arXiv subjects

Taous-Meriem Laleg-Kirati

Publications and source records attributed to Taous-Meriem Laleg-Kirati.

At least 19 recordsLinked to original sources

arXiv2026

Integrating Deep Learning and Contraction Theory for Robust Nonlinear State Estimation via Unsupervised Scientific Machine Learning

A common way to design observers is to add a correction term to a copy of the system; however, designing the correction term for nonlinear systems remains a significant long-standing challenge. Contraction theory offers a unified approach to designing this correction term by solving a matrix partial differential inequality (MPDI) and identifying a contraction metric. However, solving the MPDI for both the correction term and the contraction metric is highly challenging, both analytically and numerically. Therefore, the aim of this paper is to propose a learning-based approach to determine both the observer's correction term and the contraction metric by incorporating the contraction requirements into the learning process. The proposed approach relies on a scientific machine learning formulation that embeds the contraction conditions into the training loss function. The proposed approach is then extended to non-autonomous systems while keeping the correction term and contraction metric static to avoid generalization issues arising from time-dependent training. Computable bounds on the learning errors of the proposed observer are established as a function of the training residual and the sampling resolution. Furthermore, the robustness of the proposed observer to measurement noise and learning errors are established in an exponential input-to-state stability sense. Based on the robustness analysis, the present paper takes a further step by proposing a robust learning-based contraction nonlinear observer. The proposed observers are evaluated in numerical simulations for different contraction rates and measurement noise levels.

eess.SY↗

arXiv2026

SpaTeoGL: Spatiotemporal Graph Learning for Interpretable Seizure Onset Zone Analysis from Intracranial EEG

Accurate localization of the seizure onset zone (SOZ) from intracranial EEG (iEEG) is essential for epilepsy surgery but is challenged by complex spatiotemporal seizure dynamics. We propose SpaTeoGL, a spatiotemporal graph learning framework for interpretable seizure network analysis. SpaTeoGL jointly learns window-level spatial graphs capturing interactions among iEEG electrodes and a temporal graph linking time windows based on similarity of their spatial structure. The method is formulated within a smooth graph signal processing framework and solved via an alternating block coordinate descent algorithm with convergence guarantees. Experiments on a multicenter iEEG dataset with successful surgical outcomes show that SpaTeoGL is competitive with a baseline based on horizontal visibility graphs and logistic regression, while improving non-SOZ identification and providing interpretable insights into seizure onset and propagation dynamics.

cs.LG↗

arXiv2026

From Well-Posed Inversion to Learning Design: Physics- Informed Neural Estimation for Autonomic Regulation

Learning-based and physics-informed methods are increasingly used for inverse estimation in controlled nonlinear dynamical systems. However, in many such approaches, the theoretic requirements that make unknown-input reconstruction meaningful, namely well-posedness in the sense of Hadamard, are often disregarded or weakly addressed through generic regularization terms with no explicit guarantees. In this work, we adopt a complementary viewpoint in which these control-theoretic and structural conditions inform the estimator design and constrain its training. We thus develop a physics-informed input-state neural estimator for joint unknown-input and state estimation in nonlinear controlled systems with partial measurements. In the present work, this general framework is instantiated on a model of autonomic cardiac regulation, provides a concrete study case. The estimator is formulated as an inverse neural map conditioned on time and measured outputs, and is trained under data fidelity and dynamical consistency constraints. To ensure it complies with the same structural requirements imposed in robust estimation, we derive left-invertibility conditions by differential-algebraic elimination and embed the resulting constraints directly into the training objective. We further analyze a priori the stability of the inverse mapping to output perturbations and derive a conservative Lipschitz bound that guides the tuning of cost functional hyper-parameters. The framework is evaluated on simulated data, where ground truth data is available, and on two distinct datasets of real cardiovascular recordings. The results show that incorporating control-theoretic solvability constraints into physics-informed learning improves the reliability of inverse inference beyond forward consistency alone.

eess.SY↗

arXiv2026

An Explicit Surrogate for Gaussian Mixture Flow Matching with Wasserstein Gap Bounds

