We are Andrina Frischknecht-Grimm, Shubham Gupta, Zeno Hamers, Deepana Ishtaweera and Robert Jomar Malate and we designed, built and controlled an articulated soft robotic hand from scratch! All repositories that were used and developed during this project can be found in the RWR-GroupB GitHub organisation. This project was conducted as part of the Real-World-Robotics course at ETH Zürich, an initiative from prof. Dr. R. Katzschmann and the Soft Robotics Lab. This README file will give you an overview of the results obtained in this project and the respective GitHub directories. A detailed overview of the robotic hand, our methodology and the final results can be found in our paper: The M&M Hand: a modular and minimal hand for four-fingered dexterous manipulations.
| Simulation | On Physical Robot |
|---|---|
![]() |
![]() |
![]() |
![]() |
| Simulation | On Physical Robot |
|---|---|
![]() |
![]() |
![]() |
![]() |
For the design of the robotic hand, Fusion 360 was used. MuJoCo was used in combination with IsaacGym to simulate the robotic hand in a virtual environment for reinforcement learning purposes.










