Provably Safe Reinforcement Learning via Action Projection using Reachability Analysis and Polynomial Zonotopes

Understand

While reinforcement learning produces very promising results for many applications, its main disadvantage is the lack of safety guarantees, which prevents its use in safety-critical systems.

• In this work, we address this issue by a safety shield for nonlinear continuous systems that solve reach-avoid tasks.
• Our safety shield prevents applying potentially unsafe actions from a reinforcement learning agent by projecting the proposed action to the closest safe action.
• This approach is called action projection and is implemented via mixed-integer optimization.