Professor Skelton's Research Statement

• System Design: The Grand Challenge of System Theory
• Designing Models for Systems
• Inspirations from Art and Nature
• Designing Controlled Structures

   

• System Design

My training in control came from an electrical engineering perspective. But to solve control problems for real dynamical systems, modeling issues demanded most of my time. This prompted the pursuit of a degree in dynamics. This combination of the two disciplines flavors the point of view on everything I have to say here. What I learned about the intersection of the dynamics and control disciplines was more important than anything I learned about either discipline. While each discipline is relatively mature, their intersection is still in its infancy. A significant focus of my research is to develop scientific methods for the simultaneous design of multidisciplinary components of systems. Mathematically, this problem is characterized by a statement of performance requirements for the overall system, and the design specifications for the components must be derived from the system requirements.

It might be said that physics paved the way for technology in the first half of the 20th century, and engineering paved the way in the second half. What will drive and enable new technology as we round the corner of the new millennium? Engineering has produced sophisticated component technologies without instructions how to put them together. We have multidisciplinary problems searching for an interdisciplinary theory. So, I believe that systems will be the enabler of technology in the next few decades. I believe that the next grand challenge is to give the soul of control a body, to enlarge the tools of control theory to embrace the much more general problem of system design. Traditional control theory assumes that all other components have been designed except the control component. By system design we refer to the task of determining the design requirements of the multiple components that make up the system, given only the requirements of the system.

If you desire more specific details about these notions, refer to the plenary lecture of the American Control Conference 99, which gives several examples of the deficiencies of existing theories of control and system design, leading to the suggestion that the integration of two disciplines should not start at the state of the art of either discipline. Finally, a new paradigm (tensegrity structures) is suggested for integrating structure and control design.