1. BaSaMa & HiWi
    2.  

Robotic Cable Manipulation for Automated Disassembly and Recycling

Institut
TUM Munich Institute of Robotics and Machine Intelligence (Institut)
Typ
Masterarbeit / HiWi-Tätigkeit /
Inhalt
experimentell / theoretisch /  
Beschreibung

We are looking for motivated students to work on robotic manipulation of electrical cables in the context of automated recycling. The project aims to develop perception, planning, and control methods that allow a robot to detect, grasp, separate, and sort partially entangled cables on real hardware.

The work will be embedded in an experimental ROS 2-based robot setup and can be shaped either as a research-oriented HiWi position, a Studienarbeit, or a Master’s thesis depending on the student’s background and expected scope.

The main project direction includes:

  • 3D cable perception from RGB-D / point cloud data, including cable segmentation, centerline extraction, endpoint detection, and grasp-point or cut-point proposal.
  • Robotic manipulation planning for deformable cables, including collision-aware motion planning, joint-limit handling, and tension-aware execution.
  • Task-level strategy design for multi-step cable sorting, for example using a state machine or behavior tree for repeated grasping, pulling, regrasping, separating, and placing.
  • Force- and tension-aware manipulation skills for safe cable interaction, including tension-limited pulling, compliant execution, and failure recovery.
  • ROS 2 system integration of perception, planning, control, and sensing modules on real robot hardware.

For a HiWi position, the focus is mainly on system implementation, data collection, ROS 2 integration, benchmarking, and support for real-robot experiments.

For a Studienarbeit or Master’s thesis, the topic can be made more research-oriented. Possible directions include interactive cable parameter identification, physics-based or learned cable dynamics modeling, manipulation skills learning policy, or tension-aware manipulation control for cable separation.

 

Voraussetzungen

Candidate Profile

The ideal candidate has experience or interest in robotics, ROS 2, Python/C++, point cloud processing, motion planning, control, or machine learning. Prior experience with real robot hardware is helpful but not strictly required.

What We Offer

Work on real robotic platforms within an active European research project on automated cable recycling. Opportunity to contribute to publishable results targeting major robotics venues.

How to apply

Please send application with the subject (your intended position and topics) including your CV, transcript to Mr. Li, yu.li(at)tum.de and Dr. Sadeghian, hamid.sadeghian(at)tum.de.

Reference

  • Zhang X, Domae Y, Wan W, et al. A closed-loop bin picking system for entangled wire harnesses using bimanual and dynamic manipulation[J]. Robotics and Computer-Integrated Manufacturing, 2024, 86: 102670.
  • O. Holešovský, R. Škoviera and V. Hlaváč, "Interactive Robotic Moving Cable Segmentation by Motion Correlation," in IEEE Robotics and Automation Letters, vol. 10, no. 7, pp. 7420-7427, July 2025, doi: 10.1109/LRA.2025.3574960.
  • Süberkrüb F, Laezza R, Karayiannidis Y. Feel the tension: Manipulation of deformable linear objects in environments with fixtures using force information[C]//2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2022: 11216-11222.
  • Bartholdt, Max, et al. "A parameter identification method for static cosserat rod models: Application to soft material actuators with exteroceptive sensors." 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2021.
  • Yang Y, Stork J A, Stoyanov T. Learning differentiable dynamics models for shape control of deformable linear objects[J]. Robotics and Autonomous Systems, 2022, 158: 104258.
  • Zhang K, Li B, Hauser K, et al. Particle-grid neural dynamics for learning deformable object models from rgb-d videos[J]. arXiv preprint arXiv:2506.15680, 2025.
Möglicher Beginn
sofort
Kontakt
Yu Li, M.Sc.
yu.litum.de