Certified ML Object Detection for Surveillance Missions

TL;DR

This paper details the development of a drone detection system using machine learning object detection, aiming to meet standard dependability requirements for surveillance applications.

Contribution

It introduces a systematic development process for a ML-based drone detection system aligned with upcoming standard recommendations.

Findings

Achieved performance objectives for drone detection

Provided evidence of system dependability

Aligned development with upcoming standards

Abstract

In this paper, we present a development process of a drone detection system involving a machine learning object detection component. The purpose is to reach acceptable performance objectives and provide sufficient evidences, required by the recommendations (soon to be published) of the ED 324 / ARP 6983 standard, to gain confidence in the dependability of the designed system.