Controlling the Dynamics
Emeritus professor Dynamics and Control Henk Nijmeijer looks back on 23 years of professorship
Read more- People , Research
- 23/05/2023
Controlling the Dynamics
Henk Nijmeijer, founder of the Dynamics and Control research group, delivered his farewell lecture on May 12. The emeritus professor, who traded Twente for Eindhoven and mathematics for mechanical engineering 23 years ago, is best known as an automotive expert who focused mainly on cooperative driving, but he considers his research in the field of soft robotics just as important. He explained to Cursor how you can make trucks communicate with each other, why he is not a fan of institutions, his admiration for Huygens and his fascination for fireflies.
Nijmeijer has been a professor of Dynamics and Control at the Department of Mechanical Engineering for 23 years, but he was originally a mathematician. “Actually, I wanted to study chemistry, but there was little job opportunity in that field at the time,” smiles the emeritus professor. After obtaining his master’s degree in mathematics in Groningen, he went on to earn his doctorate there with research he conducted at CWI, the Dutch national research institute for mathematics and computer science, in Amsterdam. “Which was still in mathematics, but by then, I was already working on control systems technology, essentially my current field.” After receiving his PhD, he was appointed to the Department of Applied Mathematics at the University of Twente and started collaborating more and more with engineering departments such as electrical and mechanical engineering.
In 1997, he started working at TU/e as a part-time professor in the control engineering group within mechanical engineering, assisting with the fundamental side. “It was a perfect fit.” After three years of commuting back and forth between Twente and Eindhoven, he was offered a full-time position at TU/e and moved to Eindhoven. Another smile: “Apparently, I did something right.” At TU/e, he immediately set up a new research group: Dynamics and Control. “It reflects my background. Mathematics still plays a large role in this research area.” To illustrate, he shows a dissertation on his desk. “It’s full of mathematics, equations and modeling.”
Nevertheless, as someone with a theoretical background, the switch to the lab was quite a big step for him. “Of course I tried to attach experimental work to the research for every student and PhD candidate, but initially that was a big challenge for me personally,” he admits. It was somewhat confronting to find yourself in a lab with robots and all kinds of setups all of a sudden. “It took a while to get used to that. It was quite scary at first.”
Automated versus autonomous driving
He and several other lecturers soon came to the conclusion that an automotive track should be launched within the ME department’s master’s curriculum. It was clear from the beginning that the track needed to be broader than just mechanical engineering. “A car is full of electronics, so you also need other disciplines,” he explains. Eventually, the joint initiative led to a complete Automotive Technology (AT) program with a bachelor’s track, master’s program and EngD program. Lecturers from Computer Science, Electrical Engineering and Industrial Engineering are involved, making it a very multidisciplinary program.
Like I said beforehand - which I can only confirm now - I don’t believe in autonomous vehicles at all
Nijmeijer was also the program leader of Integrated Cooperative and Automated Vehicles, an NWO Perspectief program launched in 2016. As part of the program, more than 20 young researchers worked on the technological challenges related to the fluidity and safety of automated and cooperative driving.
It should be noted that “automated” driving is very different from “autonomous” driving, Nijmeijer emphasizes. For me, autonomous refers to fully self-driving vehicles that can drive anywhere at any time. Like I said beforehand - which I can only confirm now - I don’t believe in that at all.” According to him, driverless vehicles would only be able to drive with certain restrictions and under certain circumstances, for example in a bus lane. “Drive around in any city and you’ll see that people aren’t following traffic rules and the roads are swarming with all kinds of vehicles. Letting an autonomous vehicle drive through it would mean that it would probably just be standing still all the time,” explains the automotive expert. “And if it starts driving, make sure you get out of the way,” he says with a wink.
This is why Nijmeijer says it is only right that the hype surrounding self-driving cars has subsided somewhat and it remains to be seen whether it will even happen at all. “It probably won’t happen in our time. Maybe by the time I die, but not much sooner.”
