A talk from the AI/ML session at RSECon23 at Swansea University, on 2023-09-05.
Programme link: virtual.oxford...
Slides: drive.google.c...
Abstract:
Project Bluebird is a multi-institutional research project aiming to apply human-centered AI to the complex task of Air Traffic Control (ATC). The project, an EPSRC funded Prosperity partnership between NATS, The Alan Turing Institute and the University of Exeter, covers a range of technical challenges: from developing physics-based probabilistic models of aircraft trajectories, creating machine learning systems that collaborate with humans to make ATC decisions, and to the design of tools that promote safe, explainable, and trustworthy use of AI in ATC.
Given the ambitious aims of the project, and the extensive range of domain specific knowledge required to achieve these aims, delivering a quality software solution is a major challenge. Research software engineers (RSEs) have both expertise in professional software development practices and domain specific knowledge enabling them to produce reliable, efficient and maintainable software for research.
In this talk we will discuss the latest updates from the project and the challenges our RSE team have faced working on such a complex, multi-disciplinary project. We will present approaches to developing research software that we have successfully utilised to date, and the benefits that a diversely experienced team of RSEs can bring when creating cutting-edge, AI solutions.
16 окт 2024
