Michal Smíšek: We deal with the development of robotic systems for the research of planetary surface exploration
Young and successful Slovak scientist, mathematician and robotic engineer Ing. Michal Smíšek, PhD., works at the German Aerospace Centre (DLR). He deals with the development of robotic systems for the exploration of planetary surface and other hard-to-reach areas for humans such as the ocean floor. He is currently dealing with the research of robotic sensors, i.e. the technological components that enable a robot to perceive its surroundings.
Michal Smíšek has recently participated in a simulated lunar mission on the Etna volcano in Sicily. The mission was carried out in the scope of the ROBEX project, which demonstrates the readiness of European robotic systems. After returning, he gave us an interview.
M. BARTOŠOVIČOVÁ: Your studies at the Faculty of Informatics and Information Technologies of the Slovak University of Technology in Bratislava was obviously a good choice. How were you as a student? How did you choose your next direction?
Michal SMÍŠEK: At the time I was studying there, the Faculty of Informatics and Information Technologies of the Slovak University of Technology was a demanding school with a contemporary focus. Thanks to the acquired skills, its graduates were able to start a career in Bratislava with a relatively high starting salary and the prospect of further growth. However, I felt that FIIT usurped virtually all the student's time and gave him virtually no right for leisure activities or personal development. Because I assumed this right, I was an average to sub-average student. One that was barely passing the exams. Overall, I was not very happy with my life at that time.
After earning the bachelor's degree, I decided to join the Department of Mathematical and Computer Modelling (MPM) at the Faculty of Civil Engineering of the Slovak University of Technology. The content of the program was also very up-to-date, but compared to FIIT, the focus was much more on mathematical skills – analysis, algebra, differential equations, numeric – their computer implementations, and their practical applications. That fitted me much better. I also saw a more human approach to students. I paid for this redirection, above all, with the transfer year, which meant I had to learn all the missing mathematical skills from the bachelor's level needed to continue with the mathematical and computer modelling engineering studies. I have done this thanks to the special study plan. Today, I see it as a great investment. I have stayed at the Department of Mathematical and Computer Modelling after the engineering studies and I have earned a doctorate degree under the guidance of an excellent pedagogue, professor Karol Mikula, in the field of image processing. Today, thanks to the longer time that has passed since then, I am able to better weigh what my discontent was with the first studies and what was the satisfaction with the others. I had to get to know myself.
M. B.: When did you decide for work abroad? Did you have a specific idea what you would like to do and where?
M. S.: Towards the end of my engineering studies, I was curious about the quality of my mathematical and computer education, so I decided to test myself: I decided to spend my summer doing an internship in my field – digital image processing. I told myself that if I find a good opportunity anywhere, I’ll take it. I have sent around 60 applications all around the world. They accepted me to the Munich company DLR. At that time, I did not know it was an official German space agency. I also didn’t know that a three-month internship will develop into a four-month external cooperation and that this external cooperation would develop into a permanent job at DLR. So, I did have an idea of what I wanted to do, but not where I wanted to do it.
Logistics of robot transport, Etna field test 2017
M. B.: You have been working at the Deutsches Zentrum für Luft- und Raumfahrt (DLR) since 2011. What projects have you been working on and with what results?
M. S.: I work at the Institute of Robotics and my specialisation is sensorics. Sensors are devices by which the robot observes its state and the state of its surroundings. The robot then analyses the data from its sensors and utilises them for decisions on further actions.
In terms of hardware, I am particularly interested in cameras, which are sensors that are robots' equivalent of the eyes. I am also interested in Inertia Measurement Unit (IMU) – which is a trio of accelerometers and gyroscopes, which are robots' equivalent of the middle ear. In terms of problems, I am interested in the correlation between the data from these sensors and the status of the robot and its surroundings. We call this issue “sensor calibration”. I am also interested in how can be some sensors used synergistically if they all work properly, and in compensatory way, in case some of them works suboptimally or fails. This issue is called “fusion of sensors”. Both issues need to be resolved if we aim to build a robust and mobile robotic system that performs complex tasks at least partially in an autonomous way.
I am responsible for the development and maintenance of DLR CalLab software, which is one of the world's leading camera calibration solutions and we plan to make it publicly available for free. I am also responsible for the InFuse project funded by the European Strategic Research Cluster for Space Robotic Technologies (EC SRC), whose task is to create a universal solution for fusion of sensors usable in the upcoming orbital and planetary missions. The resulting product should also be publicly available.
Recently, I have spent a lot of time in a working group that develops mobile robots for the exploration of hard-to-reach areas, such as planetary surfaces or areas affected by disasters.
