veriFIRE: Verifying an Industrial, Learning-Based Wildfire Detection System

TL;DR

This paper discusses the verification of a real-world wildfire detection system that uses deep neural networks on an airborne platform, aiming to improve its reliability in safety-critical scenarios.

Contribution

It introduces veriFIRE, a collaborative effort to verify a complex, learning-based wildfire detection system, bridging academic verification tools with industrial safety applications.

Findings

Verification of system consistency achieved

Insights into neural network robustness in wildfire detection

Progress towards integrating verification in real-world safety systems

Abstract

In this short paper, we present our ongoing work on the veriFIRE project -- a collaboration between industry and academia, aimed at using verification for increasing the reliability of a real-world, safety-critical system. The system we target is an airborne platform for wildfire detection, which incorporates two deep neural networks. We describe the system and its properties of interest, and discuss our attempts to verify the system's consistency, i.e., its ability to continue and correctly classify a given input, even if the wildfire it describes increases in intensity. We regard this work as a step towards the incorporation of academic-oriented verification tools into real-world systems of interest.