arXiv subjects

Antonio Ferramosca

Publications and source records attributed to Antonio Ferramosca.

15 recordsLinked to original sources

arXiv2026

Integrating Biological-Informed Recurrent Neural Networks for Glucose-Insulin Dynamics Modeling

Type 1 Diabetes (T1D) management is a complex task due to many variability factors. Artificial Pancreas (AP) systems have alleviated patient burden by automating insulin delivery through advanced control algorithms. However, the effectiveness of these systems depends on accurate modeling of glucose-insulin dynamics, which traditional mathematical models often fail to capture due to their inability to adapt to patient-specific variations. This study introduces a Biological-Informed Recurrent Neural Network (BIRNN) framework to address these limitations. The BIRNN leverages a Gated Recurrent Units (GRU) architecture augmented with physics-informed loss functions that embed physiological constraints, ensuring a balance between predictive accuracy and consistency with biological principles. The framework is validated using the commercial UVA/Padova simulator, outperforming traditional linear models in glucose prediction accuracy and reconstruction of unmeasured states, even under circadian variations in insulin sensitivity. The results demonstrate the potential of BIRNN for personalized glucose regulation and future adaptive control strategies in AP systems.

cs.LG↗

arXiv2026

Infinity-norm-based Input-to-State-Stable Long Short-Term Memory networks: a thermal systems perspective

Recurrent Neural Networks (RNNs) have shown remarkable performances in system identification, particularly in nonlinear dynamical systems such as thermal processes. However, stability remains a critical challenge in practical applications: although the underlying process may be intrinsically stable, there may be no guarantee that the resulting RNN model captures this behavior. This paper addresses the stability issue by deriving a sufficient condition for Input-to-State Stability based on the infinity-norm (ISS$_{\infty}$) for Long Short-Term Memory (LSTM) networks. The obtained condition depends on fewer network parameters compared to prior works. A ISS$_{\infty}$-promoted training strategy is developed, incorporating a penalty term in the loss function that encourages stability and an ad hoc early stopping approach. The quality of LSTM models trained via the proposed approach is validated on a thermal system case study, where the ISS$_{\infty}$-promoted LSTM outperforms both a physics-based model and an ISS$_{\infty}$-promoted Gated Recurrent Unit (GRU) network while also surpassing non-ISS$_{\infty}$-promoted LSTM and GRU RNNs.

math.OC↗

arXiv2025

Robust tracking MPC for perturbed nonlinear systems -- Extended version

This paper presents a novel robust predictive controller for constrained nonlinear systems that is able to track piece-wise constant setpoint signals. The tracking model predictive controller presented in this paper extends the nonlinear MPC for tracking to the more complex case of nonlinear systems subject to bounded and not necessarily additive perturbations. The optimal control problem that is solved at each step penalizes the deviation of the predicted nominal system trajectory from an artificial reference, which is added as a decision variable, as well as the distance between the artificial reference and the setpoint. Robust feasibility is ensured by imposing conservative constraints that take into account the effect of uncertainties and convergence to a neighborhood of any feasible setpoint is guaranteed by means of an appropriate terminal cost and an extended stabilizing terminal constraint. In the case of unreachable setpoints, convergence to a neighborhood of the optimal reachable steady output is also proved.

eess.SY↗

arXiv2025

Robust contraction-based model predictive control for nonlinear systems

Model Predictive Control (MPC) is a widely known control method that has proved to be particularly effective in multivariable and constrained control. Closed-loop stability and recursive feasibility can be guaranteed by employing accurate models in prediction and suitable terminal ingredients, i.e. the terminal cost function and the terminal constraint. Issues might arise in case of model mismatches or perturbed systems, as the state predictions could be inaccurate, and nonlinear systems for which the computation of the terminal ingredients can result challenging. In this manuscript, we exploit the properties of component-wise uniformly continuous and stabilizable systems to introduce a robust contraction-based MPC for the regulation of nonlinear perturbed systems, that employs an easy-to-design terminal cost function, does not make use of terminal constraints, and selects the shortest prediction horizon that guarantees the stability of the closed-loop system.

eess.SY↗

arXiv2024

Model predictive control for tracking using artificial references: Fundamentals, recent results and practical implementation

This paper provides a comprehensive tutorial on a family of Model Predictive Control (MPC) formulations, known as MPC for tracking, which are characterized by including an artificial reference as part of the decision variables in the optimization problem. These formulations have several benefits with respect to the classical MPC formulations, including guaranteed recursive feasibility under online reference changes, as well as asymptotic stability and an increased domain of attraction. This tutorial paper introduces the concept of using an artificial reference in MPC, presenting the benefits and theoretical guarantees obtained by its use. We then provide a survey of the main advances and extensions of the original linear MPC for tracking, including its non-linear extension. Additionally, we discuss its application to learning-based MPC, and discuss optimization aspects related to its implementation.

