General Information

Overview

The University of Illinois Urbana-Champaign and North Carolina State University are pleased to announce the NSF Workshop on Control Co-Design (CCD), which is sponsored by the National Science Foundation (NSF) and will focus on bringing together a diverse set of viewpoints related to the topic of combined physical and control system design. This workshop will provide participants with the opportunity to:

• Connect with a diverse domain of researchers on an emerging subject with extremely high scholarly and societal impact;
• Collaboratively identify strategies for industry adoption of control co-design (CCD) methodologies and tools;
• Be a part of defining how CCD scientific research is done in the future;
• Shape potential future opportunities for control CCD research, and help establish a new interdisciplinary research community.

The objective of this workshop is to bring together scientific and technical expertise from the dynamics and controls, the engineering design research, and other relevant communities to craft and engage fundamental research questions at the interface of physical and control system design. We also seek involvement of individuals with significant experience in application domains that stand to benefit from advancement and adoption of CCD methods (e.g., renewable energy, aerospace, automotive, precision manufacturing, etc.). We hope that this effort will lead to the creation of a new research community that can move forward with 1) theoretical frameworks for active system design, 2) creation and analysis of integrated active system design methods with new levels of design and modeling fidelity, and 3) development of tools for engineering practice, such as design automation methods and validated design guidelines, with potential for significant societal impact. We envision continued regular interactions in different forms, including technical collaboration, future CCD community meetings, virtual micro-workshops, and identification of additional technical communities who may contribute to or benefit from CCD activities.

Background

The overall performance of actively-controlled systems can be improved, sometimes very significantly, by considering physical and control aspects of system design together in an integrated way. Examples of physical system design decisions include geometric, inertial, material, and configuration variables. Optimization of a physical system design is the subject of significant ongoing efforts in design optimization research. Control variables, on the other hand, include parameters/settings that are manipulated in real-time software, often based on sensed information. CCD methods generate significant performance gains when design coupling is strong, which refers here to how physical system design decisions influence how the control system should be designed, and vice versa.

The existence of significant design coupling in actively controlled engineering systems has been recognized for a few decades, and several research initiatives have focused on creating new methods to capitalize on this coupling to improve system performance. Notably, control structure interaction (CSI) was an area of high research activity in the 1980s and 1990s, aiming to simultaneously reduce structural mass while meeting requirements involving structural dynamics (vibration, positional accuracy, etc.) and other failure modes. While this earlier work has resulted in important integrated methods and their validation, it largely focused on specific systems such as flexible structures for space applications. Unifying theory and methods have yet to be created. Similarly, CCD work carried out within the control systems community in the 2000s established several promising frameworks for combined plant and controller design, but the theoretical guarantees have been limited to systems with restrictive structures (e.g., linear systems, unidirectional coupling) and have been validated only on relatively simple, low-order applications. In recent years CCD methods have been adopted in large research programs, including ARPA-E ATLANTIS and SHARKS programs that focus on using CCD to reduce the cost of renewable energy.

This workshop will focus on new interdisciplinary research initiatives that not only leverage earlier work, but address completely new elements at the interface of physical and control system design. This event will bring together researchers with strategically complementary expertise to begin filling the existing gaps in CCD theory, methods, human factors, and other factors influencing industry adoption. We hope that this community building effort will create new research capabilities and increase impact compared to continuing current separate research efforts.

Workshop Aims

• Begin construction of unifying theory for general integrated design methods for actively controlled dynamic engineering systems.
• Articulate the interface between the engineering design and control systems research domains, including defining terminology and language and understanding, as well as identification of scientific needs and strategies at this interface.
• Build upon earlier work in integrated design methods, while also raising the profile and general awareness of these efforts so that researchers working in various areas related to integrated active system design can combine their efforts to have greater impact.
• Identify collaborative research opportunities that could result in new theory or design tools with new capabilities.
• Critique of CCD research advancements and identify factors limiting high-impact adoption of CCD methods in systems engineering practice.