The 61st IEEE Conference on Decision and Control will be held Tuesday through Friday, December 6-9, 2022 at the Marriott and JW Marriott Cancun Resort, Cancún, Mexico. The conference will be preceded by workshops on Monday, December 5, 2022, addressing current and future topics in control systems from experts from academia, research institutes, and industry. The workshops are expected to be delivered in person.
The workshops will be offered based on viable attendance. The 61st CDC reserves the right to cancel nonviable workshops.
Questions can be directed to the Workshop Chair, Professor Maria Letizia Corradini, email@example.com
Organizers: Jing Shuang (Lisa) Li, Jing Yu, Carmen Amo Alonso, John Doyle
Date and Time: Monday, December 5, 9:00 – 16:50
Abstract: System Level Synthesis (SLS) is a very new topic in control theory and is expanding rapidly in controls and controls-adjacent fields. It allows scalable, systematic design of optimal controllers for systems with sparsity, locality, delay, and other constraints on communication and computation internal to the controller – constraints which are omnipresent in large cyberphysical systems and biological systems. The application of SLS theory to many branches of control theory (e.g. model predictive control) results in unprecedented scalability and performance/robustness guarantees. In this workshop, we aim to make SLS theory more accessible to the CDC audience, and in doing so, allow audience members to leverage SLS theory and its frontiers for the benefit of their existing research. This is an unusual and exciting opportunity; for experts in both theory and applications, the frontier of SLS theory is broad and accessible. This full-day workshop will start with a tutorial aimed at the typical CDC audience; subsequent sessions featuring guest speakers will explore advanced topics, including model predictive control, nonlinear control, and learning for control.
Organizers: Silvia Mastellone, Roy Smith, Saverio Bolognani, Dennice Gayme, Hideaki Ishii, Afef Fekih
Date and Time: Monday, December 5, 8:00 – 17:00
Location: Maya Ballroom I
Abstract: This second workshop on Diversity and Inclusion (D&I), is dedicated to implement some of the diversity strategies proposed as outcomes of the first CSS D&I Workshop that took place during the CDC 2021. The focus of this workshop will be on unconscious bias, its impact on applying for jobs and recruiting and how to face it as member of a minority group and how to prevent it as leader or as member of majority group.
The objective is twofold: on one end we offer students and young professionals from minority groups some theory and practice for developing required nontechnical skills to thrive as control engineers in academia and industry and specifically overcome the effect of unconscious bias. The second objective is to provide young leaders from any group with the knowledge and tools to prevent the effect of unconscious bias when recruiting and managing members of a research team.
(i) Provide knowledge and training to acquire skills for starting and developing a successful career in R&D, with specific focus on minority groups: interviewing, negotiating, networking, understanding group dynamics at work.
(ii) Provide knowledge and tools to overcome the effects of unconscious bias when recruiting or managing an R&D group.
(i) Students, postdocs and young researchers from minority groups preparing to face the job market and interested in acquiring skills for their future career in industry or academia.
(ii) Young professionals from academia and industry interested in acquiring skills to shape and sustainably manage a diverse and successful research group.
(iii) Members of the CSS from academia and industry interested in learning about unconscious bias and its effect on working groups and interested in fostering a creative and inclusive culture in their group, department, or institutions.
Organizers: Kunal Garg, Ricardo Sanfelice, Alvaro Cardenas
Date and Time: Monday, December 5, 8:30 – 16:30
Location: Maya Ballroom VIII
Abstract: Recent events of cyber-attacks such as Ukraine power grid hack, Iranian nuclear plants (Stuxnet), German steel mills incident, around the world have demonstrated that cyber-attacks are inevitable. More advanced attacks such as transduction attacks have led to increased risk as more and more devices and systems have become vulnerable to such external threats. Most common cyber tools used for CPS security focus on attack detection and prevention, utilizing tools such as encryption, privacy-preserving control, redundancy (in communication links, control blocks or sensors). Attack detection is a crucial aspect for security, and there has been a lot of development in development of effective attack-detection schemes, including comparing the expected behavior of the system with its actual behavior to flag an attack. On the other hand, a pure control-theoretic framework focuses on attack recovery for CPS security and takes a fault-tolerant control design approach or a robust control approach. A cyber-control-theory approach focuses on attack detection, mitigation as well as recovery of a CPS after attack and preserving crucial properties such as safety and preventing the system from failures. Some example of such tools include actuator constraining to limit how much an attacker can manipulate the system, using physics-based virtual sensors to assist feedback design under sensor-attacks, conservative, safe controller to be used at all times so that an attack is ineffective or a back-up controllers to be used when an attack is detected. However, much work still needs to be done in the field of CPS security with provable guarantees. In particular, providing guarantees on attack-mitigation and recovery is largely an open problem. Some recent developments on the matter include privacy preserving control and using machine learning-based techniques to detect and respond to adversarial attacks. Moreover, the paradigm of internet-of-things and internet-of-everything has led to an unprecedented increment in attack-surfaces, and new attack use-cases or scenarios might emerge that are currently unknown. The main goal of the workshop is to highlight recent advances and developments in the role of control theory in solving security problems of cyber-physical systems (CPS) and discuss some of the important open problems in CPS security. This workshop aims to bring together experts from cyber-security and control theory to discuss how sensors, actuators, or communications links of CPS can be attacked, and how control-theoretic tools can help prevent, minimize, and enable recovery from such attacks.
