Run-Time Safety Monitoring of Neural-Network-Enabled Dynamical Systems

TL;DR

This paper introduces a run-time safety monitoring method for neural-network-enabled dynamical systems using an interval observer with neural network components, ensuring real-time state bounds and safety guarantees.

Contribution

It develops a novel interval observer framework with neural networks for real-time safety monitoring of systems with embedded neural components, formulated via linear programming.

Findings

Successfully bounds system states in real-time

Validated on adaptive cruise control system

Demonstrates effective safety monitoring in simulations

Abstract

Complex dynamical systems rely on the correct deployment and operation of numerous components, with state-of-the-art methods relying on learning-enabled components in various stages of modeling, sensing, and control at both offline and online levels. This paper addresses the run-time safety monitoring problem of dynamical systems embedded with neural network components. A run-time safety state estimator in the form of an interval observer is developed to construct lower-bound and upper-bound of system state trajectories in run time. The developed run-time safety state estimator consists of two auxiliary neural networks derived from the neural network embedded in dynamical systems, and observer gains to ensure the positivity, namely the ability of estimator to bound the system state in run time, and the convergence of the corresponding error dynamics. The design procedure is formulated in terms of a family of linear programming feasibility problems. The developed method is illustrated by a numerical example and is validated with evaluations on an adaptive cruise control system.