We study training-free flow matching between two Gaussian mixture models (GMMs) using explicit velocity fields that transport one mixture into the other over time. Our baseline approach constructs component-wise Gaussian paths with affine velocity fields satisfying the continuity equation, which yields to a closed-form surrogate for the pairwise kinetic transport cost. In contrast to the exact Gaussian Wasserstein cost, which relies on matrix square-root computations, the surrogate admits a simple analytic expression derived from the kinetic energy of the induced flow. We then analyze how closely this surrogate approximates the exact cost. We prove second-order agreement in a local commuting regime and derive an explicit cubic error bound in the local commuting regime. To handle nonlocal regimes, we introduce a path-splitting strategy that localizes the covariance evolution and enables piecewise application of the bound. We finally compare the surrogate with an exact construction based on the Gaussian Wasserstein geodesic and summarize the results in a practical regime map showing when the surrogate is accurate and the exact method is preferable.

cs.LG↗

arXiv2025

Radio-PPG: photoplethysmogram digital twin synthesis using deep neural representation of 6G/WiFi ISAC signals

Digital twins for 1D bio-signals enable real-time monitoring of physiological processes of a person, which enables early disease diagnosis and personalized treatment. This work introduces a novel non-contact method for digital twin (DT) photoplethysmogram (PPG) signal synthesis under the umbrella of 6G/WiFi integrated sensing and communication (ISAC) systems. We employ a software-defined radio (SDR) operating at 5.23 GHz that illuminates the chest of a nearby person with a wideband 6G/WiFi signal and collects the reflected signals. This allows us to acquire Radio-PPG dataset that consists of 300 minutes worth of near synchronous 64-channel radio data, PPG data, along with the labels (three body vitals) of 30 healthy subjects. With this, we test two artificial intelligence (AI) models for DT-PPG signal synthesis: i) discrete cosine transform followed by a multi-layer perceptron, ii) two U-NET models (Approximation network, Refinement network) in cascade, along with a custom loss function. Experimental results indicate that U-NET model achieves an impressive relative mean absolute error of 0.194 with a small ISAC sensing overhead of 15.62%, for DT-PPG synthesis. Furthermore, we performed quality assessment of the synthetic DT-PPG by computing the accuracy of DT-PPG-based vitals estimation and feature extraction, which turned out to be at par with that of reference PPG-based vitals estimation and feature extraction. This work highlights the potential of generative AI and 6G/WiFi ISAC technologies and serves as a foundational step towards the development of non-contact screening tools for covid-19, cardiovascular diseases and well-being assessment of people with special needs.

cs.IT↗

arXiv2024

Internet of medical things for non-invasive and non-contact dehydration monitoring away from the hospital: state-of-the-art, challenges and prospects

Dehydration occurs when the body loses more water than it takes in. Mild dehydration can lead to fatigue, cognitive impairments, and physical complications, while severe dehydration can cause life-threatening conditions like heat stroke, kidney damage, and hypovolemic shock. Traditional bio chemistry-based clinical gold standard methods are expensive, time-consuming, and invasive. Thus, there is a pressing need to design novel non-invasive methods that could do in-situ, early and accurate detection of dehydration, which will in turn allow timely intervention. This article presents a methodological review of the literature on a range of innovative internet of medical things-based techniques for dehydration monitoring. We begin by briefly describing the pathophysiology of the dehydration problem, its clinical significance, and current clinical gold-standard methods for assessing hydration level. Subsequently, we critically examine a number of non-invasive and non-contact hydration assessment studies. We also discuss multi-modal sensing methods and assess the impact of dehydration among specific population groups (e.g., elderly, infants, athletes) and on different organs. We also provide a list of existing public and private datasets which make the backbone of machine learning-driven research on dehydration monitoring. Finally, we provide our opinion statement on the challenges and future prospects of non-invasive and non-contact hydration monitoring.

eess.SP↗

arXiv2024

Unsupervised Physics-Informed Neural Network-based Nonlinear Observer design for autonomous systems using contraction analysis

Contraction analysis offers, through elegant mathematical developments, a unified way of designing observers for a general class of nonlinear systems, where the observer correction term is obtained by solving an infinite dimensional inequality that guarantees global exponential convergence. However, solving the matrix partial differential inequality involved in contraction analysis design is both analytically and numerically challenging and represents a long-lasting challenge that prevented its wide use. Therefore, the present paper proposes a novel approach that relies on an unsupervised Physics Informed Neural Network (PINN) to design the observer's correction term by enforcing the partial differential inequality in the loss function. The performance of the proposed PINN-based nonlinear observer is assessed in numerical simulation as well as its robustness to measurement noise and neural network approximation error.

