Searcharxiv⌕ Search

arXiv subjects

Francesco De Lellis

Publications and source records attributed to Francesco De Lellis.

At least 19 recordsLinked to original sources
arXiv2026

ROPE: A Novel Method for Real-Time Phase Estimation of Complex Biological Rhythms

Accurate phase estimation -- the process of assigning phase values between $0$ and $2π$ to repetitive or periodic signals -- is a cornerstone in the analysis of oscillatory signals across diverse fields, from neuroscience to robotics, where it is fundamental, e.g., to understanding coordination in neural networks, cardiorespiratory coupling, and human-robot interaction. However, existing methods are often limited to offline processing and/or constrained to one-dimensional signals. In this paper, we introduce ROPE, which, to the best of our knowledge, is the first phase-estimation algorithm capable of (i) handling signals of arbitrary dimension and (ii) operating in real-time, with minimal error. ROPE identifies repetitions within the signal to segment it into (pseudo-)periods and assigns phase values by performing efficient, tractable searches over previous signal segments. We extensively validate the algorithm on a variety of signal types, including trajectories from chaotic dynamical systems, human motion-capture data, and electrocardiographic recordings. Our results demonstrate that ROPE is robust against noise and signal drift, and achieves significantly superior performance compared to state-of-the-art phase estimation methods. This advancement enables real-time analysis of complex biological rhythms, opening new pathways, for example, for early diagnosis of pathological rhythm disruptions and developing rhythm-based therapeutic interventions in neurological and cardiovascular disorders.

eess.SP↗
arXiv2025

Adaptive Optimal Control for Avatar-Guided Motor Rehabilitation in Virtual Reality

A control-theoretic framework for autonomous avatar-guided rehabilitation in virtual reality, based on interpretable, adaptive motor guidance through optimal control, is presented. The framework faces critical challenges in motor rehabilitation due to accessibility, cost, and continuity of care, with over 50% of patients inability to attend regular clinic sessions. The system enables post-stroke patients to undergo personalized therapy in immersive virtual reality at home, while being monitored by clinicians. The core is a nonlinear, human-in-the-loop control strategy, where the avatar adapts in real time to the patient's performance. Balance between following the patient's movements and guiding them to ideal kinematic profiles based on the Hogan minimum-jerk model is achieved through multi-objective optimal control. A data-driven "ability index" uses smoothness metrics to dynamically adjust control gains according to the patient's progress. The system was validated through simulations and preliminary trials, and shows potential for delivering adaptive, engaging and scalable remote physiotherapy guided by interpretable control-theoretic principles.

eess.SY↗
arXiv2025

A bioreactor-based architecture for in vivo model-based and sim-to-real learning control of microbial consortium composition

Microbial consortia offer significant biotechnological advantages over monocultures for bioproduction. However, industrial deployment is hampered by the lack of scalable architectures to ensure stable coexistence between populations. Existing strategies rely on genetic modifications, which impose metabolic load, or environmental changes, which can reduce production. We present a versatile control architecture to regulate density and composition of a two-strain consortium without genetic engineering or drastic environmental changes. Our bioreactor-based control architecture comprises a mixing chamber where both strains are co-cultured and a reservoir sustaining the slower-growing strain. For both chambers we develop model-based and sim-to-real learning controllers. The control architecture is then validated in vivo on a two-strain Escherichia coli consortium, achieving precise and robust regulation of consortium density and composition, including tracking of time-varying references and recovery from perturbations.

eess.SY↗
arXiv2025

Hierarchical Learning-Based Control for Multi-Agent Shepherding of Stochastic Autonomous Agents

Multi-agent shepherding represents a challenging distributed control problem where herder agents must coordinate to guide independently moving targets to desired spatial configurations. Most existing control strategies assume cohesive target behavior, which frequently fails in practical applications where targets exhibit stochastic autonomous behavior. This paper presents a hierarchical learning-based control architecture that decomposes the shepherding problem into a high-level decision-making module and a low-level motion control component. The proposed distributed control system synthesizes effective control policies directly from closed-loop experience without requiring explicit inter-agent communication or prior knowledge of target dynamics. The decentralized architecture achieves cooperative control behavior through emergent coordination without centralized supervision. Experimental validation demonstrates superior closed-loop performance compared to state-of-the-art heuristic control methods, achieving 100\% success rates with improved settling times and control efficiency. The control architecture scales beyond its design conditions, adapts to time-varying goal regions, and demonstrates practical implementation feasibility through real-time experiments on the Robotarium platform.

eess.SY↗
arXiv2025

Breaking the mould of Social Mixed Reality - State-of-the-Art and Glossary

This article explores a critical gap in Mixed Reality (MR) technology: while advances have been made, MR still struggles to authentically replicate human embodiment and socio-motor interaction. For MR to enable truly meaningful social experiences, it needs to incorporate multi-modal data streams and multi-agent interaction capabilities. To address this challenge, we present a comprehensive glossary covering key topics such as Virtual Characters and Autonomisation, Responsible AI, Ethics by Design, and the Scientific Challenges of Social MR within Neuroscience, Embodiment, and Technology. Our aim is to drive the transformative evolution of MR technologies that prioritize human-centric innovation, fostering richer digital connections. We advocate for MR systems that enhance social interaction and collaboration between humans and virtual autonomous agents, ensuring inclusivity, ethical design and psychological safety in the process.

