ACC 2012 will feature a number of workshops on emerging and important topics in control systems.
Both half-day (afternoon) and full-day workshops will be offered on Tuesday, June 26. Please note that workshops are (a) subject to
cancellation for lack of registrants, and (b) subject to capacity limits. Prospective attendees are therefore advised
to register for their intended workshops early. See the
Registration page to sign up!
Descriptions of the workshops follow below the list of workshops.
Half-Day Workshops (1pm - 5pm)
Control of Power Inverters for Renewable Energy and Distributed Generation [Power Inverters]
Organizer: Qing-Chang Zhong
Controlling Software Execution: An Emerging Application Area for Control Engineering [Software Execution]
Organizers: Yin Wang, Stéphane Lafortune, and Spyros Reveliotis
Modeling, Control, and Optimization for Aerospace Systems [Aerospace Systems]
Organizers: Hugh Liu, Xavier Louis, Luis Rodrigues, Behzad Samadi, Serge Tremblay
Real-time Optimization of Nonlinear Dynamical Systems with Extremum-Seeking Control [Real-time Optimization]
Organizers: Martin Guay and Denis Dochain
Full-Day Workshops (8:30am - 5pm)
Analysis and Design of Cyber-Physical Transportation Systems: Challenges, Progress, and Future Directions [Transportation Systems]
Organizer: Domitilla Del Vecchio
Controlling Green Buildings: Challenges and Opportunities [Green Buildings]
Organizers: Francesco Borrelli, and Meli Stylianou
Health Management, Fault-tolerant Control, and Cooperative Control of Unmanned Aircraft [Unmanned Aircraft]
Organizers: Youmin Zhang, Camille Alain Rabbath, YangQuan Chen, Christopher Edwards, Cameron Fulford, Hugh H.-T. Liu, Liang Tang, Didier Theilliol, and Antonios Tsourdos
Nonlinear Regression Modeling: A Practical Guide [Nonlinear Regression]
Organizer: R. Russell Rhinehart
Quasilinear Control: Analysis and Design of Systems with Nonlinear Actuators and Sensors [Quasilinear Control]
Organizers: Semyon M. Meerkov, Pierre T. Kabamba, Yongsoon Eun, and ShiNung Ching
Robust Networked Infrastructures: Reliability and Security from Game Theoretic Perspectives [Networked Infrastructures]
Organizers: Saurabh Amin, Tamer Basar, Shankar Sastry, Galina Schwartz, Hamidou Tembine, and Jean Walrund
Smart Grid Markets: Integration of Renewables, Pricing, Modeling, and Optimization [Smart Grids]
Organizer: Anuradha Annaswamy
For further information on the ACC 2012 workshops please contact the conference
Control of Power Inverters for Renewable Energy and Distributed Generation
Organizer: Qing-Chang Zhong (University of Sheffield)
Location: Chaudiere Room
Energy and sustainability are now on the top agenda of many governments. Smart grids have become one of the major enablers to address energy and sustainability issues. The integration of renewable energy, distributed generation and hybrid electrical vehicles into smart grids, through power inverters, is one of the most important research areas for smart grids. Power electronic systems, at the heart of smart grids, have provided an exciting stage for control engineers. The attention of the workshop will be paid to fully appreciate the beauty of the integration of control and power electronics with applications in smart grids. Various problems around inverters, e.g. power quality issues, provision of a neutral line, grid-connection, synchronisation and parallel operation of inverters, will be addressed with innovative concepts such as synchronverters (inverters that mimic synchronous generators), C-inverters (inverters with capacitive output impedances), robust droop control, sinusoid-locked loops etc. It will help researchers who want to move into the area of smart grids establish a solid technical foundation for modeling, optimisation and control of smart grids. Most of the artful control strategies to be presented will be demonstrated with experimental results and, hence, the workshop will also help practitioners understand how advanced control strategies could improve system performance. The workshop also provides an excellent opportunity for researchers, PhD students and postdoctoral fellows who work in the area to get familiar with the latest developments.
Controlling Software Execution: An Emerging Application Area for Control Engineering (external website)
Organizers: Yin Wang (HP Labs), Stéphane Lafortune (University of Michigan), and Spyros Reveliotis (Georgia Tech)
Location: Matapedia Room
Computer and software engineering is a rich application area for control systems technology. Indeed, classical control theory, based on continuous-variable and time-driven models, has been applied to computer systems problems, such as throughput stabilization, and has achieved commercial success. Many important problems in computer and software engineering, however, are discrete in nature and naturally fall into the realm of discrete-variable and event-driven systems, i.e., Discrete Event Systems (DES). The growth of software defects has paralleled and shadowed the rapid advancement in computer technology. Relevant issues such as safety and consistency are inherently discrete-event problems. In the last few years, several applications of DES to computer systems problems have emerged. In particular, the paradigm of controlling software execution to avoid software defects at runtime has received significant attention in three communities: control engineering, programming languages, and operating systems. The objective of this workshop is to present this paradigm to a control audience, with relevant background on software issues and DES theory, using recent work on deadlock avoidance in multithreaded software by the organizers as the guiding case study.
