Robotics Internship

Robotics Internship:
- Are you interested in controlling complex and fast dynamic systems?- Have you already worked on projects involving robotics?- Do you like interacting with an international team while working independently?

Sony AI is looking for a talented Robotics Intern to join our Zurich-based team for a 6-month internship (early start dates preferred).

Sony AI is dedicated to using AI to help unleash human imagination and creativity. We believe in researching and developing AI techniques that empower the imagination and creativity of artists, makers and creators around the world. Our aim is to advance AI so that it augments - and works in harmony with - humans to benefit society.

Role:
The intern will work in a diverse team to support robotics, AI and perception researchers by improving the safety and robustness of the robot in use. This will involve estimating the distance between different rigid bodies as well as the behavior of the physical actuators of the robot accordingly. The primary goal of this work is to achieve safe and robust robot control while minimizing the gap between simulation and real-world, thereby maximizing the utility of any policies learnt in simulation.

**Responsibilities**:
- Fast/ real-time estimation of distance between bodies in simulation.- Simulating rigid/soft body dynamics in the presence of external and internal forces.- Ensuring the smooth transfer of policies from simulation to physical robots.- Fast policy adaptation based on safety metrics.- Developing a simulation environment for training intelligent robotic systems which encapsulate internal and external physical constraints.

Competencies:
- Creative problem solver, to be able to think on their feet and come up with unique solutions based on an in-depth state-of-art literature review.- Software development skills (C++, Python).- Strong mathematical and robotics fundamentals, including linear algebra and calculus.- Mathematical modelling of physical systems- Modelling of physical systems using machine learning and data-driven approaches- Organized and detailed-orientated mindset.- Motivated and energetic- Confident in English- Bonus points for:

- Experience with game/physics engine development.
- Publications in controlling robotics systems.
- Experience with different simulation environments like Bullet, Gazebo, Unity, MuJoCo, Isaac Sim, Drake.
- Experience of differentiable simulators and physical modelling
- Familiarity with transferring policies from simulation to the real world
- Experience with ROS2