cs.HC↗
arXiv2025

Online Phase Estimation of Human Oscillatory Motions using Deep Learning

Accurately estimating the phase of oscillatory systems is essential for analyzing cyclic activities such as repetitive gestures in human motion. In this work we introduce a learning-based approach for online phase estimation in three-dimensional motion trajectories, using a Long Short- Term Memory (LSTM) network. A calibration procedure is applied to standardize trajectory position and orientation, ensuring invariance to spatial variations. The proposed model is evaluated on motion capture data and further tested in a dynamical system, where the estimated phase is used as input to a reinforcement learning (RL)-based control to assess its impact on the synchronization of a network of Kuramoto oscillators.

eess.SY↗
arXiv2025

Hierarchical Policy-Gradient Reinforcement Learning for Multi-Agent Shepherding Control of Non-Cohesive Targets

We propose a decentralized reinforcement learning solution for multi-agent shepherding of non-cohesive targets using policy-gradient methods. Our architecture integrates target-selection with target-driving through Proximal Policy Optimization, overcoming discrete-action constraints of previous Deep Q-Network approaches and enabling smoother agent trajectories. This model-free framework effectively solves the shepherding problem without prior dynamics knowledge. Experiments demonstrate our method's effectiveness and scalability with increased target numbers and limited sensing capabilities.

cs.LG↗
arXiv2024

Emergent Cooperative Strategies for Multi-Agent Shepherding via Reinforcement Learning

We present a decentralized reinforcement learning (RL) approach to address the multi-agent shepherding control problem, departing from the conventional assumption of cohesive target groups. Our two-layer control architecture consists of a low-level controller that guides each herder to contain a specific target within a goal region, while a high-level layer dynamically selects from multiple targets the one an herder should aim at corralling and containing. Cooperation emerges naturally, as herders autonomously choose distinct targets to expedite task completion. We further extend this approach to large-scale systems, where each herder applies a shared policy, trained with few agents, while managing a fixed subset of agents.

eess.SY↗
arXiv2024

Learning-based cognitive architecture for enhancing coordination in human groups

As interactions with autonomous agents-ranging from robots in physical settings to avatars in virtual and augmented realities-become more prevalent, developing advanced cognitive architectures is critical for enhancing the dynamics of human-avatar groups. This paper presents a reinforcement-learning-based cognitive architecture, trained via a sim-to-real approach, designed to improve synchronization in periodic motor tasks, crucial for applications in group rehabilitation and sports training. Extensive numerical validation consistently demonstrates improvements in synchronization. Theoretical derivations and numerical investigations are complemented by preliminary experiments with real participants, showing that our avatars can integrate seamlessly into human groups, often being indistinguishable from humans.

eess.SY↗
arXiv2024

Guaranteeing Control Requirements via Reward Shaping in Reinforcement Learning

In addressing control problems such as regulation and tracking through reinforcement learning, it is often required to guarantee that the acquired policy meets essential performance and stability criteria such as a desired settling time and steady-state error prior to deployment. Motivated by this necessity, we present a set of results and a systematic reward shaping procedure that (i) ensures the optimal policy generates trajectories that align with specified control requirements and (ii) allows to assess whether any given policy satisfies them. We validate our approach through comprehensive numerical experiments conducted in two representative environments from OpenAI Gym: the Inverted Pendulum swing-up problem and the Lunar Lander. Utilizing both tabular and deep reinforcement learning methods, our experiments consistently affirm the efficacy of our proposed framework, highlighting its effectiveness in ensuring policy adherence to the prescribed control requirements.

eess.SY↗
arXiv2023

In vivo learning-based control of microbial populations density in bioreactors

A key problem toward the use of microorganisms as bio-factories is reaching and maintaining cellular communities at a desired density and composition so that they can efficiently convert their biomass into useful compounds. Promising technological platforms for the real time, scalable control of cellular density are bioreactors. In this work, we developed a learning-based strategy to expand the toolbox of available control algorithms capable of regulating the density of a \textit{single} bacterial population in bioreactors. Specifically, we used a sim-to-real paradigm, where a simple mathematical model, calibrated using a few data, was adopted to generate synthetic data for the training of the controller. The resulting policy was then exhaustively tested in vivo using a low-cost bioreactor known as Chi.Bio, assessing performance and robustness. In addition, we compared the performance with more traditional controllers (namely, a PI and an MPC), confirming that the learning-based controller exhibits similar performance in vivo. Our work showcases the viability of learning-based strategies for the control of cellular density in bioreactors, making a step forward toward their use for the control of the composition of microbial consortia.