More specifically, the goals of the workshop are the following:
- Introduce computer systems problems to the control community, and discuss possible applications of control theory in this domain.
- Educate the attendees about how to apply DES control technology to real computer systems problems, using the circular-mutex-wait deadlock avoidance project as an example.
- Discuss the relevant control theory, its benefits that are especially relevant in computer science applications, and the missing gaps that need to be addressed for successful application.
- Stimulate involvement of control researchers and students in this emerging area of research, both on the applications front and on the theoretical front.
Modeling, Control, and Optimization for Aerospace Systems
Organizers: Hugh Liu (University of Toronto), Xavier Louis (Thales), Luis Rodrigues (Concordia University), Behzad Samadi (Amirkabir University of Technology), Serge Tremblay (CAMAQ)
Speakers: Naira Hovakimyan (University of Illinois at Urbana-Champaign),
Kash Khorasani (Concordia University),
Hugh Liu (University of Toronto, Institute for Aerospace Studies),
Jamila Raouf (Concordia University),
Luis Rodrigues (Concordia University),
Behzad Samadi (MapleSoft)
Arkadiy Turevskiy (Mathworks),
Enric Xargay (University of Illinois at Urbana-Champaign).
Location: Hochelaga 2 Room
Montréal is a well known aerospace city in North America and the most important in Canada.
Aerospace systems and control is one of the two main themes of the American Control Conference and it is the topic of this workshop, which will try to bridge the gap between theory and practice of aerospace control systems. The workshop brings together organizers from both industry and academia and has a multidisciplinary program involving modeling, control and optimization for aerospace systems. Aerospace is one of the most important applications of control. It drove several control theoretical achievements from which optimal control, developed in the space race era of the sixties, is probably the most successful. The workshop will present cutting edge contributions at the theoretical level applied to a vast number of aerospace problems, such as convex optimization of aerospace systems, flutter active supression, autopilot design for energy efficiency and turbulence, inverse optimal autopilot design, modeling of actuator dynamics, fly-by-wireless and flight simulation. A flight simulation demonstration will be performed.
Real-time Optimization of Nonlinear Dynamical Systems with Extremum-Seeking Control
Organizers: Martin Guay (Queen's University, Canada) and Denis Dochain (Université Catholique de Louvain, Belgium)
Location: Hochelaga 3 Room
In this workshop, we provide an introduction to the problem of real-time optimization (RTO) of nonlinear dynamical systems. Real-time optimization (RTO) has become a leading technology for steady-state process optimization in the process industry. RTO is used as a supervisory control technique to compute, in real-time, optimal set-points to be tracked by the process operation.
The workshop first explores the problem of real-time optimization using perturbation-based extremum-seeking control techniques. The convergence properties of these techniques are explored to highlight some of its inherent limitations. The presentation focuses on some recent developments in the area of extremum-seeking control that have led to performance improvements beyond the standard requirements for time-scale separation and averaging.
One of the requirements of perturbation-based techniques is that the cost function of interest is measurable. In most cases, however, the measurement of the objective function may be limited to an inferential model that depends on some of the system's parameters. In such situation, an alternative to the standard extremum-seeking control must be considered. One of the objectives of the workshop is the formalization of the integrated design of RTO and control systems as a model-based adaptive extremum-seeking control (AESC) task. The main idea advocated is to integrate the competing tasks by using the objective function of the AESC system to formulate a Lyapunov function for the control system. The resulting integrated control system achieves the steady-state optimization objectives with guaranteed transient performance. AESC has proven to be an effective technology in a number of areas including bioprocess control, chemical reactor control, building systems control and fuel-cell control. We will provide a comprehensive introduction to leading solutions of the AESC problem. In doing so, we establish a number of results in the area of nonlinear adaptive control, constrained system control and periodic system control.