Cooperative driving
He says there are much more realistic possibilities in the field of cooperative driving. The so-called Cooperative Adaptive Cruise Control (CACC), the successor to Adaptive Cruise Control, allows vehicles to communicate with each other via a wireless connection. As a result, cars can be driven much closer together while maintaining safety. This ensures more driving comfort, improved traffic fluidity and reduced fuel consumption. “It’s a very interesting discipline and we’re really good at it here in Eindhoven.”
We showed the potential in the GCDC, but that’s something completely different from bringing such a car to the market
In 2011 and 2016, the Grand Cooperative Driving Challenge (GCDC), aka the “Olympic Games” of cooperative driving, took place in the Netherlands. For this event, the TU/e team converted a DAF truck and a number of passenger cars and drove on the A270 between Helmond and Eindhoven using CACC technology. “We showed the potential in the GCDC, but that’s something completely different from bringing such a car to the market, which involves a lot more work,” Nijmeijer says. “The same thing happened with Lightyear; you could have predicted that it wasn’t going to work out this way. You need at least half a billion euros to make it a commercial success.”
However, cooperative driving has a very different cost structure and therefore has a better chance of success, he believes. “You don’t have to modify the whole vehicle, just create a sort of Wi-Fi protocol between the vehicles, a wireless connection that allows them to communicate with each other,” he explains. “But it only works if the vehicle in front of you also has this protocol, otherwise it’s useless.”
“A few years ago, we worked on a European project with partners such as Ericsson and KPN to see if it would be possible to run the communication between vehicles via your smartphone,” he continues. That could be a next step. “It would be easier because then you don’t have to install an extra box in the car. At the same time, however, the project ownership gets a little more complicated, so the business case is still being developed.”
Something that could be done in the short term, is applying this technology to freight traffic. For example, you could have trucks driving in a sort of column behind each other, with much smaller headway distances than normal. “It would have to be in the left lane, though, because otherwise problems would arise due to merging cars. And preferably at night, because there’s less traffic on the road,” Nijmeijer explains. If such a thing is to be implemented, it is advisable to do so at a European level, he stresses. “Otherwise, it’s not going to achieve much.” Steps are currently being taken and testing is underway. “I think I can cautiously say that it could happen within five or ten years."
Institutes
Nijmeijer thinks multidisciplinary collaboration is extremely important. In fact, he is convinced that the most important discoveries of the coming decades will be at the interfaces of different disciplines. “The question is whether institutes are absolutely necessary for this,” he argues. The emeritus professor does not mince his words in stating that he is personally not a fan of institutions. “You can always question whether they add value. Even here at TU/e, you can ask that question.”
If almost all departments are attached to an institute and everyone is collaborating with each other, as is the case with EAISI, I wonder if it’s even worthwhile
For example, he feels that the EAISI (Eindhoven Artificial Intelligence Systems Institute) is not an adequate institute. “If almost all departments are attached to an institute and everyone is collaborating with each other, as is the case with EAISI, I wonder if it’s even worthwhile,” he argues. “If you have institutes at all, they should exude strength and there should be a certain exclusivity to them. ICMS is a good example of that, but I have my doubts about EAISI.”
Nijmeijer did on multiple occasions - including during his farewell lecture - make a case for the establishment of an automotive center. “We have excellent automotive programs here at TU/e, so it would be a good thing to bring more visibility to the field,” he believes. “Officially it falls under EAISI, but I think that’s far too broad a conglomerate, whereas automotive is a very specific research area.”
Soft robotics
Nijmeijer does call his collaboration within ICMS on Soft robotics “very valuable.” In 2018, he and several colleagues set up a 4 TU program dedicated to Soft robotics. “Many robotics are pick-and-place machines in factories, no humans get near them. But when the robot is used to help people, things could get quite dangerous. Before you know it, you’re knocked down by a heavy robotic arm,” he explains. Soft robotics deals with the question of how to make robots more mechanically flexible and softer. This is essential to enable safe interaction between humans and robots. “It’s incredibly important because that’s how you can bring robots closer to humans.”
