The Underactuated Lab at #DFKI #RoboticsInnovationCenter is committed to pushing the limits of our #robots to achieve animal-like physical/athletic intelligence. Our lab generally hosts a group of exchange students every summer with diverse backgrounds from all over Europe with a shared vision of developing athletic intelligence for underactuated robotic systems (for example, the last year’s proceedings/videos can be found here). The Summer of Underactuation 2023 event continues this tradition from last year and intends to highlight the scientific work of the exchange students working with the underactuated lab to everyone working at DFKI-RIC. The student topics included studying monkey-like brachiation movements, parkour with (monopod) hopping leg, the latest insights into reinforcement learning for achieving dynamic movements, controllers for floating robots in space and in water, and understanding hybrid rovers. This video provides a summary of student interviews during the summer of underactuation event which touches upon their topics, future goals and their experience of working at DFKI Robotics Innovation Center and life in Bremen. The student talks can be found here:
https://robotik.dfki-bremen.de/en/research/research-facilities-labs/underactuated-lab/
https://dfki-ric-underactuated-lab.github.i/summer_of_underactuation_2023/
https://robotik.dfki-bremen.de/en/startpage
If you are interested to work in this area and have the right educational background, feel free to contact us for next summer!
The Underactuated Robotics lab brings together different applied and fundamental research teams at DFKI to perform groundbreaking research in the field of underactuated robotics for dynamic robots such as brachiation robots or legged robots. Every year we host a group of students from all over the world at our lab for internships and master thesis.
This year at the summer of Underactuation, we had nine students who presented their work and most of them go forward and also publish their work in multiple conferences and journal publications. My name is Shubham Vyas. I’m a space robotics researcher here at the DFKI, and I also lead the underactuated robotics lab.
I’m working on developing control strategies to reduce the vertical motion of the robots and their bodies when they navigate rigid surfaces. I’m working on a control system for dynamic Parkour motion on the hopping leg, and the hopping leg is hop around in a circle with the gantry
And I obstruct its way with some obstacles and the goal is to overtake those and get to the finish. At DFKI, my master thesis is about optimal control for satellites, so steering satellites as efficient as possible without hitting any obstacles. I’m doing my master’s thesis here DFKI.
The project is about modeling the parallel continuum robot and to see how those flexible structure deform under forces. I am currently working on a symbolic library for code generation of the kinematics and dynamics of open chain robots. Despite all the amazing results which have been done in the last decades
And also some more recent advancements in bipedal walking, there’s still so much which we don’t really comprehend and which we don’t really work with right now, especially regarding hydrodynamic effects for underwater robotics. So I just find it exciting to feel like in a few years maybe we already made great advancements
In this field and I maybe can be part of it too. Nowadays, it’s common to see robots that can move around using wheels or robots that can walk around as well. They became common, but it’s quite rare to see a robot that can walk like a monkey and brachiate across ladder bars
That’s what makes me passionate and fascinated by this project. I’m fascinated about my project because double pendulum system is highly non-linear and chaotic system It’s very hard for classical control method, but I’m trying to solve it with reinforcement based control, state of the art control method
The most fascinating part about my work is that I can use my mathematical knowledge to combine it with real world systems, which is the real motivation of studying math and applying it in real world scenarios. So, the best day
Of my time here at DFKI was when DFKI sent me out to the European Space Agency in the Netherlands, where I was able to test my system. And then, of course, when the moment the controller showed it is working the first time. That was, of course, the best day.
My best day DFKI was when I signed my contract as a research assistant because it was open day at DFKI and there were all these robots around me and I thought it was quite amazing to see them. And I felt at home. Bremen is a very bike friendly city.
I can do everything by bike and for my free time. There are also a lot of opportunities. I found a sports club. I can go swimming at different places. Also with the burger park, that’s a really big forest. So running there is perfect Bremen doesn’t really lack anything.
It’s a great city to have, it’s really a surprisingly vibrant and lively city for it being rather small. So, it’s pretty exciting to live here.