Analysis and Design of Cyber-Physical Transportation Systems: Challenges, Progress, and Future Directions (external website)
Organizer: Domitilla Del Vecchio (MIT)
Speakers: Domitilla Del Vecchio (MIT), Derek Caveney (Toyota), Umit Ozguner (Ohio State University), Matthias Althoff (CMU), Raj Rajkumar (CMU), Hasma Balakrishnan (MIT), Jonathan P. How (MIT), Emilio Frazzoli (MIT), Alexandre Bayen (UC, Berkeley)
Location: Richelieu Room
The decreasing costs of embedded computing and communication technologies are taking several engineering systems toward increased levels of autonomy. A remarkable example is that of transportation systems. The growing employment of cyber infrastructure, both for in-vehicle and intra-vehicle applications, is leading the way to an exciting nearby future in which vehicles will be "connected" with each other and with the surrounding infrastructure through wireless networks. These vehicle networks will be capable of coordinating with each other, predicting and avoiding collisions, increasing traffic throughput, improving fuel efficiency, and optimizing routes in real-time based on distributed estimation of global congestion information. In these cyber-physical transportation systems, human operators will be fully supported and assisted in all their decisions, and they will be warned and guided toward optimal and safe behavior.
There are a large number of challenges for control and systems theory that must be overcome in order for cyber-physical transportation systems to reach their full potential. First of all, these systems are software-intensive while being physical, that is, they are "hybrid". Yet, they must be designed so that their physical and computational parts interoperate correctly. Additionally, these systems have large state spaces and hence any control/optimization task has to overcome complexity bottlenecks. Finally, they involve human operators, and hence humans should be accounted for in any analysis and design task. In this workshop, we will provide an overview of the current state of the art in cyber-physical transportation systems, technical and technological challenges, and future research directions. Specifically, the workshop will provide a technology overview, including state of the art in driver-assist technology and current trends in automotive companies and the government, formal methods for safety verification and design both for inter-vehicle and intra-vehicle systems, scheduling and routing problems for both air and ground vehicle networks, and design problems in the presence of human operators.
Controlling Green Buildings: Challenges and Opportunities (external website)
Organizers: Francesco Borrelli (UC Berkeley), and Meli Stylianou (CanmetENERGY)
Speakers: Researchers from UC Berkeley, Lawrence Berkeley National Laboratory, ETH Zurich, United Technology Research Center and Honeywell.
Location: CanmetENERGY, 1615 Lionel Boulet Boul., Varennes, Québec
Commercial buildings are plants that process air and water in order to provide comfort for their occupants. The components used are similar to those employed in the process industry: chillers, boilers, heat exchangers, pumps and fans. The control design complexity resides in adapting to time-varying user loads, occupant requirements and quickly responding to weather changes. Today this is easily achieved by over sizing the building components and using simple control strategies.
In recent years, the drive towards energy efficient buildings, has added an interesting dimension to the distributed control problem of commercial buildings. In this context, high performance, green buildings are expected to maintain comfort and satisfaction of their occupants while optimizing energy, being robust to intermittencies in the renewable energy generation and be responsive to signals from the smart grid. Achieving all this in often under-monitored environments is a big challenge. New methodologies and tools for controlling this emerging building type are required.
This workshop will present an overview of high performance green buildings, both from a systems and a control perspective, and will present new trends, ideas and recent results from research and development projects. Topics covered will include:
- The use of weather and load prediction to manage the operation of the building's electromechanical and thermal storage systems;
- Control of building-integrated photovoltaic installations, combined heat and power systems, and other renewable energy systems;
- Optimisation of single and multi-building performance subject to smart grid signals;
- Innovative ways to overcome interdisciplinary barriers to implement advanced control techniques in building systems.
Important note: This workshop will be held offsite at CanmetENERGY, 1615 Lionel Boulet Boulevard, Varennes, Québec. A bus will depart from the Mansfield St. entrance of the Fairmont Queen Elizabeth hotel promptly at 8:30 a.m.
Health Management, Fault-tolerant Control, and Cooperative Control of Unmanned Aircraft (external webiste - North America, external website - Europe)
Organizers and Speakers: Youmin Zhang (Concordia University), Camille Alain Rabbath (Defence R&D Canada), YangQuan Chen (Utah State University), Christopher Edwards (University of Leicester), Cameron Fulford (Quanser Inc.), Hugh H.-T. Liu (University of Toronto), Liang Tang (Impact Technologies, LLC), Didier Theilliol (Nancy Universite), Antonios Tsourdos (Cranfield University)
Location: Hochelaga 5 Room
Unmanned systems including Unmanned Aerial Vehicles/Systems (UAVs or UAS), Unmanned Ground Vehicles (UGVs), and Unmanned Underwater Vehicles (UUVs) etc are gaining more and more attention during the last a few years due to their important contributions and cost-effective applications in several tasks such as monitoring, surveillance, search, rescue, as well as military and security applications. Benefited from the recent significant advances and development of UAVs, development and application of fault-tolerant control as well as cooperative control techniques with small/miniature UAVs have been emerged and developed quickly in recent years, since UAVs provide a cheap and operative experimental testbed for development, implementation, and testing the latest developed health management, fault-tolerant, and cooperative guidance, navigation and control techniques.
