Neural Control and Certificate Repair via Runtime Monitoring

TL;DR

This paper introduces a runtime monitoring framework for certifying and repairing neural network control policies and certificates in unknown system dynamics, enhancing safety and reliability in control tasks.

Contribution

It proposes a novel method for certifying and repairing neural control policies using runtime data, addressing reliability issues in black-box systems.

Findings

Successfully repaired neural policies in autonomous control tasks.

Improved safety rates of neural control policies.

Demonstrated effectiveness on two autonomous systems.

Abstract

Learning-based methods provide a promising approach to solving highly non-linear control tasks that are often challenging for classical control methods. To ensure the satisfaction of a safety property, learning-based methods jointly learn a control policy together with a certificate function for the property. Popular examples include barrier functions for safety and Lyapunov functions for asymptotic stability. While there has been significant progress on learning-based control with certificate functions in the white-box setting, where the correctness of the certificate function can be formally verified, there has been little work on ensuring their reliability in the black-box setting where the system dynamics are unknown. In this work, we consider the problems of certifying and repairing neural network control policies and certificate functions in the black-box setting. We propose a novel framework that utilizes runtime monitoring to detect system behaviors that violate the property of interest under some initially trained neural network policy and certificate. These violating behaviors are used to extract new training data, that is used to re-train the neural network policy and the certificate function and to ultimately repair them. We demonstrate the effectiveness of our approach empirically by using it to repair and to boost the safety rate of neural network policies learned by a state-of-the-art method for learning-based control on two autonomous system control tasks.