Home // International Journal On Advances in Intelligent Systems, volume 15, numbers 1 and 2, 2022 // View article

A Reference Ontology for Collision Avoidance Systems and Accountability Integrated with DAIDALUS

Authors:
David Martín-Lammerding
José Javier Astrain
Alberto Córdoba

Keywords: Semantic reasoning; ontology; DAIDALUS; UAS; knowledge; conflicts; anti-collision; sensor; embedded; air traffic.

Abstract:
Unmanned Aerial Systems (UASs) will be deployed massively in urban areas to implement logistic and security applications, with lower cost and more flexibility than manned aircraft. An increasing number of UAS missions requires an improvement of their safety by equipping UASs with Collision Avoidance Systems (CASs). CAS implementations require data from the UAS environment to identify conflicts and to perform an avoidance maneuver if required. UAS generates heterogeneous data from multiple sources, like the Flight Control Unit (FCU), the Global Navigation Satellite System (GNSS), a radio receiver, an onboard-camera, etc. However, each CAS implementation represents, processes and stores conflict data in a different way. Therefore, there is a lack of standards that simplify their development and homologation. To solve this situation, we present a reference knowledge model for any CAS for UAS implemented as a novel application ontology, called Dronetology- cas, and its integration with DAIDALUS, a CAS developed by NASA. Dronetology-cas provides a unified semantic representa- tion within an ontology-based triplet store designed to run in a Single Board Computer (SBC). Its semantic model provides advantages, such as interoperability between systems, machine- processable data and the ability to infer new knowledge. It is implemented using semantic web standards, which contribute to simplify an operational safety audit. Additionally, we integrate Dronetology-cas with DAIDALUS and verify it with two scenarios where conflict data from external sources are considered to improve UAS safety.

Pages: 60 to 70

Copyright: Copyright (c) to authors, 2022. Used with permission.

Publication date: June 30, 2022

Published in: journal

ISSN: 1942-2679