The workshop will demonstrate the state-of-the-art techniques and development in health management, fault diagnosis, fault-tolerant guidance, navigation and control (GNC), as well as multi-vehicle cooperative GNC techniques. Overview of past, current and future research activities and research and development outcomes on the health management, fault diagnosis, fault-tolerant control, and cooperative control applications with emphasis to UAVs will be presented, which include aircraft such as quadrotor rotary- and fixed-wing UAVs; Linear and nonlinear techniques for modeling, fault diagnosis, fault-tolerant control, path and trajectory planning/re-planning, cooperative/formation flight guidance, navigation and control, based on a set of quadrotor helicopter and fixed- wing UAVs testbeds, with practical application scenarios on persistent surveillance and coverage control with multiple unmanned aircraft, will be presented; Multiple UAS operations toward verifiable autonomy and assessment of the potential insertion of UAS in the air transportation system, as well as reliability and safety aspects of UAVs will also be discussed, based on the experience and latest development from world-leading researchers (North America and Europe).
Audience will have opportunity to visit a set of quadrotor helicopter UAVs and fixed-wing UAVs at the Networked Autonomous Vehicles Lab (NAVL) of Concordia University for REAL flight demonstrations of fault-tolerant and cooperative control techniques presented in the workshop.Nonlinear Regression Modeling: A Practical Guide
Organizer and Speaker: R. Russell Rhinehart (Oklahoma State University)
Location: St. Charles Room
This workshop will be a practical guide for nonlinear regression modeling. Although theoretical analysis behind techniques will be revealed, the takeaway will be:
- Techniques for defining the regression objective,
- Choosing an optimization approach,
- Design of experiments for data generation,
- Statistically-based, data-based model validation/discrimination, and
- Criteria for selecting an appropriate model order/complexity.
A pre-textbook monograph of approximately 200-pages will be used as the workshop notes. Exercises and code can be implemented in any environment, but Excel/VBA will used as in-workshop examples and exercises. Participants are invited to bring a laptop with Excel version 2010 or higher.
Models based on data are often central for model-based control, forecasting, training simulators, analysis and diagnosis, mechanism validation, and supervisory optimization. For many of these applications nonlinear models are preferred in order to capture the process/device behavior. Regression is the procedure of fitting models to data, and nonlinear regression means the adjustable model coefficients do not appear linearly within the model.
Since processes/devices are neither ideal nor stationary, model coefficients need to be adjusted periodically to make the models fit experimental data. Even for seemingly linear models, a variable delay introduces a nonlinear model coefficient.
Workshop topics will include equation structures, optimization of parameter values in the presence of constraints and local traps, choosing optimization stopping criteria based on model properties, data pre-processing and post-processing, data-based model validation, discrimination between models, design of experiments that support validation outcomes, propagation of uncertainty, and model utility evaluation.
This is not the standard linear regression approach to develop response surface model structures, and classic experimental designs such as Latin Square, Box, and Star plans. This workshop will focus on techniques for nonlinear regression.
Quasilinear Control: Analysis and Design of Systems with Nonlinear Actuators and Sensors (external website)
Organizers: Semyon M. Meerkov (University of Michigan), Pierre T. Kabamba (University of Michigan), Yongsoon Eun (Xerox Innovation Group), ShiNung Ching (MIT)
Location: Harricana Room
Quasilinear Control (QLC) is a set of methods for performance analysis and design of feedback systems with nonlinear actuators and sensors. The approach of QLC is based on the method of stochastic linearization, which reduces the nonlinearities of actuators and sensors to quasilinear gains. Unlike the usual, Jacobian linearization, stochastic linearization is global. The price to pay is that the quasilinear gain depends not on the operating point but on all functional blocks and exogenous signals of the system.
Using this approximation, QLC extends most methods of linear control theory to systems with nonlinear instrumentation. These include: system type, error coefficients, transients characteristics, admissible domains, root locus, LQR/LQG, H?, and LMI. The resulting equations consist of usual control-theoretic relationships (e.g., Lyapunov and Ricatti equations) coupled with transcendental equations, which account for the quasilinear gains; a bisection algorithm for solving these equations is provided. In addition, QLC addresses problems that are specific for systems with nonlinear actuators and sensors. These include so-called Instrumented LQR/LQG, where optimal controller is designed simultaneously with optimal actuator and sensor, and partial and complete performance recovery, where the degradation of linear performance is either contained or completely eliminated.