eess.SY↗

arXiv2024

Image contrast enhancement based on the Schrödinger operator spectrum

In this study, we propose a novel image contrast enhancement method based on projecting images onto the squared eigenfunctions of the two-dimensional Schrödinger operator. This projection relies on a design parameter, $γ$, which controls pixel intensity during image reconstruction. The method's performance is evaluated using color images. The selection of $γ$ values is guided by priors based on fuzzy logic and clustering, preserving the spatial adjacency information of the image. Additionally, multi-objective optimization using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is employed to determine the optimal values of $γ$ and the semi-classical parameter, $h$, from the 2D-SCSA. Results demonstrate that the proposed method effectively enhances image contrast while preserving the inherent characteristics of the original image, producing the desired enhancement with minimal artifacts.

cs.CV↗

arXiv2024

Non-Contact Acquisition of PPG Signal using Chest Movement-Modulated Radio Signals

We present for the first time a novel method that utilizes the chest movement-modulated radio signals for non-contact acquisition of the photoplethysmography (PPG) signal. Under the proposed method, a software-defined radio (SDR) exposes the chest of a subject sitting nearby to an orthogonal frequency division multiplexing signal with 64 sub-carriers at a center frequency 5.24 GHz, while another SDR in the close vicinity collects the modulated radio signal reflected off the chest. This way, we construct a custom dataset by collecting 160 minutes of labeled data (both raw radio data as well as the reference PPG signal) from 16 healthy young subjects. With this, we first utilize principal component analysis for dimensionality reduction of the radio data. Next, we denoise the radio signal and reference PPG signal using wavelet technique, followed by segmentation and Z-score normalization. We then synchronize the radio and PPG segments using cross-correlation method. Finally, we proceed to the waveform translation (regression) task, whereby we first convert the radio and PPG segments into frequency domain using discrete cosine transform (DCT), and then learn the non-linear regression between them. Eventually, we reconstruct the synthetic PPG signal by taking inverse DCT of the output of regression block, with a mean absolute error of 8.1294. The synthetic PPG waveform has a great clinical significance as it could be used for non-contact performance assessment of cardiovascular and respiratory systems of patients suffering from infectious diseases, e.g., covid19.

eess.SP↗

arXiv2024

You can monitor your hydration level using your smartphone camera

This work proposes for the first time to utilize the regular smartphone -- a popular assistive gadget -- to design a novel, non-invasive method for self-monitoring of one's hydration level on a scale of 1 to 4. The proposed method involves recording a small video of a fingertip using the smartphone camera. Subsequently, a photoplethysmography (PPG) signal is extracted from the video data, capturing the fluctuations in peripheral blood volume as a reflection of a person's hydration level changes over time. To train and evaluate the artificial intelligence models, a custom multi-session labeled dataset was constructed by collecting video-PPG data from 25 fasting subjects during the month of Ramadan in 2023. With this, we solve two distinct problems: 1) binary classification (whether a person is hydrated or not), 2) four-class classification (whether a person is fully hydrated, mildly dehydrated, moderately dehydrated, or extremely dehydrated). For both classification problems, we feed the pre-processed and augmented PPG data to a number of machine learning, deep learning and transformer models which models provide a very high accuracy, i.e., in the range of 95% to 99%. We also propose an alternate method where we feed high-dimensional PPG time-series data to a DL model for feature extraction, followed by t-SNE method for feature selection and dimensionality reduction, followed by a number of ML classifiers that do dehydration level classification. Finally, we interpret the decisions by the developed deep learning model under the SHAP-based explainable artificial intelligence framework. The proposed method allows rapid, do-it-yourself, at-home testing of one's hydration level, is cost-effective and thus inline with the sustainable development goals 3 & 10 of the United Nations, and a step-forward to patient-centric healthcare systems, smart homes, and smart cities of future.

eess.SP↗

arXiv2024

Modulating function based algebraic observer coupled with stable output predictor for LTV and sampled-data systems

