From swarms of robots to crowds of people, from biological populations to networks of autonomous vehicles, the ability to shape their emergent collective behavior has become one of the defining control challenges of our time. Yet classical control, designed one agent at a time, simply does not scale. In this talk, I will argue that meeting this challenge requires a fundamental shift in perspective: rather than controlling agents, we must learn to control distributions, densities, and emergent patterns. I will introduce the multi-scale control paradigm, a framework that moves seamlessly between the microscopic world of individual agents and the macroscopic world of the population they form. I will discuss two powerful ideas: continuification, which lifts agent-level dynamics into partial differential equations amenable to elegant macroscopic control design and convergence analysis; and robust density control, which makes these strategies work in the messy reality of heterogeneous agents, uncertainty, and disturbances. Together, they open the door to scalable, decentralized strategies that remain provably effective when deployed at scale. I will showcase the approach through some relevant engineering applications from coverage control to swarm density regulation and close with a glimpse of an exciting complementary frontier: indirectly steering collective behavior through leaders, herders, and shepherds.

Mario di Bernardo is Professor of Automatic Control at the University of Naples Federico II, Italy and Visiting Professor of Nonlinear Systems and Control at the University of Bristol, U.K. He currently serves as Deputy pro-Vice Chancellor for Internationalization at the University of Naples and coordinates the research area and PhD program on Modeling and Engineering Risk and Complexity of the Scuola Superiore Meridionale, the new School of Advanced Studies located in Naples. On 28th February 2007 he was bestowed the title of Cavaliere of the Order of Merit of the Italian Republic for scientific merits from the President of Italy. His research interests include the analysis, synchronization and control of complex network systems; piecewise-smooth dynamical systems; nonlinear dynamics and nonlinear control with applications to engineering and computational biology. He authored or co-authored more than 220 international scientific publications including more than 110 papers in scientific journals, a research monograph and two edited books.

Network dynamics often take place on large-scale networks that are only partially known, because of privacy constraints or because the information is too costly to collect. Even when the network structure is fully available, designing effective interventions becomes increasingly challenging as the network size grows. In this seminar, we present a framework to approximate the behavior of network dynamics on large random networks by low-dimensional discrete-time systems. Our approach describes aggregate quantities of interest—such as the average opinion in opinion dynamics or the fraction of adopters in diffusion processes—while relying only on limited statistical information about the network. Building on these approximation results, we develop intervention strategies that scale efficiently with network size and do not require complete knowledge of the network structure. 

Leonardo Cianfanelli received the B.Sc. In Physics and Astrophysics in 2014 from Università di Firenze, Italy, the M.S. in Physics of Complex Systems in 2017 from Università di Torino, Italy, and the PhD in Pure and Applied Mathematics in 2022 from Politecnico di Torino, Italy. He was Visiting Student at the Laboratory for Information and Decision Systems, Massachusetts Institute of Technology in 2018–2020, and Research Assistant at Politecnico di Torino in 2021-2025. He is currently Assistant Professor at the Department of Mathematical Sciences, Politecnico di Torino, Italy. His research focuses on control in network systems, with application to transportation, epidemics and socio-technical systems.

Are you writing a research paper and wondering where to submit? Do you have a paper currently under consideration and wondering what is going on behind the scenes? Philip Coatsworth, an Associate Editor at the journal Communications Engineering, will demystify the publication process and give insider tips on how to enable a smooth and successful publication journey, from writing through to acceptance.

Dr Philip Coatsworth is an Associate Editor at Communications Engineering, a selective open access journal in the Nature Portfolio covering all areas of engineering. He joined the journal in February 2025 and previously obtained a PhD in Bioengineering from Imperial College London, where he conducted research on using electrochemical sensors to monitor living plant roots.

Collective decision-making in social systems is increasingly shaped by complex interaction mechanisms, including higher-order social influences and algorithmic personal recommendations. This talk presents two recent contributions that investigate how such mechanisms affect the emergence of collective decisions. First, we study a nonlinear opinion dynamics model with cooperative higher-order interactions represented by a hypernetwork. Saturated influence functions govern the dynamics, and a parameter captures the community’s social effort. We show that higher-order interactions unfold a pitchfork bifurcation, creating bistability and potentially disrupting consensus. We then consider social networks in which algorithms influence individuals through personal recommendations. To capture confirmation bias, we introduce an analytically tractable model that describes how agents process personal recommendations, based on their prior beliefs and sensitivity to external information. We analyze both broadcast and personalized recommendation policies and show that the former can also induce multiple stable equilibria, potentially leading to unintended collective outcomes. Together, these results illustrate how nonlinear social influence mechanisms can shape collective behavior in complex social systems.