This requires adapting materials and developing new, less rigid ones. Designing soft materials is one of the research areas within ICMS, which illustrates Nijmeijer’s connection to the institute. However, the use of these new materials could mean that the accuracy of the robots deteriorates. Nijmeijer: “From my own field of research, I then start thinking: if the robots are flexible, how are you going to model them and make sure that you achieve a certain precision that is sufficient.”
Synchronization
Other key areas of interest in Nijmeijer’s work include synchronization, the phenomenon where two systems display the same behavior or identical dynamics. “The interesting thing about synchronization is that it occurs not only in processes and machines, but also in nature, among animals. Well-known examples are schools of fish swimming in synchrony or a flock of birds flying in a perfect V formation. What we want to know is: how does that happen?” The question might be simple, but it’s all the more difficult to scientifically explain the countless examples of this phenomenon.
The phenomenon of synchronization was first described by the Dutch scientist Christiaan Huygens, whom Nijmeijer greatly admires. In 1665, he discovered that two pendulum clocks hanging side by side from the same wooden beam would eventually start to synchronize, regardless of the initial positions. Due to a lack of the necessary scientific knowledge, Huygens was unable to prove it at the time, but he suspected that the beam acted as a medium between the clocks and that synchronization was achieved through vibrations in the beam.
In 2016, Nijmeijer and several Mexican colleagues conducted a study on this effect now known as “Huygens' synchronization”. For this study, they conducted an experiment with two custom-designed, identical clocks that they placed side by side on a wooden table. The pendulums were set in motion in opposite directions, but within thirty minutes they oscillated in synchronicity, albeit at a slightly lower frequency. From that moment on, the pendulums continued to swing back and forth synchronously. The analysis the scientists then conducted proved that Huygens was right and that the 350-year-old explanation was correct.
Fireflies that flash light at the same time or birds that fly in a V formation - we have no idea how it works
This study explained a lot, but there are still more questions than answers when it comes to synchronized motion. “It’s a subject that intrigues me tremendously. Fireflies that flash light at the same time or birds that fly in a V formation,” says Nijmeijer. “We have no idea how it works exactly but it leads us back to complexity, toward ICMS. To the question of what life is and how the behavior of systems is brought about by interactions between the components they consist of. I think these are extremely interesting questions.”
Regularly dynamic
When Nijmeijer was appointed professor 23 years ago, the title of his inaugural speech was “Dynamisch geregeld” [dynamically controlled], referring to the new Dynamics and Control group he set up at TU/e at the time. His farewell lecture was supposed to be titled “Geregeld dynamisch” [regularly dynamic], were it not for the fact that the lecture had to be delivered in English. Nijmeijer had not counted on that when he came up with this title years ago. In Dutch it is still the perfect title, he thinks, because it reflects how his interest in different areas of research changed over the years, but it also indicates that he is not going to fully bid farewell to dynamics after his farewell lecture. Even after receiving his emeritus status, he plans to stay "regularly dynamic”.
Henk Nijmeijer (1955) received his MSc degree (1979) and PhD degree (1983), both in Mathematics, from the University of Groningen. He was affiliated with the CWI, Amsterdam (1980- 1983), and the Applied Mathematics Department of the University of Twente (1983-2000). In 2000, he was appointed a full professor at TU/e. He served as the Graduate Program Director of the TU/e Automotive Systems program in the period 2016-2021. Since January 2015, he has also been the Scientific Director of the Dutch Institute of Systems and Control (DISC) and chairs the Dutch Mechanical Engineering Council. He is a core member of the ICMS with the focus area “complexity and soft robotics”. Since 2021, he has been the Field Chief Editor of the newly established journal Frontiers in Control Engineering.
Discussion