In this workshop, the method of stochastic linearization and the theory of QLC will be covered using the recently published textbook: S. Ching, Y. Eun, C. Gokcek, P. Kabamba, and S. Meerkov, "Quasilinear Control", Cambridge University Press, 2011.
Robust Networked Infrastructures: Reliability and Security from Game Theoretic Perspectives
Organizers: Saurabh Amin (MIT), Tamer Basar (UIUC), Shankar Sastry (UC Berkeley), Galina Schwartz (UC Berkeley), Hamidou Tembine (Supelec), Jean Walrand (UC Berkeley)
Location: Hochelaga 6 Room
Modern communication and physical infrastructure networks inherit the vulnerabilities of commercial IT solutions, and are subject to correlated hardware malfunctions and software flaws. There is an urgent need to develop proactive mechanisms to enhance the survivability of physical and cyber-based processes of the next generation networked systems in the face of both random and adversarial failures. Due to the presence of interdependencies, the detection and isolation of security failures (attacks) and reliability failures (faults) is, in general, costly. Moreover, many critical networked systems are predominantly managed by profit-driven, private entities. Due to the presence of incomplete and asymmetric information, the cyber economic incentives for provisioning of reliability and security of many networked systems are suboptimal. The development of new mechanisms for increasing the survivability of networked systems requires the integration of tools from network games, robust control, and computer security. The focus of this full-day workshop is on game theoretic analysis of interdependencies between malicious attacks (security) and random faults (reliability) in networked infrastructure systems.
Numerous recent workshops have addressed the subject of robust control in networked environments. This workshop will systematize the body of existing literature, and review and harmonize the existing terminology. Further, we will especially focus on strategic interactions of the parties, whose choices affect the systems' reliability and security. To achieve this goal, we will employ game theoretic tools. The workshop will focus on foundational aspects of the field, will suggest the future research directions, and highlight the promising research topics. Our workshop targets academic researchers and industry practitioners who are interested in contributing to the emerging field of secure and dependable networked systems. We will focus on the following topics:
- Incentive mechanisms for improving reliability and security
- Analysis of interdependent network risks
- Congestion management in the presence of security & privacy risks
- Infrastructure security as a public good
- Applications to communication infrastructures, electricity, transportation
Smart Grid Markets: Integration of Renewables, Pricing, Modeling, and Optimization (external website)
Organizers: Anuradha Annaswamy (MIT) and Arman Kiani (Technische Universitaet Munchen)
Speakers: A. M. Annaswamy (MIT), P. Varaiya (UC Berkeley), S. Meyn (U Florida), M. Dahleh (MIT),
D. Bienstock (Columbia University), and A. Kiani (Technische Universitaet Muenchen)
Location: Hochelaga 4 Room
Motivated by the growing energy needs amidst compelling sustainability and environmental concerns, a new architecture for energy management, labeled Smart Grid, is emerging where increasingly energy generation, transmission and distribution are expected to be controlled by cyber-enabled and cyber-secure components. A major component of this architecture is the design of electricity markets, which pertains to the optimal scheduling of power generation based on bids from all participants. The challenge is to carry out this scheduling with a high level of integration of renewable generation sources, a formidable task due to intermittency and uncertainty of the latter. The emergence of Demand Response, a new concept of controlling loads via cyber based communication, control and economic signals, and new technologies related to Advanced Metering Infrastructure including smart meters and smart devices, have introduced tools for exchanging information about all participants and making decisions in a distributed framework. This workshop will provide a snapshot of the electricity market, its role in designing a smart grid, the main challenges, and promising approaches to address them.
Offered over a whole day, the workshop will present an overview of current market practices, challenges related to intermittency and uncertainty associated with the integration of renewables, and new methods and mechanisms pertaining to dynamic pricing, dynamic modeling, demand response, stability, robustness, and optimization. Fundamental issues that affect stability, volatility, and risk-mitigation in smart grid markets will be discussed. Following are some of the topics that will be covered:
- Dynamic modeling and stability of wholesale markets under congestion constraints
- Risk Limiting Dispatch for near real-time operations
- Service vs. Commodity point-of-view of electricity and relation to volatility
- Pricing mechanisms and stability of markets
- Effect of forecast horizons, ramping constraints, and predictive control on market stability