Building the world's fastest humanoid and turning Munich into a hub for innovation in Robotics.

Model-free energy-efficient locomotion of legged robots through physics-informed reinforcement learning.

Authored seminal work on braided electromagnets and invented a novel braided linear elastic actuator, resulting in a patent.

Modified a simulation of the JenaFox robot to train a learned controller through Reinforcement Learning:- Achieved efficient running and velocity tracking- Adapted and tested the controller on robotic hardware- Resulted in successful real-world locomotion

My responsibilities included the design of power and control PCBs for legged robots and the Implementation of a neuromorphic controller for a biped robot.

My goal was to enhance the teleoperation and whole-body control of DLR's humanoid robot, Justin. In this project, I:- Implemented control algorithms for bilateral teleoperation with haptic feedback- Developed the necessary communication protocols- Created a Unity-based user interface that integrates Justin's existing technology

My thesis focused on developing the Motion CANVAAS haptic input device, designed for adaptable teleoperation control architecture for humanoid robots.

Developed and optimized robot control and haptic rendering algorithms for LWR-based teleoperation systems.

Development and construction of an autonomous vehicle for the use of a humanoid robot "Roboy". My main responsibilities were the development of PCBs in Eagle, the CAD Model of the vehicle, its assembly and the implementation of classical motor control systems.

Modeling of business revenues for incoming companies to the german market.