The control policies are learned in a physics simulator and then deployed on real robots.
Recently, a safe Bayesian optimization algorithm, called SafeOpt, has been developed, which guarantees that the performance of the system never falls below a critical value; that is, safety is defined based on the performance function.
Our model class is a generalisation of nonlinear mixed-effects (NLME) dynamical systems, the statistical workhorse for many experimental sciences.
We are proceeding towards the age of automation and robotic integration of our production lines [5].