eess.SY↗
arXiv2022

CT-DQN: Control-Tutored Deep Reinforcement Learning

One of the major challenges in Deep Reinforcement Learning for control is the need for extensive training to learn the policy. Motivated by this, we present the design of the Control-Tutored Deep Q-Networks (CT-DQN) algorithm, a Deep Reinforcement Learning algorithm that leverages a control tutor, i.e., an exogenous control law, to reduce learning time. The tutor can be designed using an approximate model of the system, without any assumption about the knowledge of the system's dynamics. There is no expectation that it will be able to achieve the control objective if used stand-alone. During learning, the tutor occasionally suggests an action, thus partially guiding exploration. We validate our approach on three scenarios from OpenAI Gym: the inverted pendulum, lunar lander, and car racing. We demonstrate that CT-DQN is able to achieve better or equivalent data efficiency with respect to the classic function approximation solutions.

cs.LG↗
arXiv2022

External control of a genetic toggle switch via Reinforcement Learning

We investigate the problem of using a learning-based strategy to stabilize a synthetic toggle switch via an external control approach. To overcome the data efficiency problem that would render the algorithm unfeasible for practical use in synthetic biology, we adopt a sim-to-real paradigm where the policy is learnt via training on a simplified model of the toggle switch and it is then subsequently exploited to control a more realistic model of the switch parameterized from in-vivo experiments. Our in-silico experiments confirm the viability of the approach suggesting its potential use for in-vivo control implementations.

eess.SY↗
arXiv2021

Intermittent non-pharmaceutical strategies to mitigate the COVID-19 epidemic in a network model of Italy via constrained optimization

This paper is concerned with the design of intermittent non-pharmaceutical strategies to mitigate the spread of the COVID-19 epidemic exploiting network epidemiological models. Specifically, by studying a variational equation for the dynamics of the infected in a network model of the epidemic spread, we derive, using contractivity arguments, a condition that can be used to guarantee that, in epidemiological terms, the effective reproduction number is less than unity. This condition has three advantages: (i) it is easily computable; (ii) it is directly related to the model parameters; (iii) it can be used to enforce a scalability condition that prohibits the amplification of disturbances within the network system. We then include satisfaction of such a condition as a constraint in a Model Predictive Control problem so as to mitigate (or suppress) the spread of the epidemic while minimizing the economic impact of the interventions. A data-driven model of Italy as a network of three macro-regions (North, Center, and South), whose parameters are identified from real data, is used to illustrate and evaluate the effectiveness of the proposed control strategy.

q-bio.PE↗
arXiv2020

Tutoring Reinforcement Learning via Feedback Control

We introduce a control-tutored reinforcement learning (CTRL) algorithm. The idea is to enhance tabular learning algorithms by means of a control strategy with limited knowledge of the system model. By tutoring the learning process, the learning rate can be substantially reduced. We use the classical problem of stabilizing an inverted pendulum as a benchmark to numerically illustrate the advantages and disadvantages of the approach.

math.OC↗
arXiv2020

Intermittent yet coordinated regional strategies can alleviate the COVID-19 epidemic: a network model of the Italian case

The COVID-19 epidemic that emerged in Wuhan China at the end of 2019 hit Italy particularly hard, yielding the implementation of strict national lockdown rules (Phase 1). There is now a hot ongoing debate in Italy and abroad on what the best strategy is to restart a country to exit a national lockdown (Phase 2). Previous studies have focused on modelling possible restarting scenarios at the national level, overlooking the fact that Italy, as other nations around the world, is divided in administrative regions who can independently oversee their own share of the Italian National Health Service. In this study, we show that regionalism, and heterogeneity between regions, is essential to understand the spread of the epidemic and, more importantly, to design effective post Lock-Down strategies to control the disease. To achieve this, we model Italy as a network of regions and parameterize the model of each region on real data spanning almost two months from the initial outbreak. Using the model, we confirm the effectiveness at the regional level of the national lockdown strategy implemented so far by the Italian government to mitigate the spread of the disease and show its efficacy at the regional level. We also propose that differentiated, albeit coordinated, regional interventions can be effective in Phase 2 to restart the country and avoid future recurrence of the epidemic, while avoiding saturation of the regional health systems and mitigating impact on costs. Our study and methodology can be easily extended to other levels of granularity (provinces or counties in the same region or states in other federal countries, etc.) to support policy- and decision-makers.

physics.soc-ph↗
arXiv2019

Control-Tutored Reinforcement Learning

We introduce a control-tutored reinforcement learning (CTRL) algorithm. The idea is to enhance tabular learning algorithms so as to improve the exploration of the state-space, and substantially reduce learning times by leveraging some limited knowledge of the plant encoded into a tutoring model-based control strategy. We illustrate the benefits of our novel approach and its effectiveness by using the problem of controlling one or more agents to herd and contain within a goal region a set of target free-roving agents in the plane.

math.OC↗