Robot Control (RoCo)

Controlling a robot, e.g. a robot arm or a humanoid, is witnessing a paradigm shift by combining physical simulation with machine learning, typically reinforcement learning to solve a large repertoire of challenges that previously have been impossible to solve. In such a scenario the simulator provides a model or prior for the robot to interact and learn before its deployment to the real world. This field of simulation-based reinforcement learning (SiRL) addresses many interesting issues such as motion control, motion imitation, and system identification in advanced autonomous robotics manipulation and navigation, prediction, detection or mimicking of human motion, that may benefit sectors like health-care, industrial manufacturing and entertainment. Many more examples exist. This course will prepare students for working with this world-changing technology.

This course is ideal for students who wish to work with robotics, simulation and practical reinforcement learning in their master's thesis. It is also helpful  for students who want to work with character animation or content creation for XR.

Students are expected to solve weekly practical exercises mostly programming with python, but some C++ could be involved if preferred, as well as some lab-space work as well. Students are expected to be self-driven in getting the software to work on their own devices, like installing an auto differentiation framework and making it run. Students could be exposed to working with vision or motion capture data.