Safe and Reliable Training of Learning-Based Aerospace Controllers
This addresses safety-critical issues in aerospace and similar domains, though it appears incremental with multiple methods presented.
The paper tackles the problem of ensuring safety in deep reinforcement learning controllers for aerospace systems by introducing a design-for-verification approach using k-induction and neural Lyapunov Barrier certificates, with results demonstrated through a case study.
In recent years, deep reinforcement learning (DRL) approaches have generated highly successful controllers for a myriad of complex domains. However, the opaque nature of these models limits their applicability in aerospace systems and safety-critical domains, in which a single mistake can have dire consequences. In this paper, we present novel advancements in both the training and verification of DRL controllers, which can help ensure their safe behavior. We showcase a design-for-verification approach utilizing k-induction and demonstrate its use in verifying liveness properties. In addition, we also give a brief overview of neural Lyapunov Barrier certificates and summarize their capabilities on a case study. Finally, we describe several other novel reachability-based approaches which, despite failing to provide guarantees of interest, could be effective for verification of other DRL systems, and could be of further interest to the community.