GoSafeOpt: Scalable safe exploration for global optimization of dynamical systems

Learning optimal control policies directly on physical systems is challenging. Even a single failure can lead to costly hardware damage. Most existing model-free learning methods that guarantee safety, i.e., no failures, during exploration are limited local optima. This work proposes GoSafeOpt as the first provably safe and algorithm safely discover globally for with high-dimensional state space. We demonstrate superiority of over competing in simulation experiments robot arm.