eess.SY↗

arXiv2024

Model Predictive Control for setpoint tracking

The main objective of tracking control is to steer the tracking error, that is the difference between the reference and the output, to zero while the plant's operation limits are satisfied. This requires that some assumptions on the evolution of the future values of the reference must be taken into account. Typically a simple evolution of the reference is considered, such as step, ramp, or parabolic reference signals. It is important to notice that the tracking problem considers possible variations in the reference to be tracked, such as steps or slope variations of the ramps. Then the tracking control problem is inherently uncertain, since the reference may differ from what is expected. If the value of the reference is changed, then there is no guarantee that the feasibility and stability properties of the resulting control law hold. This report presents the MPC for tracking (MPCT) approach, which ensures recursive feasibility and asymptotic stability of the setpoint when the value of the reference is changed.

math.OC↗

arXiv2024

CHoKI-based MPC for blood glucose regulation in Artificial Pancreas

This work presents a Model Predictive Control (MPC) for the artificial pancreas, which is able to autonomously manage basal insulin injections in type 1 diabetic patients. Specifically, the MPC goal is to maintain the patients' blood glucose level inside the safe range of 70-180 mg/dL, acting on the insulin amount and respecting all the imposed constraints, taking into consideration also the Insulin On Board (IOB), to avoid excess of insulin infusion. MPC uses a model to make predictions of the system behaviour. In this work, due to the complexity of the diabetes disease that complicates the identification of a general physiological model, a data-driven learning method is employed instead. The Componentwise Hölder Kinky Inference (CHoKI) method is adopted, to have a customized controller for each patient. For the data collection phase and also to test the proposed controller, the virtual patients of the FDA-accepted UVA/Padova simulator are exploited. The proposed MPC is also tested on a modified version of the simulator, that takes into consideration also the variability of the insulin sensitivity. The final results are satisfying since the proposed controller reduces the time in hypoglycemia (which is more dangerous) if compared to the outcome obtained with the standard constant basal insulin therapy provided by the simulator, satisfying also the time in range requirements and avoiding long-term hyperglycemia events.

eess.SY↗

arXiv2023

GLISp-r: A preference-based optimization algorithm with convergence guarantees

Preference-based optimization algorithms are iterative procedures that seek the optimal calibration of a decision vector based only on comparisons between couples of different tunings. At each iteration, a human decision-maker expresses a preference between two calibrations (samples), highlighting which one, if any, is better than the other. The optimization procedure must use the observed preferences to find the tuning of the decision vector that is most preferred by the decision-maker, while also minimizing the number of comparisons. In this work, we formulate the preference-based optimization problem from a utility theory perspective. Then, we propose GLISp-r, an extension of a recent preference-based optimization procedure called GLISp. The latter uses a Radial Basis Function surrogate to describe the tastes of the decision-maker. Iteratively, GLISp proposes new samples to compare with the best calibration available by trading off exploitation of the surrogate model and exploration of the decision space. In GLISp-r, we propose a different criterion to use when looking for new candidate samples that is inspired by MSRS, a popular procedure in the black-box optimization framework. Compared to GLISp, GLISp-r is less likely to get stuck on local optima of the preference-based optimization problem. We motivate this claim theoretically, with a proof of global convergence, and empirically, by comparing the performances of GLISp and GLISp-r on several benchmark optimization problems.

math.OC↗

arXiv2023

Set-Point Tracking MPC with Avoidance Features

This work proposes a finite-horizon optimal control strategy to solve the tracking problem while providing avoidance features to the closed-loop system. Inspired by the set-point tracking model predictive control (MPC) framework, the central idea of including artificial variables into the optimal control problem is considered. This approach allows us to add avoidance features into the set-point tracking MPC strategy without losing the properties of an enlarged domain of attraction and feasibility insurances in the face of any changing reference. Besides, the artificial variables are considered together with an avoidance cost functional to establish the basis of the strategy, maintaining the recursive feasibility property in the presence of a previously unknown number of regions to be avoided. It is shown that the closed-loop system is recursively feasible and input-to-state-stable under the mild assumption that the avoidance cost is uniformly bounded over time. Finally, two numerical examples illustrate the controller behavior.

eess.SY↗

arXiv2023

Set-based MPC for discrete-time LTI systems with maximal domain of attraction and minimal predictive control horizon

