On the picture: Sergey Kolyubin
Laboratories – Robotics and Cyberphysical Systems
Timecode – 00:18
dmitrykabanov: We usually interact remotely – for example, using comments for the university blog on Habré. Previously, these were photo tours: around the laboratory of cyberphysical systems and FabLab – Alexey Shchekoldin showed us it). Then he demonstrated his tutorial development – robot SMARR with VR and AR elements. Among other things, have you acted as the scientific advisor for this project?
Sergey Kolyubin: Yes that’s right. I was Alexey’s scientific advisor in graduate school. It was a practice-oriented project within a cycle of such works at the university. Now there is another set of student teams under the guidance of graduate students. They are preparing prototypes of their solutions with a commercialization perspective. During this process, such practice-oriented R&D helps to absorb the culture of production, acquainting children with the life cycle of their developments and related activities – for example, with the nuances of drawing up documentation.
Dmitry: In addition to this project, there were many others – you participated with a number of teams in international conferences on robotics, including went to TechCrunch…
Together with this, you take part in the management of the whole mega-faculty, plus you work as a boss department of magistracy… Do you manage to combine this activity, and also stay in touch with younger colleagues and help them?
Sergei: I agree that there is a workload, but I try to be creative in solving any problems. As soon as I understand that there is an opportunity to move away from bureaucratic affairs and spend time in the laboratory with the guys, I try to do it.
Robotics laboratory and development opportunities
Timecode – 02:09
Dmitry: By the way, in one of the photo excursions on Habré there was also a robotics laboratory, where we were shown various gripping devices and industrial manipulators.
Sergei: Yes, this is my laboratory.
Dmitry: What has changed since then? What is the laboratory working on now?
Sergei: This is an international scientific laboratory “Biomechatronics and Energy Efficient Robotics“. We run it together with the professor Stefano Stramigioli, which represents the University of Twente, is a well-known expert in this field, he runs large robotic hubs and recently became a member of the Netherlands Academy of Sciences. This person motivates us and sets challenging tasks. There are other areas – projects RSFsupported by appropriate grants; there are orders from companies; there is also a “sandbox” – initiative projects within the framework of scientific work with undergraduates and graduate students.
In the latter case, the line is wide enough – someone does hand rehabilitation systems, prostheses and classification systems by encephalogram and myography, others – design systems for controlling robots with elastic elements, others – are dealing with the topic of autonomous control, someone is working on haptic systems with remote control and feedback. Some are also involved in our projects for the Russian Science Foundation – developing energy efficient walking and galloping robots.
I give creative freedom. The main condition is the compliance of the project with the direction of laboratory development and the presence of a focus. If we talk about the latter, then we are dealing with systems with “flexible” elements for working with an unstructured environment. This is a situation where we cannot describe all the properties of the environment, and the system must have internal adaptability.
The second key point is the energy efficiency of the system. All this is achieved at the level of algorithms and at the level of design to make the system smart in terms of hardware as well.
How to join these projects
Audio version of the interview in Apple Podcasts
Dmitry: What would you recommend to those who would like to join such a work?
Sergei: There are certain prerequisites at the entrance. Roughly speaking, in my laboratory there are three types of specialists, if we talk about competencies: “mechanics”, “electronics engineers” and “algorithms”. The first – previously studied, done graduation projects or worked in companies related to the design of systems – were engaged in “mechanical engineering”. The second – they know how to program controllers, understand what sensors are, how to organize data processing at a low level. Still others are responsible for higher-level algorithms and the development of robots at the system level.
Robotics is an interdisciplinary field. In addition to techies, experts from the field of medicine and industrial design work here. So, we started a European project from the Strategic Partnerships line from Erasmus. It involves not only representatives of ITMO University, but our colleagues from universities Leuven, Twente, Ozjegin University and Ford Otosan, the brand’s largest truck division. Together we create a graduate educational course for masters and graduate students, and in the course of this process we build closer cooperation between organizations and research groups.
The project itself is about wearable collaborative robotics. In addition to the knowledge that technology developers need for industrial exoskeletons and collaborative manipulators for manufacturing, we touch on the areas related to medical technology and surgical robots involved in operations. We understand that such developments are impossible without practicing medical specialists – you need to understand the problems in detail, engineering work is only 40% of such projects.
