Searcharxiv⌕ Search

arXiv subjects

Marc Thomas

Publications and source records attributed to Marc Thomas.

16 recordsLinked to original sources
arXiv2026

Human-in-the-Loop Testing of AI Agents for Air Traffic Control with a Regulated Assessment Framework

We present a rigorous, human-in-the-loop evaluation framework for assessing the performance of AI agents on the task of Air Traffic Control, grounded in a regulator-certified simulator-based curriculum used for training and testing real-world trainee controllers. By leveraging legally regulated assessments and involving expert human instructors in the evaluation process, our framework enables a more authentic and domain-accurate measurement of AI performance. This work addresses a critical gap in the existing literature: the frequent misalignment between academic representations of Air Traffic Control and the complexities of the actual operational environment. It also lays the foundations for effective future human-machine teaming paradigms by aligning machine performance with human assessment targets.

cs.HC↗
arXiv2026

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

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.

cs.AI↗
arXiv2026

Conditioning Aircraft Trajectory Prediction on Meteorological Data with a Physics-Informed Machine Learning Approach

Accurate aircraft trajectory prediction (TP) in air traffic management systems is confounded by a number of epistemic uncertainties, dominated by uncertain meteorological conditions and operator specific procedures. Handling this uncertainty necessitates the use of probabilistic, machine learned models for generating trajectories. However, the trustworthiness of such models is limited if generated trajectories are not physically plausible. For this reason we propose a physics-informed approach in which aircraft thrust and airspeed are learned from data and are used to condition the existing Base of Aircraft Data (BADA) model, which is physics-based and enforces energy-based constraints on generated trajectories. A set of informative features are identified and used to condition a probabilistic model of aircraft thrust and airspeed, with the proposed scheme demonstrating a 20% improvement in skilfulness across a set of six metrics, compared against a baseline probabilistic model that ignores contextual information such as meteorological conditions.

eess.SY↗
arXiv2026

Fast Surrogate Models for Adaptive Aircraft Trajectory Prediction in En route Airspace

Trajectory prediction (TP) is crucial for ensuring safety and efficiency in modern air traffic management systems. It is, for example, a core component of conflict detection and resolution tools, arrival sequencing algorithms, capacity planning, as well as several future concepts. However, TP accuracy within operational systems is hampered by a range of epistemic uncertainties such as the mass and performance settings of aircraft and the effect of meteorological conditions on aircraft performance. It can also require considerable computational resources. This paper proposes a method for adaptive TP that has two components: first, a fast surrogate TP model based on linear state space models (LSSM)s with an execution time that was 6.7 times lower on average than an implementation of the Base of Aircraft Data (BADA) in Python. It is demonstrated that such models can effectively emulate the BADA aircraft performance model, which is based on the numerical solution of a partial differential equation (PDE), and that the LSSMs can be fitted to trajectories in a dataset of historic flight data. Secondly, the paper proposes an algorithm to assimilate radar observations using particle filtering to adaptively refine TP accuracy. Comparison with baselines using BADA and Kalman filtering demonstrate that the proposed framework improves system identification and state estimation for both climb and descent phases, with 46.3% and 64.7% better estimates for time to top of climb and bottom of descent compared to the best performing benchmark model. In particular, the particle filtering approach provides the flexibility to capture non-linear performance effects including the CAS-Mach transition.

cs.CE↗
arXiv2026

A Probabilistic Digital Twin of UK En Route Airspace for Training and Evaluating AI Agents for Air Traffic Control

