Analysis of alerting performance for detect and avoid of unmanned aircraft systems

Abstract

As the use of Unmanned Air Systems (UAS) increases, the need to address integration challenges within the National Airspace (NAS) becomes increasingly more important. To help ensure safe operation of UAS into the NAS, minimum Detect and Avoid (DAA) standards for class 3–5 UAS are being developed for operations within class D, E, and G airspace. DAA for UAS is intended to perform a similar function as that of a manned pilot's requirement to see and avoid other aircraft. In the case of DAA, surveillance sensor information is used to provide the UAS and remotely located Pilot in Command (PIC) with the situational awareness necessary to assist in remaining well-clear from other aircraft. While DAA poses several unique challenges, of particular focus for this research, is developing quantifiable alerting standards by which UAS shall comply when operating near proximate aircraft. This research is a detailed evaluation of the various aspects of alerting when intruder aircraft are predicted to violate well-clear. Missed, late, and early alerts are all realities, with the potential to negatively impact DAA performance and safety, and thus need to be understood, characterized, and mitigated against. Analysis of Alerting Performance for Detect and Avoid of Unmanned Aircraft Systems explores and characterizes several aspects of DAA alerting, including surveillance and navigation errors, and encounter geometry. This paper places emphasis on the impact that several currently proposed required navigation/surveillance sources have on alerting performance; currently proposed required surveillance sensors include Automatic Dependent Surveillance — Broadcast (ADS-B), active surveillance transponder, and airborne radar, with the focus of this research being place on radar and ADS-B. One of the major highlights of this paper is the alerting performance trade-off between desirable (i.e. correct) and undesirable (i.e. late, missed, etc.) alerting characteristics, and the effect that navigation and surveillance errors have on this trade-off. The research will highlight the impact which both radar and ADS-B accuracy have on alerting, as well as provide insight into how encounter geometry, impact DAA alerting performance. This paper outlines the influence these considerations have on alerting algorithm approaches being considered for DAA standards within the NAS. The analysis performed herein will be used to inform and influence alerting standards and requirements, as related to the UAS DAA functionality.