This paper presents a novel set-based model predictive control for tracking, which provides the largest domain of attraction, even with the minimal predictive/control horizon. The formulation - which consists of a single optimization problem - shows a dual behavior: one operating inside the maximal controllable set to the feasible equilibrium set, and the other operating at the N-controllable set to the same equilibrium set. Based on some finite-time convergence results, asymptotic stability of the resulting closed-loop is proved, while recursive feasibility is ensured for any change of the setpoint. The properties and advantages of the proposal have been tested on simulation models.

eess.SY↗

arXiv2022

A unified surrogate-based scheme for black-box and preference-based optimization

Black-box and preference-based optimization algorithms are global optimization procedures that aim to find the global solutions of an optimization problem using, respectively, the least amount of function evaluations or sample comparisons as possible. In the black-box case, the analytical expression of the objective function is unknown and it can only be evaluated through a (costly) computer simulation or an experiment. In the preference-based case, the objective function is still unknown but it corresponds to the subjective criterion of an individual. So, it is not possible to quantify such criterion in a reliable and consistent way. Therefore, preference-based optimization algorithms seek global solutions using only comparisons between couples of different samples, for which a human decision-maker indicates which of the two is preferred. Quite often, the black-box and preference-based frameworks are covered separately and are handled using different techniques. In this paper, we show that black-box and preference-based optimization problems are closely related and can be solved using the same family of approaches, namely surrogate-based methods. Moreover, we propose the generalized Metric Response Surface (gMRS) algorithm, an optimization scheme that is a generalization of the popular MSRS framework. Finally, we provide a convergence proof for the proposed optimization method.

math.OC↗

arXiv2021

Optimal control strategies to tailor antivirals for acute infectious diseases in the host

Several mathematical models in SARS-CoV-2 have shown how target-cell model can help to understand the spread of the virus in the host and how potential candidates of antiviral treatments can help to control the virus. Concepts as equilibrium and stability show to be crucial to qualitatively determine the best alternatives to schedule drugs, according to effectivity in inhibiting the virus infection and replication rates. Important biological events such as rebounds of the infections (when antivirals are incorrectly interrupted) can also be explained by means of a dynamic study of the target-cell model. In this work, a full characterization of the dynamical behavior of the target-cell models under control actions is made and, based on this characterization, the optimal fixed-dose antiviral schedule that produces the smallest amount of dead cells (without viral load rebounds) is computed. Several simulation results - performed by considering real patient data - show the potential benefits of both, the model characterization and the control strategy.

math.OC↗

arXiv2020

Dynamical Characterization of Antiviral Effects in COVID-19

Mathematical models describing SARS-CoV-2 dynamics and the corresponding immune responses in patients with COVID-19 can be critical to evaluate possible clinical outcomes of antiviral treatments. In this work, based on the concept of virus spreadability in the host, antiviral effectiveness thresholds are determined to establish whether or not a treatment will be able to clear the infection. In addition, the virus dynamic in the host -- including the time-to-peak and the final monotonically decreasing behavior -- is chracterized as a function of the treatment initial time. Simulation results, based on nine real patient data, show the potential clinical benefits of a treatment classification according to patient critical parameters. This study is aimed at paving the way for the different antivirals being developed to tackle SARS-CoV-2.

math.DS↗

arXiv2020

Discrete-time MPC for switched systems with applications to biomedical problems

Switched systems in which the manipulated control action is the time-depending switching signal describe many engineering problems, mainly related to biomedical applications. In such a context, to control the system means to select an autonomous system - at each time step - among a given finite family. Even when this selection can be done by solving a Dynamic Programming (DP) problem, such a solution is often difficult to apply, and state/control constraints cannot be explicitly considered. In this work a new set-based Model Predictive Control (MPC) strategy is proposed to handle switched systems in a tractable form. The optimization problem at the core of the MPC formulation consists in an easy-to-solve mixed-integer optimization problem, whose solution is applied in a receding horizon way. Two biomedical applications are simulated to test the controller: (i) the drug schedule to attenuate the effect of viral mutation and drugs resistance on the viral load, and (ii) the drug schedule for Triple Negative breast cancer treatment. The numerical results suggest that the proposed strategy outperform the schedule for available treatments.

math.OC↗

arXiv2020

Characterization of SARS-CoV-2 Dynamics in the Host

While many epidemiological models have being proposed to understand and handle COVID-19, too little has been invested to understand how the virus replicates in the human body and potential antiviral can be used to control the replication cycle. In this work, using a control theoretical approach, validated mathematical models of SARS-CoV-2 in humans are properly characterized. A complete analysis of the main dynamic characteristic is developed based on the reproduction number. The equilibrium regions of the system are fully characterized, and the stability of such a regions, formally established. Mathematical analysis highlights critical conditions to decrease monotonically SARS-CoV-2 in the host, such conditions are relevant to tailor future antiviral treatments. Simulation results show the potential benefits of the aforementioned system characterization.

math.DS↗