What are the features of an interdisciplinary approach
Audio version of the interview in Google podcasts
Dmitry: An interdisciplinary approach is a feature of ITMO University. It can be found on all top programs.
Sergei: Yes. This approach aligns with our strategic goals and our development agenda. This is the understanding of the brand, which should be maintained at all levels, linked with the message that we bring to the outside world, and included in the daily work of employees.
Dmitry: From the point of view of a participant in the educational process, is this interdisciplinary approach somehow combined with an individual trajectory of development or some kind of focus that would make it possible to clearly understand one’s specialization, but at the same time take into account those necessary competencies that need to be taken from related industries?
Sergei: The willingness to look wide is a must. Science is going through another round of development. There were generalists – philosophers and mathematicians, then specialization went, now there is a new round when specialists with universal competencies are in demand.
This approach is possible with the parallel obtaining of two degrees. Many universities have it. We have it – bioinformatics… It requires equal knowledge in computer science, molecular biology and genetics. There is a program for infochemistrycombining competencies in information technology and chemistry. There are less obvious examples. For example, I studied with control systems. But it is impossible to build a control system for something without understanding the subject – what you control. For an electromechanical system, physics, electricity and related processes are needed. If you make a control system for some kind of chemical reactor, you need to understand, at least at the level of abstraction, what is happening there.
If we take cybernetics, then it is an interdisciplinary science in itself. Robotics is a clear example in this regard. There are many such examples across ITMO University.
Balance of theory and practice
Audio version of the interview on Yandex.Music
Dmitry: You are talking about general competences – is this the basis that you look at at the stage of accepting people? For example, according to the tasks of the competition “I am a professional” you can see that these are quite fundamental things that correspond to such requests.
Sergei: Yes, we use different tools – not just exams and portfolio competition.
Dmitry: But on the other hand, when looking at such tasks, doubts may arise whether a person will be able to practice enough to realize all this knowledge – for example, in the framework of cooperation with a company in a particular field.
Sergei: I understand. Now there is a great demand for a quick transition to practice. When I was a student, it was motivating. Control systems and graphs on the screen are theory, and the robot is driving or not, solving its problem or not – this is already a visual demonstration. You see the result and evaluate the usefulness of the development. But only simple tasks can be solved quickly.
More complex challenges cannot be answered without fundamental knowledge. You will simply spend multiple times more time, but if you have an understanding and knowledge of formulas and methods, everything will go faster.
We just got used to getting information quickly and achieving results. But sometimes you need to spend a few weeks studying the literature and sorting out the theory. You, as a specialist, “cost” exactly as much as the training of such a specialist. If you know how to solve only basic tasks, implemented with a minimum level of competence, you will not be able to claim significant income. There is another threat: what some people previously received money for are now starting to make algorithms in a good form. The level of costs in the long run may not be comparable. Therefore, it is important to be able to solve complex problems from the point of view of long-term “convolution tension”. Fundamental preparation is useful for this.
Dmitry: If we go back to laboratories – robotics and cyber-physical systems, then the layman can imagine them as some kind of office, where people in white coats and goggles do something monotonous around the clock.
Sergei: We don’t have that. One of our values is respect for the individual. The main thing is implementation in a professional sense. But if safety precautions require, you need to equip accordingly. Of course, we want to see specialization for each of the specialists. Then you can form a team with a good level of interchangeability and be sure that each of its members is strong in something different.
Dmitry: Can everyone have a personal project or are they always common?
Sergei: This is the essence of laboratory work. A good example is the cyber-physical systems laboratory. It exists because we have set a task at the faculty level – to gather specialists in different fields, to provide them with infrastructure so that they can solve joint problems, communicate, and form mutual understanding. It is more difficult to do this without specialized locations, but it is also important to have personal space to work on your tasks. At the same time go out and communicate with colleagues. The optimal balance between one and the other is art.
Other episodes of our podcast on Habré:
- How to approach syncing AR content with a stadium-wide show
- What scientific data visualization and gamedev have in common
- Quantum hacking and key distribution
- Digest: AI systems, neural networks, data analysis and ML – discussing personal experience