A framework for assuring the accuracy and fidelity of an AI-enabled Digital Twin of en route UK airspace

TL;DR

The paper tackles the challenge of assuring accuracy and fidelity for an AI-enabled Digital Twin of en route UK airspace used to train and evaluate AI ATC agents. It introduces the Trustworthy and Ethical Assurance (TEA) framework to construct a modular assurance case that links data pipelines, virtual environment fidelity, probabilistic trajectory prediction, and AI-agent interoperability to emerging ATM and AI guidance. Key contributions include a top-level Goal Claim (G1) and in-depth deep-dives on GPT-5-based synthetic scenario generation and a physics-informed probabilistic trajectory predictor for descending aircraft, with rigorous evidence and assumptions mapping to thresholds. The work provides an extensible template for regulatory-ready assurance of probabilistic Digital Twins in ATM, supporting stakeholder engagement and paving the way for future live-data assimilation and per-agent assurance considerations.

Abstract

Digital Twins combine simulation, operational data and Artificial Intelligence (AI), and have the potential to bring significant benefits across the aviation industry. Project Bluebird, an industry-academic collaboration, has developed a probabilistic Digital Twin of en route UK airspace as an environment for training and testing AI Air Traffic Control (ATC) agents. There is a developing regulatory landscape for this kind of novel technology. Regulatory requirements are expected to be application specific, and may need to be tailored to each specific use case. We draw on emerging guidance for both Digital Twin development and the use of Artificial Intelligence/Machine Learning (AI/ML) in Air Traffic Management (ATM) to present an assurance framework. This framework defines actionable goals and the evidence required to demonstrate that a Digital Twin accurately represents its physical counterpart and also provides sufficient functionality across target use cases. It provides a structured approach for researchers to assess, understand and document the strengths and limitations of the Digital Twin, whilst also identifying areas where fidelity could be improved. Furthermore, it serves as a foundation for engagement with stakeholders and regulators, supporting discussions around the regulatory needs for future applications, and contributing to the emerging guidance through a concrete, working example of a Digital Twin. The framework leverages a methodology known as Trustworthy and Ethical Assurance (TEA) to develop an assurance case. An assurance case is a nested set of structured arguments that provides justified evidence for how a top-level goal has been realised. In this paper we provide an overview of each structured argument and a number of deep dives which elaborate in more detail upon particular arguments, including the required evidence, assumptions and justifications.

Figures (10)

• Figure 1: Context for the assurance case. G=Goal Claim, C=Context.
• Figure 2: Strategies for the assurance case. G=Goal Claim, S=Strategy.
• Figure 3: Abridged argument over the accuracy and fidelity of the data pipeline. S=Strategy, P=Property Claim.
• Figure 4: Abridged argument over accuracy and fidelity of the virtual environment. S=Strategy, P=Property Claim, E=Evidence, A=Assumption, J=Justification. Deep dive is highlighted by the bold lines and arrows.
• Figure 5: GPT-5 solving a benchmark with 8 aircraft successfully: there are no interactions in this scenario. Aircraft are never adjacent, which indicates an interaction. This is the simplified grid representation in which each grid point is 20 NMI. Snapshots show two time units from the start of a scenario. Aircraft are colour coded with their current location shown by the solid circle, their trace by the transparent circles, and their paths by the dashed lines.