Angela Fontan is an Assistant Professor with the Department of Decision and Control Systems, School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Sweden. She is also affiliated with Digital Futures, and she is a WASP Fellow, funded by the Knut and Alice Wallenberg Foundation. She received a B.Sc. degree in Information Engineering in 2013 and a M.Sc. degree (with honor) in Automation Engineering in 2016, from the University of Padova, Italy. She received a Ph.D. degree in Electrical Engineering with specialization in Automatic Control in September 2021, from the Division of Automatic Control, Department of Electrical Engineering, Linköping University, Sweden. From 2021 to 2024, she was a Postdoctoral researcher at the Department of Decision and Control Systems, KTH Royal Institute of Technology. Her research interests are in the area of networked cyber-physical-human systems and nonlinear dynamics over networks, with applications to social networks, and autonomous and collective decision-making.

A key aspect of Nash equilibrium seeking is that the agents have access to the other participants’ decision variables. However, this assumption clashes with real life applications since it implicitly relies on the fact that the agents are willing to share private information and at the same time trust what the others communicate. In this talk, we consider instead what happens when the communication is affected by the agents behavior and trust level. We will show that an equilibrium can be reached even if the agents are stubborn and with different leaning mechanisms, as long as some communication is allowed between (some of) them. This provides a step toward understanding strategic behavior of the agents in a more realistic setting.

Barbara Franci is an Assistant Professor with the Department of Mathematical Sciences at Politecnico di Torino, Italy. She received her Bachelor’s and Master’s degrees in Pure Mathematics from University of Siena, Siena, Italy, respectively in 2012 and 2014. Then, she received her PhD from Politecnico di Torino and Universitá di Torino (joint program), Italy, in 2018. In September-December 2016 she visited the Department of Mechanical Engineering, University of California, Santa Barbara, USA. After the PhD, she was a PostDoc at the Delft Center for Systems and Control, Delft University of Technology, Delft, The Netherlands, until 2021. From 2021 till 2025, she was Assistant Professor with the Department of Advanced Computing Sciences at Maastricht University, The Netherlands. She was awarded in 2017 with the Quality Award by the Academic Board of Politecnico di Torino. Her research interests are on game theory and its applications.

Scalable multi-robot systems promise robust collective performance, but increasing the number of robots does not simply make a system better. Across models of scalability in computing, robotics, and networked systems, performance depends on a balance between cooperation, interference, information sharing, and shared-resource depletion; adding more agents can produce superlinear gains, sublinear returns, or collapse. Recent evidence suggests that large-scale multi-robot systems can exhibit a sharp two-phase regime near critical system sizes, where performance separates into near-optimal and near-failed outcomes, raising the possibility that apparently successful robot swarms may operate on the edge of delayed breakdown. This talk synthesizes work on scalability modeling, critical system-size effects, and task specialization in multi-robot systems. I will discuss how minimal models can help describe transitions between productive scaling, transient overload, and collapse, and how task specialization may shift these transitions by reorganizing collective work. Recent evolutionary robotics results show that specialization has an evaluation cost, because multiple specialist behaviors must be optimized separately, but that this cost becomes easier to justify as swarm size increases. Together, these results suggest that the future of scalability in robot swarms may depend as much on scaling down participation, interference, and behavioral complexity as on carefully scaling up the number of robots.

Since 2022, Heiko Hamann has been Professor of Cyber-physical Systems at the University of Konstanz, Germany, and a member of the Centre for the Advanced Study of Collective Behaviour. He has worked on swarm robotics and collective behavior for more than 20 years, with research interests spanning swarm robotics, bio-hybrid systems, evolutionary robotics, and the modeling of complex systems. His work often connects robotics with biology and the behavioral sciences, including collaborations with ethologists, plant biologists, architects, and psychologists. He is the author of Swarm Robotics: A Formal Approach (2nd edition, 2026) and has served as editor-in-chief of the journal Swarm Intelligence since 2023.