This paper presents the first probabilistic Digital Twin of operational en route airspace, developed for the London Area Control Centre. The Digital Twin is intended to support the development and rigorous human-in-the-loop evaluation of AI agents for Air Traffic Control (ATC), providing a virtual representation of real-world airspace that enables safe exploration of higher levels of ATC automation. This paper makes three significant contributions: firstly, we demonstrate how historical and live operational data may be combined with a probabilistic, physics-informed machine learning model of aircraft performance to reproduce real-world traffic scenarios, while accurately reflecting the level of uncertainty inherent in ATC. Secondly, we develop a structured assurance case, following the Trustworthy and Ethical Assurance framework, to provide quantitative evidence for the Digital Twin's accuracy and fidelity. This is crucial to building trust in this novel technology within this safety-critical domain. Thirdly, we describe how the Digital Twin forms a unified environment for agent testing and evaluation. This includes fast-time execution (up to x200 real-time), a standardised Python-based ``gym'' interface that supports a range of AI agent designs, and a suite of quantitative metrics for assessing performance. Crucially, the framework facilitates competency-based assessment of AI agents by qualified Air Traffic Control Officers through a Human Machine Interface. We also outline further applications and future extensions of the Digital Twin architecture.

cs.CE↗
arXiv2025

Probabilistic Simulation of Aircraft Descent via a Physics-Informed Machine Learning Approach

This paper presents a method for generating probabilistic descent trajectories in simulations of real-world airspace. A dataset of 116,066 trajectories harvested from Mode S radar returns in UK airspace was used to train and test the model. Thirteen aircraft types with varying performance characteristics were investigated. It was found that the error in the mean prediction of time to reach the bottom of descent for the proposed method was less than that of the the Base of Aircraft Data (BADA) model by a factor of 10. Furthermore, the method was capable of generating a range of trajectories that were similar to the held out test dataset when analysed in distribution. The proposed method is hybrid, with aircraft drag and calibrated airspeed functions generated probabilistically to parameterise the BADA equations, ensuring the physical plausibility of generated trajectories.

eess.SY↗
arXiv2023

Learning Generative Models for Climbing Aircraft from Radar Data

Accurate trajectory prediction (TP) for climbing aircraft is hampered by the presence of epistemic uncertainties concerning aircraft operation, which can lead to significant misspecification between predicted and observed trajectories. This paper proposes a generative model for climbing aircraft in which the standard Base of Aircraft Data (BADA) model is enriched by a functional correction to the thrust that is learned from data. The method offers three features: predictions of the arrival time with 26.7% less error when compared to BADA; generated trajectories that are realistic when compared to test data; and a means of computing confidence bounds for minimal computational cost.

eess.SY↗
arXiv2023

Context-Aware Generative Models for Prediction of Aircraft Ground Tracks

Trajectory prediction (TP) plays an important role in supporting the decision-making of Air Traffic Controllers (ATCOs). Traditional TP methods are deterministic and physics-based, with parameters that are calibrated using aircraft surveillance data harvested across the world. These models are, therefore, agnostic to the intentions of the pilots and ATCOs, which can have a significant effect on the observed trajectory, particularly in the lateral plane. This work proposes a generative method for lateral TP, using probabilistic machine learning to model the effect of the epistemic uncertainty arising from the unknown effect of pilot behaviour and ATCO intentions. The models are trained to be specific to a particular sector, allowing local procedures such as coordinated entry and exit points to be modelled. A dataset comprising a week's worth of aircraft surveillance data, passing through a busy sector of the United Kingdom's upper airspace, was used to train and test the models. Specifically, a piecewise linear model was used as a functional, low-dimensional representation of the ground tracks, with its control points determined by a generative model conditioned on partial context. It was found that, of the investigated models, a Bayesian Neural Network using the Laplace approximation was able to generate the most plausible trajectories in order to emulate the flow of traffic through the sector.

eess.SY↗
arXiv2022

A Probabilistic Model for Aircraft in Climb using Monotonic Functional Gaussian Process Emulators

Ensuring vertical separation is a key means of maintaining safe separation between aircraft in congested airspace. Aircraft trajectories are modelled in the presence of significant epistemic uncertainty, leading to discrepancies between observed trajectories and the predictions of deterministic models, hampering the task of planning to ensure safe separation. In this paper a probabilistic model is presented, for the purpose of emulating the trajectories of aircraft in climb and bounding the uncertainty of the predicted trajectory. A monotonic, functional representation exploits the spatio-temporal correlations in the radar observations. Through the use of Gaussian Process Emulators, features that parameterise the climb are mapped directly to functional outputs, providing a fast approximation, while ensuring that the resulting trajectory is monotonic. The model was applied as a probabilistic digital twin for aircraft in climb and baselined against BADA, a deterministic model widely used in industry. When applied to an unseen test dataset, the probabilistic model was found to provide a mean prediction that was 21% more accurate, with a 34% sharper forecast.

