Apply before April 30, 2025

Background

Uncertainty poses a significant challenge in modern control systems, particularly for real-time networked applications. This research project explores cutting-edge online learning techniques using distributed Gaussian Processes (GPs) to enhance the performance of networked control systems under uncertainty [1,2]. You will design a computationally efficient, real-time learning framework to enable adaptive and robust control, focusing on Euler-Lagrange systems with unknown dynamics. The approach integrates a data-driven model, accounts for communication delays, and adapts to dynamic network conditions [3]. Stability will be rigorously analyzed using the Lyapunov theorem, with the proposed controller validated through simulations and experiments on FRANKA robots.

Your Tasks

• Develop an online GP-based learning framework for networked control systems
• Investigate real-time inference methods for GPs to ensure computational efficiency
• Tackle challenges such as communication delays and distributed learning
• Implement and evaluate the framework in simulation and on real robotic hardware

Requirements

• Highly self-motivated and self-independent
• Solid knowledge on Machine Learning, Control Theory and Robotics
• Python and C++ and/or MATLAB programming experience

Application Process

To apply, send your CV, transcript, and supporting documents to Dr. Zewen Yang (zewen.yang(at)tum.de) and Dr. Hamid Sadeghian (hamid.sadeghian(at)tum.de), before 30 April 2025.  

Chair of Robotics and Systems Intelligence (RSI)

Munich Institute of Robotics and Machine Intelligence (MIRMI)

Technical University of Munich

Reference:

[1] Zewen Yang, Xiaobing Dai, Hirche Sandre. "Asynchronous Distributed Gaussian Process Regression." In the Thirty-Ninth Conference on Artificial Intelligence (AAAI), Philadelphia, Pennsylvania, USA, 2025.

[2] Zewen Yang, Songbo Dong, Armin Lederer, Xiaobing Dai, Siyu Chen, Stefan Sosnowski, Georges Hattab, and Sandra Hirche. "Cooperative Learning with Gaussian Processes for Euler-Lagrange Systems Tracking Control under Switching Topologies." In 2024 American Control Conference (ACC), pp. 560-567. IEEE, 2024.

[3] Xiao Chen, Youssef Michel, Hamid Sadeghian, and Sami Haddadin. "Network-aware Shared Autonomy in Bilateral Teleoperation." In 2024 IEEE-RAS 23rd International Conference on Humanoid Robots (Humanoids), pp. 888-894. IEEE, 2024.

Internship/Master thesis | Application deadline: May 20, 2025

Recent advancements in robotics, especially concerning humanoids and quadrupeds, are largely due to the adoption of novel actuator technologies. These technologies are involving high power density BLDC motors with a lower gear ratio, with the aim of maintaining good proprioceptive feedback for control. [1] However, active development of new actuation concepts is ongoing. One of the important directions is an augmentation of such actuators with mechanical springs for storing and releasing energy at the dynamic peaks.

Our work focuses on the development and testing of one such actuator, a version of the Parallel Elastic Actuator. This particular task includes the manufacturing of a simple test-actuator that integrates a backdrivable motor with a spring and a torque sensor connected in parallel. Further, a performance characterization of the motor will be conducted (Bode plot analysis, etc.). Additionally, an impedance controller will be evaluated on such a setup.

 

What you will gain:

• Experience in Modeling and Control of Robotics systems
• Experience building, prototyping
• Best design practices for torque sensor integration in actuators
• Insights into our System Development and access to our community
• BLDC motor control

Requirements from candidates:

• Mechanical Engineering background
• Completed classical control courses (or some project experience in control)
• Any CAD software for the Part designs (such as Solidworks, Fusion 360,etc.)
• Matlab skills
• Basic skills in Electronics
• Plus are:
  • Understanding how Motors work
  • Familiarity with GIT
  • Working skills in Ubuntu operating system

 

To apply, you can send your CV, and short motivation to:

 

Supervisors

M.Sc. Vasilije Rakcevic vasilije.rakcevic(at)tum.de

M.Sc. Tomas Slimak  tomas.slimak(at)tum.de

 

 

 

 

[1] P. M. Wensing, A. Wang, S. Seok, D. Otten, J. Lang and S. Kim, "Proprioceptive Actuator Design in the MIT Cheetah: Impact Mitigation and High-Bandwidth Physical Interaction for Dynamic Legged Robots," in IEEE Transactions on Robotics, vol. 33, no. 3, pp. 509-522, June 2017, doi: 10.1109/TRO.2016.2640183.