 |
| Professor Claudio De Persis explains his latest nonlinear model, showing how temperature sensors can communicate. Photo: Gijs van Ouwerkerk |
Nonlinear Control Specialist De Persis was born in Veroli in central Italy. While he still maintains his tenured position at Sapienza University of Rome, De Persis joined the UT on November 1 as Chair of Mechanical Automation and Mechatronics within the Faculty of Engineering Technology (CTW). While sitting in his spacious office at the Horst, he shared his thoughts about his interdisciplinary field, recent projects, and his native Italy.
What does a nonlinear control specialist do?
‘We combine engineering and mathematics to build models used to analyze the behavior of systems and help to regulate and control them. For example, you might want to control an airplane’s trajectory during flight, or conditions that maintain animal populations, or a town’s heating system. I try to create models that are described by nonlinear equations. Nonlinear models are richer than other models and can handle complex dynamics.’
Can you describe your creative process?
‘First I try to understand the problem. I do this by talking with the research engineers. Then I formulate the various conditions into mathematical terms and attempt to describe the problem using mathematical relations and equations. Math is just a matter of language. And then, with these equations, I try to solve the overall problem.’
‘For example, I recently worked in a project supported by the Danish Research Council for Science and Technology and Production Sciences which involved several companies and universities. One of the companies manages a large district heating system which acts as a centralized power plant, distributing hot water to apartment blocks. The problem arose: Can the controllers that regulate the system reconfigure themselves to react independently and smoothly when a new apartment block is added to that system?’
‘We derived a nonlinear model for the system and designed the control algorithms. It was a success and now the model will be implemented. We hope to have the actual system in place by 2011.’
Are you ever surprised when the theoretical algorithms actually work in the field?
‘In fact, the engineers and I were very surprised by our success in Denmark. The model was complex. We are using small pipes with turbulent flow which makes the equations highly complex. Also, there are many uncertain plant parameters and it is impossible to compute all the unknown variables. Theoretical speculation can be a risky investment, but when it works it leads to real innovation. Our control algorithm was just fine. We didn’t even have to fine-tune it.’
I imagine you spend a lot of time thinking. Do you ever have one of those Eureka! moments in the bathtub when you have a flash of intuition like Archimedes?
‘Yes, flashes of intuition are typical for me. I might be boarding a train wondering where the best seat is when I suddenly have a solution to a problem I’ve been working on. When I am not focusing on a problem then I can often see the core obstacle blocking me from finding the solution. Sometimes I dream a solution.’
‘But it can also work the other way. For example, I might be at a concert, in the middle of the performance, and suddenly I realize I made a wrong assumption in the proof from two weeks earlier.’
I keep thinking of Leonardo da Vinci. As an Italian, did you ever find inspiration from him?
(laughing) ‘Oh, no! He is too great. He wasn’t really a mathematician. He was a great artist and mechanic, but no…I could never even come close to him. I found inspiration from my mentor and professor who I regard as a pure figure of science. My advisor Professor Alberto Isidori patiently taught me how to approach a problem in a true methodological way and helped me develop a refined attitude to all that I do.’
You are now commuting back and forth from Rome. How do you manage all that traveling?
‘I’m used to it. I’m often traveling. Last week I was in Shanghai. I keep myself busy on these flights with emails, preparing presentations, and reviewing papers. [De Persis is an editor of the International Journal of Robust and Nonlinear Control.] I don’t need much to work—just a paper and pencil. Colleagues and supporting sponsors usually help run the modeling experiments. But the fundamental research is easily done on a plane.’
‘Of course, I miss my family. I spend about 70 percent of my time here and the rest in Rome where my wife Raffaella and six-year-old daughter Elisa live. I hope they can join me here next year.’
How did you end up at the University of Twente?
‘In Italy, it is very difficult to be promoted or find money to support your research and create your own group. It is really a static situation. The brain-drain phenomenon in Italy is parallel to that in third-world countries. I know this from my own experience, my colleagues and my students. Things are stuck and it’s difficult for younger Italians to find their own way.’
‘I was very happy to come here. The UT is a very open community. Its international flavor is excellent.’
How are you finding life in the Netherlands so far?
‘I really don’t see much difference between the Netherlands and Italy. I spent a number of years in the USA and I experienced more of a difference there. I’m a big believer in the European Union. People talk about the Dutch directness, but honestly, I think Dutch people are much more polite than Italians. My only problem so far is with the coffee. I have my own espresso machine here and my own tricks for brewing the best cup.’
Italians are known for their good food and wine. Do you have a favorite Italian dish?
‘Yes, of course! In fact, my wife comes from a noble family. And in the 1850s her family had a number of different chefs working for them. Her aunt recently collected their recipes into a book. My favorite dish is called Toteri, pasta filled with pork, veal and a special tomato sauce which you need two days to prepare. They are about the size of your fist and, depending on my appetite, I can manage up to four of them, they taste so delicious.’
‘My wife and I try to make these old, traditional recipes when we have time. But my real passion is history. I love reading about ancient Greek and Roman history and now I am interested in the history of the Balkans.’
Sputnik to Robotics
Mechanical automation and mechatronics is concerned with the design and development of methods, and equipment for the control and automation of mechanical systems and physical processes. ‘When my mentor, Professor Isidori would go to a conference on nonlinear control theory in the 1970s, there might be 20 experts worldwide,’ said De Persis. ‘Today it is a very attractive field.’
Historically, nonlinear theory began to be more widely applied to automatic control and feedback systems in the 1960s. In the Soviet Union, there was a great deal of activity in nonlinear controls design which helped them launch Sputnik, the first satellite, in 1957. This created tremendous activity in the United States in automatic controls designed for aircraft and, eventually, robotics.
At the UT, research focuses on technologies that enhance performance of high-tech equipment, such as machining systems for laser materials processing, smart mounts for vibration isolation in precision machinery and MEMS-based precision manipulators. ‘I’m hopeful that we can establish working relationships with other departments to create innovative nonlinear control designs,’ said De Persis.