Organizers: Andreas A. Malikopoulos
Date and Time: Monday, December 5, 8:00 – 17:30
Location: Maya Ballroom V (lectures) and Maya Ballroom VI (poster session/demos)
Abstract: Cyber-physical systems (CPS), in many instances, represent systems of systems with an informationally decentralized structure such as emerging mobility systems, networked control systems, mobility markets, smart power grids, power systems, social media platforms, and cooperation of robots. In such systems, there is a large volume of data with a dynamic nature which is added to the system gradually in real time and not altogether in advance. As the volume of data increases, the domain of the control strategies also increases, and thus it becomes challenging to search for an optimal strategy. Even if an optimal strategy is found, implementing such strategies with increasing domains is burdensome. Model-based control approaches cannot effectively facilitate optimal solutions with performance guarantees due to the discrepancy between the model and the actual CPS. On the other hand, traditional supervised learning approaches cannot always facilitate robust solutions using data derived offline. By contrast, applying reinforcement learning approaches directly to the actual CPS might impose significant implications on safety and robust operation of the system. The goal of this workshop is to stimulate a discussion on how we can combine offline model- based control with online learning approaches, and thus circumvent the challenges in deriving optimal strategies for CPS with particular emphasis on CPS applications related to emerging mobility systems.
Organizers: Dragan Nesic, Mathieu Granzotto, Romain Postoyan
Date and Time: Monday, December 5, 8:00 – 17:00
Location: Mexico and Cozumel
Abstract: Learning (i.e., estimation) and optimization have a long history in control systems theory, forming some of its pillars. Recent cross-fertilizations between the artificial intelligence and control systems communities have seen a renewed interest in data-driven and optimization-based control, such as reinforcement learning, and opened up numerous new research opportunities. This one-day workshop will highlight some of the most recent developments by world experts and will foster interactions between these closely related exciting research areas.
Organizers: Anastasios Tsiamis, Ingvar Ziemann, Nikolai Matni, George Pappas
Date and Time: Monday, December 5, 8:30 – 17:30
Location: Maya Ballroom IV
Abstract: Machine learning methods are at an ever increasing pace being integrated into domains that have classically been within the purview of controls. There is a wide range of examples, including perception-based control, agile robotics, and autonomous driving and racing. As exciting as these developments may be, they have been most pronounced on the experimental and empirical sides. To deploy these systems safely, stably, and robustly into the real world, we argue that a principled and integrated theoretical understanding of a) fundamental limitations and b) statistical optimality is needed. With this workshop we seek to bring together theoretical research across machine learning and control theory to address these issues in tasks like system identification, learning-based control and estimation. This workshop will serve as a good reference for a wider audience as to what the current state of the art in this intersection is. We also aim to make existing technical tools in this area more accessible to an audience with a control theoretic background, as they require mathematical tools not typically included in a control theorist’s training (e.g., high-dimensional statistics and learning theory).
The objective is to grow and foster the new interdisciplinary community around machine learning and control, with an emphasis on how mathematical methods from learning theory interact with classical control theoretic notions. To do so, we have invited speakers from both control and machine learning backgrounds. Talks will emphasize the interplay between control-theoretic notions, such as controllability, safety, stability, robustness, and learning theoretic notions, such as sample complexity, regret, and excess risk. We will revisit recent results for linear systems but also present recent advances in the case of more general nonlinear systems.
