Certified ML Object Detection for Surveillance Missions
This work addresses the need for certified and dependable ML-based object detection in surveillance missions, particularly for drone detection, but appears incremental as it focuses on a development process for standard compliance.
The paper tackles the problem of developing a drone detection system with a machine learning object detection component to meet performance objectives and provide evidence for compliance with the ED 324 / ARP 6983 standard, aiming to gain confidence in system dependability.
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.