In this talk, we explore the role of monotonicity in enabling modular control design for the stabilization and multi-stabilization of networked nonlinear systems. We begin by showing that the variational systems of monotone systems can be embedded into positive systems. Building on this embedding, we address a network stabilization problem by enforcing monotonicity and exponential dissipativity of the individual network components. We then extend this approach to a network multi-stabilization problem. In both cases, the overall problem decomposes into the analysis of each network component and the in- or out-degree of each interconnection.

Yu Kawano received his M.S. and Ph.D. degrees in Engineering from Osaka University, Japan, in 2011 and 2013, respectively. He was a Postdoctoral Researcher at Kyoto University, Japan, from 2013 to 2016, and at the University of Groningen, the Netherlands, from 2016 to 2019. He was an Associate Professor from 2019 to 2026 and has been a Full Professor since 2026 at the Graduate School of Advanced Science and Engineering, Hiroshima University. He has also held visiting research positions at the Tallinn University of Technology (Estonia), the University of Groningen (the Netherlands), the University of Pavia (Italy), the Indian Institute of Technology Bombay (India), and the University of California, Santa Barbara (USA). His research interests include nonlinear systems, complex networks, model reduction, and privacy in control systems. He serves as an Associate Editor for Systems & Control Letters, IEEE Transactions on Systems, Man, and Cybernetics: Systems, the IEEE CSS Conference Editorial Board, and the EUCA Conference Editorial Board.

The Entangled Brain framework promotes the idea that we need to understand the brain as a complex, entangled system. Why does a complex systems perspective, one that entails emergent properties, matter for brain science? We discuss principles of brain organization that inform the question of the interactional complexity of the brain: (1) massive combinatorial anatomical connectivity; (2) highly distributed functional coordination; (3) non-linear, reentrant processing; and (4) networks/circuits as functional units. To motivate the challenges of mapping structure and function, we discuss neural circuits illustrating the high anatomical and functional interactional complexity typical in the brain. We discuss implications for brain science, including the need to characterize decentralized and heterarchical anatomical–functional organization. Finally, we discuss implications for artificial intelligence agents and robotics, and argue that approaches that integrate domains typically treated separately (perception, action, cognition, emotion, motivation) are necessary to develop systems that can handle open-ended environments, and to attain autonomy in complex, time-varying conditions.

Luiz Pessoa received a bachelor’s degree in Computer Science and a masters in Computer Engineering from the Federal University of Rio de Janeiro, Brazil. He obtained a PhD in computational neuroscience at Boston University, USA. He then returned to Brazil for a few years where he was a professor of Computer of Systems Engineering at the Federal University of Rio de Janeiro. Subsequently, he took a U-turn and received further postdoctoral training at the National Institute of Mental Health, USA. Prior to his current position, he held faculty positions at Brown University and Indiana University, Bloomington. Since 2011, he has been at the Department of Psychology, University of Maryland, College Park, where he is full Professor and Director of the Maryland Neuroimaging Center. His research interests center around the interactions between emotion/motivation and perception/cognition. His most recent book is “The entangled brain: How perception, cognition, and emotion are woven together” by MIT Press (2022) and is aimed at a general audience interested in science.

Scaling laws are ubiquitous in mechanics, from material strength to turbulence. These laws describe system behavior via power-laws, connecting specific properties to size. While foundational in physics, recent studies have identified scaling laws in living social systems that currently lack rigorous mechanistic understanding. Toward a methodology for unveiling the underpinnings of these complex systems, we tackle two distinct problems. First, we examine the scaling of metabolic rate in insect colonies, measured in the laboratory. Grounded in a “reverse social contagion” hypothesis, we establish an experimentally validated compartmental model for colony energy savings. Second, we analyze firearm prevalence across U.S. cities—a much less structured scenario where only ecological data are available. Using multidimensional data, we demonstrate the possibility of informing plausible modeling hypotheses through causal discovery and, consequently, formulating network-theoretic models.