Michal Smíšek and LRU during Etna field test 2016
M. B.: Which one of the projects was the most challenging for you?
M. S.: Definitely the ROBEX project concerning the above-mentioned working group on mobile robots. This project has been gigantic in terms of number of people allocated, duration and ambitions. Our task was to demonstrate the readiness of European robotic technologies for exploration of the extraterrestrial planetary surface – specifically, to partially autonomously execute a series of operations necessary for a seismological survey of the Moon.
Last summer, for 10 days, we organised a test of our LRU planetary rovers (Lightweight rover unit) on the Etna volcano in Sicily. The slopes of Etna have been recognised in previous studies as the most similar environment to the lunar one in Europe from a seismological and geological point of view. Our robots, which were until that time only laboratory prototypes, have for the first time tasted what it's like to roll around the real volcanic desert, where the sand and rocks are electrically conductive and magnetic. It was a valuable experience not only for robots, but also for us people. We often worked for 10 hours without proper offices, food and toilets. If there was a problem, we solved it. When another arose, we solved it as well and so on. Eventually, this iterative process has shifted both “systems” – both robots and engineers – to greater robustness.
Base at Etna during Etna field test 2017
However, the last year's campaign was only a beta preparation. The real demo supervised by rigorous project reviewers from the management of DLR, Helmholtz Association and ESA, with clearly defined tasks that robot had to complete, took place just now – in June and July of this year. We operated in the same spot, this time, however, for four weeks. This year, we transported virtually all our laboratory to Etna in several cargo containers, and we arrived in the assemblage of about 20 robotics engineers. It may sound like a lot, but it was necessary to be in a team, where we can fix virtually any defect that can arise in such a complex system.
In addition to us, robotics engineers, our colleagues taking care of the landing module, communication systems and seismometers also took part in the test. All together, we were about 40 people. This time, working conditions were more acceptable thanks to a better container for workstation placement. Nevertheless, the conditions were still harsh. In cold and windy weather, in tight conditions, with few computers, unstable power supply, rarefied atmosphere, volcanic vapours and vapours from gasoline generators, we “slogged away” even twelve hours a day. All together, we had three weekend days during four weeks.
I think I can say for the entire group of mobile robotics that everyone was operating far beyond their comfort zone and without any external motivation to use the limited time on Etna to improve all possible aspects of our robotic systems. But once again, it could be seen on the results: our robots not only managed to do all the sequences required to demonstrate the feasibility of the robotic lunar seismological mission, but we also successfully performed a number of self-assigned bonus missions – long travers, sample collection and return, multi-robot collaboration. So, on our part, the “trip” to Etna, as well as the whole ROBEX project, was very successful.
My responsibility on Etna during both years was the calibration of cameras, robot localisation control and other support activities.
Etna eruption during Etna field test 2017
M. B.: Do you have any possibility to influence the subject of the research? What would you prefer to work on?
M. S.: The organisational structure at DLR's Institute of Robotics is surprisingly flat. We use the Bavarian motto “Wer ko, der ko / Wer kann, der kann”, which can be translated as “If you've got it, flaunt it”. In practice, this means that if, for example, the head of the Institute has some idea of how something should be done, but a student will do it differently and demonstrate that this way is better, the final product will be done in the way the student did it and the head of the Institute accepts it. This flatness in organisational structures and the focus on objective quality of solutions has provided our institute with a reputation of the world robotic capacity and already brought many publications, patents, projects, and contracts. When formulating the strategy of our institute, it is one of our basic characteristics, which we do not even plan to surrender in case we will reorganise the structures.
I can say that I have the possibility to influence research and I do influence it. I do work on what I prefer to work on the most.
View of Etna from the plane
M. B.: What are your plans for the future?
M. S.: I think that from a career point of view, I find myself in a good life situation. Unlike younger colleagues, I do not have to spend time writing a dissertation, because I have already done that. Unlike older colleagues, I do not have family nor children, so I can fully focus on science, project management and routine engineering.
I have experience with projects that have finished software solutions in the end and with such that are connected to European space ambitions. In addition, I have spent 40 days on external system testing.
I would like to invest my efforts into fulfilling one of the great dreams of mankind for at least past hundred years: the permanent settlement of some of the extraterrestrial objects of the solar system. It is a great collective ambition, but I think that with the right ideas, funding, management, technology, and above all the enthusiasm and co-responsibility, we will be able to achieve it.
Interview prepared and published by: Marta Bartošovičová, NCP S&T within SCSTI
Photos provided by: Ing. Michal Smíšek, PhD.
Young Slovak scientist Ing. Michal Smíšek, PhD., who works at the German Aerospace Centre (DLR), will be the guest of Science in the CENTRE – Summer special on Thursday 27 July 2017 at 5PM. He will present the topic Robots in the service of the solar system exploration in the Slovak Centre of Scientific and Technical Information, Lamačská cesta 8/A in Bratislava.
The summer special of the July's Science in the CENTRE in Bratislava is also one of the accompanying events of the European Researchers' Night 2017.Back to top