduces a model-based Fault Detection and Diagnosis (FDD) strategy specifically targeting Loss-of-Efficiency (LoE) faults in EMA-driven TVC systems. The proposed approach leverages the nullspace method, a linear model-based FDD technique, while handling the vehicle-induced loads and uncertainties. The system dynamics and synthesis model under LoE fault conditions are thoroughly explained. The FDD system is validated through Monte-Carlo simulation campaigns using both a perturbed linear TVC model and a high-fidelity nonlinear model.

Keywords: Fault detection and isolation, Model-based methods, Aeronautics / aerospace
Abstract: This paper deals with the investigation of a UAV crash applying post accident system identification and modelbased fault detection. After crashing a UAV initial examination of flight log data showed that the possible cause was a left elevon stuck fault. This is verified applying flight data based system identification, then model-based fault detection and isolation. As the verification was successful this possibly shows a new method for future accident investigations where the detection of the cause of accident is not straightforward from the flight log or pilot observations.

Keywords: Supervisory control, Fault tolerant control / fault recovery, Aeronautics / aerospace
Abstract: This paper aims at demonstrating how the dwelltime fault-tolerant control (FTC) approach can be used to accommodate faults occurring in control surfaces of high speed aircraft. The application support is the F-8 aircraft that performs a coordinated turn. Global exponential stability of the FTC law is proven, considering the coupling between the fault diagnosis and control units. Beyond the application case, the paper demonstrates how the structured H infinity approach can be embedded within the dwell-time FTC theory. A simulation campaign considering sensor noise, control surfaces saturation and uncertainties in mass, inertia and aerodynamic coefficients, demonstrates the potential of the proposed approach.