Maurizio Porfiri received the M.Sc. and Ph.D. degrees in engineering mechanics from Virginia Tech, Blacksburg, VA, USA, and the dual Ph.D. degrees in theoretical and applied mechanics from the Sapienza University of Rome, Rome, Italy, and the University of Toulon, La Garde, France, in 2005. He is currently an Institute Professor with the Tandon School of Engineering, New York University, New York, NY, USA, with appointments in mechanical, aerospace and biomedical engineering. He is also the Director of the Center for Urban Science and Progress. He founded the Dynamical Systems Laboratory in 2006, leading research in theory and applications of complex systems. He has authored more than 400 journal publications with more than 20 000 citations. Dr. Porfiri was the recipient of the NSF CAREER Award and recognition in Popular Science’s “Brilliant 10.” He is a Fellow of ASME.

Many networked systems exhibit collective behaviors shaped by a small number of agents whose states, actions, or signals persist over time and leave a detectable footprint on the rest of the system. These agents may correspond to stubborn individuals in opinion dynamics, bots or coordinated accounts in online platforms, informed agents in robotic swarms, or anchors and stable signal sources in distributed sensing and localization problems. This talk addresses the problem of inferring such persistent agents from partial observations of collective dynamics at equilibrium, without requiring prior knowledge of the underlying interaction network. The motivating model is a DeGroot-type influence system in which regular agents update their states through local interactions, while a subset of agents maintains persistent behavior. At equilibrium, the states of regular agents can be expressed as combinations of the persistent agents’ states, with weights encoding their influence on the rest of the network. From noisy equilibrium observations collected across multiple discussions, trials, or operating conditions, the identification task is formulated as a low-rank approximation problem aimed at recovering both the persistent agents and their influence matrix. The talk will introduce an efficient computational approach based on interpolative decomposition, together with theoretical guarantees for exact recovery and robustness to noisy equilibrium observations. Although the formulation is rooted in opinion dynamics, the underlying perspective is broader. Persistent agents can be interpreted as behavioral sources, anomalous or coordinated actors, informed leaders, or stable anchors depending on the application domain.

Chiara Ravazzi (Senior Member, IEEE) is a Senior Researcher at the Italian National Research Council (CNR-IEIIT) and adjunct professor at the Politecnico di Torino. She obtained the Ph.D. in Mathematical Engineering from Politecnico di Torino in 2011. In 2010, she spent a semester as a visiting scholar at the Massachusetts Institute of Technology (LIDS), and from 2011 to 2016, she worked as a post-doctoral researcher at Politecnico di Torino (DISMA, DET). She joined the Institute of Electronics and Information Engineering and Telecommunications (IEIIT) of the National Research Council (CNR) in the role of a Tenured Researcher (2016-2022). Furthermore, she served as an Associate Editor for IEEE Transactions on Signal Processing from 2019 to 2023, and she currently holds the same position for IEEE Transactions on Control Systems Letters (since 2021) and the European Journal of Control (since 2023). She has achieved the national scientific qualification as a full professor in the field of Automatica (09/G1).

Complex social systems are characterized by a deep intertwining between opinion formation and decision-making processes. However,  these two processes have been typically studied as separate problems in the literature, limiting the possibility to capture and study the complexity of real-world social systems. This talk aims at bridging this gap by presenting a novel game-theoretic co-evolutionary model for actions and opinions. After briefly presenting the wide range of emergent behaviors that our model can predict, I will show how this model can be used to design effective intervention policies to unlock social change.

Lorenzo Zino is an Assistant Professor with the Department of Electronics and Telecommunications, Politecnico di Torino, Italy, since 2022. He received the BS, MS, and PhD in Applied Mathematics from Politecnico di Torino. He held Research Fellowships at Politecnico di Torino (Italy), the University of Groningen (The Netherlands), and New York University (US), and visiting positions at Lund University (Sweden), Curtin University (Australia), Adelaide University (Australia), and Hiroshima University (Japan). His research interests include modeling, analysis, and control of dynamics over networks, applied probability, and game theory. He has co-authored more than 100 international scientific publications, including 60 journal papers. He is Senior Member of the IEEE and the recipient of the 2024 Best Young Author Journal Paper Award from the IEEE CSS Italy Chapter. He is member of the Editorial Board of Scientific Reports, Associate Editor of the Journal of the International Journal of Control and IEEE Control Systems Letters, and member of the CEB for IEEE CSS and EUCA