This paper proposes an algebraic observer-based modulating function approach for linear time-variant systems and a class of nonlinear systems with discrete measurements. The underlying idea lies in constructing an observability transformation that infers some properties of the modulating function approach for designing such algebraic observers. First, we investigate the algebraic observer design for linear time-variant systems under an observable canonical form for continuous-time measurements. Then, we provide the convergence of the observation error in an L2-gain stability sense. Next, we develop an exponentially stable sampled-data observer which relies on the design of the algebraic observer and an output predictor to achieve state estimation from available measurements and under small inter-sampling periods. Using a trajectory-based approach, we prove the convergence of the observation error within a convergence rate that can be adjusted through the fixed time-horizon length of the modulating function and the upper bound of the sampling period. Furthermore, robustness of the sampled-data algebraic observer, which yields input-to-state stability, is inherited by the modulating kernel and the closed-loop output predictor design. Finally, we discuss the implementation procedure of the MF-based observer realization, demonstrate the applicability of the algebraic observer, and illustrate its performance through two examples given by linear time-invariant and linear time-variant systems with nonlinear input-output injection terms.

eess.SY↗

arXiv2023

Tracking Control of Optical Beam Transceivers using Mean Field Models

This paper proposes mean field models to maintain an accurate line-of-sight and tracking between transceivers mounted in mobile unmanned aerial vehicles (UAVs) platforms in the presence of underlying mechanical vibration effects. We describe a two-way optical link beam tracking control that coordinates mobile UAVs in a network architecture to provide reliable network structure, distributed connectivity, and communicability, enhancing terrestrial public safety communication systems. We derive the optical transceiver trajectory tracking problem in which each agent dynamic and cost function is coupled with other optical beam transceiver agent states via a mean field term. We propose two optimal mean field beam tracking control frameworks through decentralized and centralized strategies in which the optical transceivers compete to reach a Nash equilibrium and cooperate to attain a social optimum, respectively. The solutions of these strategies are derived from forward-backward ordinary differential equations and rely on the linearity Hamilton-Jacobi-Bellman Fokker-Planck (HJB-FP) equations and stochastic maximum principle. Moreover, we numerically compute the solution pair of the resulting joint equations using Newton and fixed point iteration methods to verify the existence and uniqueness of the equilibrium and social optimum.

eess.SY↗

arXiv2023

Model-based versus model-free feeding control and water quality monitoring for fish growth tracking in aquaculture systems

The high concentration level of the environmental factors, such as a high ammonia concentration and pH level, affect the water quality, affecting fish's survival and mass death. Therefore, there is a critical need to develop control strategies to determine optimal, efficient, and reliable feeding and water quality monitoring processes. In this paper, we revisit the representative fish growth model describing the total biomass change by incorporating the fish population density and mortality. Since the measurement data of the total biomass and population from the aquaculture systems are limited and difficult to obtain, we validate the new dynamic population model with the individual fish growth data for tracking control purposes. We specifically focus on relative feeding as a manipulated variable to design traditional and optimal control to track the desired weight reference within the sub-optimal temperature and dissolved oxygen profiles under different levels of unionized ammonia exposure. Then, we propose a Q-learning approach that learns an optimal feeding control policy from the simulated data of the fish growth weight trajectories while managing the ammonia effects. The proposed Q-learning feeding control prevents fish mortality and achieves good tracking errors of the fish weight under the different levels of unionized ammonia. However, it maintains a relative food consumption that potentially underfeeds the fish. Finally, we propose an optimal algorithm that optimizes the feeding and water quality of the dynamic fish population growth process. We also show that the model predictive control decreases fish mortality and reduces food consumption in all different cases of unionized ammonia exposure.

eess.SY↗

arXiv2023

Feeding control and water quality monitoring in aquaculture systems: Opportunities and challenges

Aquaculture systems can benefit from the recent development of advanced control strategies to reduce operating costs and fish loss and increase growth production efficiency, resulting in fish welfare and health. Monitoring the water quality and controlling feeding are fundamental elements of balancing fish productivity and shaping the fish growth process. Currently, most fish-feeding processes are conducted manually in different phases and rely on time-consuming and challenging artificial discrimination. The feeding control approach influences fish growth and breeding through the feed conversion rate; hence, controlling these feeding parameters is crucial for enhancing fish welfare and minimizing general fishery costs. The high concentration of environmental factors, such as a high ammonia concentration and pH, affect the water quality and fish survival. Therefore, there is a critical need to develop control strategies to determine optimal, efficient, and reliable feeding processes and monitor water quality. This paper reviews the main control design techniques for fish growth in aquaculture systems, namely algorithms that optimize the feeding and water quality of a dynamic fish growth process. Specifically, we review model-based control approaches and model-free reinforcement learning strategies to optimize the growth and survival of the fish or track a desired reference live-weight growth trajectory. The model-free framework uses an approximate fish growth dynamic model and does not satisfy constraints. We discuss how model-based approaches can support a reinforcement learning framework to efficiently handle constraint satisfaction and find better trajectories and policies from value-based reinforcement learning.

