GoSafe: Globally Optimal Safe Robot Learning

TL;DR

GoSafe extends safe Bayesian optimization for robotic policy learning by exploring outside initial safe zones, using backup controllers to ensure safety and convergence to the global optimum, validated through hardware experiments.

Contribution

It introduces a method to explore beyond initial safe regions in Bayesian optimization for robotics, ensuring safety and global optimality.

Findings

Successfully explores outside initial safe zones

Guarantees convergence to the global optimum

Validated through hardware experiments

Abstract

When learning policies for robotic systems from data, safety is a major concern, as violation of safety constraints may cause hardware damage. SafeOpt is an efficient Bayesian optimization (BO) algorithm that can learn policies while guaranteeing safety with high probability. However, its search space is limited to an initially given safe region. We extend this method by exploring outside the initial safe area while still guaranteeing safety with high probability. This is achieved by learning a set of initial conditions from which we can recover safely using a learned backup controller in case of a potential failure. We derive conditions for guaranteed convergence to the global optimum and validate GoSafe in hardware experiments.