Organizers: Pablo Borja, Cosimo Della Santina
Date and Time: Monday, December 5, 9:00 – 16:00
Location: Maya Ballroom II
Abstract: In recent years, we have witnessed how emerging technologies have expanded the possibilities in designing and manufacturing complex mechanical systems, e.g., smart materials sensors and actuators, nanomechanical devices, and soft robots. These new systems offer solutions to problems in different essential areas, including, but not limited to, medical applications, robotics, aerospace applications, and environmental monitoring.
However, the complexity of such systems has introduced new challenges for the control and systems community, where the necessity of suitable modeling approaches and efficient control methods is becoming essential.
In this regard, energy-based methods for modeling and control are powerful tools since, despite their complexity, the behavior of the mentioned systems is ruled by physical quantities such as energy and dissipation.
This workshop aims to disseminate state-of-the-art results in the modeling and control of mechanical systems using energy approaches. To this end, the proposed program consists of 7 talks, which cover a broad range of topics including, but not limited to, modeling and control of flexible structures; tuning techniques and performance analysis of energy-based controllers; the synergy between energy-based control methodologies and neural networks; fluidic actuation in mechanical systems; and the use of energy-based methods in robotic surgery.
Organizers: Mauro Salazar, Raphael Stern
Date and Time: Monday, December 5, 9:00 – 17:30
Location: Maya Ballroom III
Abstract: Public debate about the future of mobility and transportation is increasingly informed by predictions about the impact of Connected, Autonomous and Electrified Vehicles (CAEVs). As CAEVs are approaching market-readiness and are beginning to be deployed, there are still several socio-technical challenges to be addressed. Many of those are being identified as research topics in control systems, for instance:
• How can we design profitable and sustainable mobility systems that leverage CAEVs?
• What are the socio-technical challenges we are facing and how can control engineers contribute to address them?
• How should we adapt our infrastructure to adequately accommodate future autonomous, connected and electrified mobility systems, whilst leveraging existing transportation modes?
• How can we ensure that such technologies benefit all members of society, improving equity and fairness rather than undermining them?
• How can we ensure that such technologies are leveraged to help meet sustainability goals?
This workshop will gather experts from control systems, transportation, mechanical, electrical, computer and civil engineering, robotics, and social science in order to:
1. identify challenges and opportunities in control systems for the future of transportation that are triggered by the advent of CAEVs,
2. identify modeling and control methodologies to address them,
3. share insights from early deployments and turn such insights into an actionable research roadmap.
From a technical perspective, we will discuss socio-technical and control problems from the individual vehicle-level up to the transportation-system-level, ranging from human-robot interaction to incentive and tolling schemes, and from network control problems to the combination of new (and old!) technologies within sustainable and human-centered mobility systems. Overall, the goal of the workshop is to provide researchers in control with an overview of current and future mobility challenges to be addressed by our community, and, at the same time, exchange new research results, ideas and visions on how our control community can contribute to sustainable and human-centered mobility systems and, ultimately, our society.
Organizers: Thomas Beckers, Janine Matschek, Sandra Hirche, Rolf Findeisen
Date and Time: Monday, December 5, 8:15 – 18:00
Location: Maya Ballroom VII
Abstract: One challenge in controller design is to achieve the desired performance and guarantee safe operation, e.g., via the satisfaction of constraints despite the presence of disturbances. One way to deal with uncertainties is obtaining an estimate via machine learning techniques, such as Gaussian processes. Gaussian processes have been used increasingly as a data-driven technique within the past two decades due to many beneficial properties such as the bias-variance trade-off and the close relation to Bayesian mathematics.
In contrast to most of the methods, Gaussian process models provide a regression function and a measure for the uncertainty of the prediction. This powerful property makes them very attractive for many applications in control, e.g., model predictive control, robust control, reinforcement learning, and general optimization tasks, as the uncertainty measure provides convergence, performance, and safety guarantees. However, fusing/embedding machine learning, especially Gaussian processes in a closed-loop control system, poses several challenges, such as closed-loop uncertainty propagation or real-time feasible online learning.
This tutorial-style workshop aims to provide insight into the fundamentals behind Gaussian processes for modeling and control and sketches some of the open challenges and opportunities using Gaussian processes for modeling and control. Experts/lecturers with experience in Gaussian processes and (optimization-based) control from academia and industry will introduce Gaussian processes’ basics and spotlight Gaussian processes’ opportunities for the control community and recent advances in learning-based control under uncertainties in general. The workshop targets an audience from graduate level to experienced theoretical and practically oriented control engineers who aim to improve their knowledge in controller design under uncertainties leveraging Gaussian processes and machine learning.