Working Student (m/f/d) Contact-Rich Manipulation and RL

About the role

We are looking for a Working Student (m/f/d) Contact-Rich Manipulation and RL. In this role, you will prototype and benchmark classical and learning-based approaches to contact-rich manipulation tasks, including insertion, assembly, and surface following, across simulation and real robot hardware.

Your Responsibilities

• Approach Survey: Survey recent literature on contact-rich manipulation and benchmark classical and learning-based approaches against each other on tasks such as insertion, assembly, and peg-in-hole.
• Force/Torque Feedback: Explore force and torque feedback strategies for contact state estimation and reactive control.
• Simulation Prototyping: Prototype and compare RL, MPC, and imitation learning pipelines in simulation, including learning from demonstration through behavior cloning and DAgger.
• Transfer Investigation: Investigate sim-to-real transfer through domain randomization, system identification, and reality gap analysis.
• Hardware Experiments: Run experiments on real robot hardware and systematically document which behaviors transfer from simulation and which do not.

Essential Skills

• Academic Background: Currently enrolled in a Master's programme in Robotics, Computer Science, or Mechanical Engineering.
• Technical Foundation: Solid foundation in at least one of robot control, motion planning, or machine learning, with enough depth to read, implement, and critically evaluate approaches from literature.
• Python Proficiency: Working proficiency in Python for prototyping, scripting simulation pipelines, and running experiments.
• Simulation: Ability to set up and run robot manipulation experiments in simulation environments such as MuJoCo or similar.
• Literature Engagement: Ability to read recent research papers, extract implementable ideas, and compare approaches in a structured and reproducible way.

Beneficial Skills

• Force/Torque Sensing: Familiarity with force and torque feedback for contact state estimation or reactive control.
• ROS2: Basic familiarity with ROS2 for hardware integration and experiment execution.
• Real Hardware: Prior experience running experiments on robot hardware, including in a lab or course context.

What we offer

• Practical learning opportunities to complement your studies.
• Dynamic high-tech company combined with financial soundness and world class investors.
• Join an interdisciplinary, international team with 60+ different nationalities in a collaborative work environment.
• Corporate Benefits Program that covers health, mobility and learning with 100 € net per month.
• Modern office facilities with a rooftop terrace overlooking Munich, free drinks & fruits, and regular company events contribute to a good working environment.