eess.SY↗

arXiv2023

Non-Asymptotic State and Disturbance Estimation for a Class of Triangular Nonlinear Systems using Modulating Functions

Dynamical models are often corrupted by model uncertainties, external disturbances, and measurement noise. These factors affect the performance of model-based observers and as a result, affect the closed-loop performance. Therefore, it is critical to develop robust model-based estimators that reconstruct both the states and the model disturbances while mitigating the effect of measurement noise in order to ensure good system monitoring and closed-loop performance when designing controllers. In this article, a robust step by step non-asymptotic observer for triangular nonlinear systems for the joint estimation of the state and the disturbance is developed. The proposed approach provides a sequential estimation of the states and the disturbance in finite time using smooth modulating functions. The robustness of the proposed observer is both in the sense of model disturbances and measurement noise. In fact, the structure of triangular systems combined with the modulating function-based method allows the estimation of the states independently of model disturbances and the integral operator involved in the modulating function-based method mitigates the noise. Additionally, the modulating function method shifts the derivative from the noisy output to the smooth modulating function which strengthens its robustness properties. The applicability of the proposed modulating function-based estimator is illustrated in numerical simulations and compared to a second-order sliding mode super twisting observer under different measurement noise levels.

eess.SY↗

arXiv2022

Spatial Distribution Patterns of Clownfish in Recirculating Aquaculture Systems

Successful aquaculture systems can reduce the pressure and help secure the most diverse and productive Red Sea coral reef ecosystem to maintain a healthy and functional ecosystem within a sustainable blue economy. Interestingly, recirculating aquaculture systems are currently emerging in fish farm production practices. On the other hand, monitoring and detecting fish behaviors provide essential information on fish welfare and contribute to an intelligent production in global aquaculture. This work proposes an efficient approach to analyze the spatial distribution status and motion patterns of juvenile clownfish \textit{(Amphiprion bicinctus)} maintained in aquaria at three stocking densities (1, 5, and 10 individuals/aquarium). The estimated displacement is crucial in assessing the dispersion and velocity to express the clownfish's spatial distribution and movement behavior in a recirculating aquaculture system. Indeed, we aim to compute the velocity, magnitude, and turning angle using an optical flow method to assist aquaculturists in efficiently monitoring and identifying fish behavior. We test the system design on a database containing two days of video streams of juvenile clownfish maintained in aquaria. The proposed displacement estimation reveals good performance in measuring clownfish's motion and dispersion characteristics leading to assessing the potential signs of stress behaviors. We demonstrate the effectiveness of the proposed technique for quantifying variation in clownfish activity levels between recordings taken in the morning and afternoon at different stocking densities. It provides practical baseline support for online predicting and monitoring feeding behavior in ornamental fish aquaculture.

cs.CV↗

arXiv2022

Accelerating Extremum Seeking Convergence by Richardson Extrapolation Methods

In this paper, we propose the concept of accelerated convergence that has originally been developed to speed up the convergence of numerical methods for extremum seeking (ES) loops. We demonstrate how the dynamics of ES loops may be analyzed to extract structural information about the generated output of the loop. This information is then used to distil the limit of the loop without having to wait for the system to converge to it.

eess.SY↗

arXiv2021

Data-Driven State Estimation for Light-Emitting Diode Underwater Optical Communication

Light-Emitting Diodes (LEDs) based underwater optical wireless communications (UOWCs), a technology with low latency and high data rates, have attracted significant importance for underwater robots. However, maintaining a controlled line of sight link between transmitter and receiver is challenging due to the constant movement of the underlying optical platform caused by the dynamic uncertainties of the LED model and vibration effects. Additionally, the alignment angle required for tracking is not directly measured and has to be estimated. Besides, the light scattering propagates beam pulse in water temporally, resulting in time-varying underwater optical links with interference. We address the state estimation problem by designing an LED communication system that provides the angular position and velocity information to overcome the challenges. In this way, we leverage the power of deep learning-based observer design to explore the LED communication's state space properly. Simulation results are presented to illustrate the performance of the data-driven LED state estimation.

eess.SY↗