cs.CE↗
arXiv2020

Effect of ageing on the properties of the W-containing IRIS-TiAl alloy

The effects of ageing at 800 $^\circ$C on the properties of the IRIS alloy (Ti$_{49.9}$Al$_{48}$W$_2$B$_{0.1}$) are studied. The initial microstructure of this alloy densified by Spark Plasma Sintering (SPS) is mainly composed of lamellar colonies which are surrounded by $\gamma$ grains. The evolutions of the alloy strength and creep resistance resulting from this ageing treatment are measured by the related mechanical tests. The microstructural changes are investigated by scanning and transmission electron microscopies and by X-ray diffraction. The main structural evolutions consist in a shrinkage of the lamellar areas and in a precipitation of $\beta_0$ phase, which is accompanied by a moderate segregation of tungsten and a decrease of the $\alpha_2$ lamellar width. However, these evolutions are relatively limited and the microstructural stability is found to result mainly from the low diffusivity of tungsten. Conversely, a moderate effect of this ageing treatment on mechanical properties, at room and high temperatures, is measured. Such experimental results are interpreted and discussed in terms of the microstructural evolutions and of the deformation mechanisms which are activated at different temperatures under various solicitations.

cond-mat.mtrl-sci↗
arXiv2016

Anatomy of the Inert Two Higgs Doublet Model in the light of the LHC and non-LHC Dark Matter Searches

The inert Two Higgs Doublet Model (i2HDM) is a theoretically well-motivated example of a minimal consistent Dark Matter(DM) model which provides mono-jet, mono-Z, mono-Higgs and Vector-Boson-Fusion+Missing Transverse Momentum signatures at the LHC, complemented by signals in direct and indirect DM search experiments. In this paper we have performed a detailed analysis of the constraints in the full 5D parameter space of the i2HDM, coming from perturbativity, unitarity, electroweak precision data, Higgs data from LHC, DM relic density, direct/indirect DM detection and LHC mono-jet analysis, as well as implications of experimental LHC studies on disappearing charged tracks relevant to high DM mass region. We demonstrate the complementarity of the above constraints and present projections for future LHC data and direct DM detection experiments to probe further i2HDM parameter space. The model is implemented into the CalcHEP and micrOMEGAs packages, which are publicly available at the HEPMDB database, and is ready for a further exploration in the context of the LHC, relic density and DM direct detection.

hep-ph↗
arXiv2016

Dark Matter characterization at the LHC in the Effective Field Theory approach

We have studied the complete set of dimension 5 and dimension 6 effective operators involving the interaction of scalar, fermion and vector Dark Matter (DM) with SM quarks and gluons, to explore the possibility to distinguish these operators and characterise the spin of DM at the LHC. We have found that three factors - the effective dimension of the operator, the structure of the SM part of the operator and the parton densities of the SM particles connected to the operator - uniquely define the shape of the (unobservable) invariant mass distribution of the DM pair and, consequently, the shape of the (observable) MET distribution related to it. Using $\chi^2$ analysis, we found that at the LHC, with a luminosity of 300 fb$^{-1}$, certain classes of EFT operators can be distinguished from each other. Hence, since DM spin is partly correlated with the factors defining the shape of MET, the LHC can potentially shed a light also on DM spin. We have also observed a drastic difference in the efficiencies (up to two orders of magnitude) for large MET cuts scenarios with different DM spin, thus indicating that the DM discovery potential strongly depends on it. The study we perform here can be applied more generally than within the EFT paradigm, where the DM mediator is not produced on-the-mass-shell, such as the case of t-channel mediator or mediator with mass below $2M_{DM}$, where the invariant mass of the DM pair is not fixed.

hep-ph↗
arXiv2015

Towards model-independent exclusion of light Stops

Understanding the extent to which experimental searches are sensitive to Light Stops (LST) scenarios is essential to resolve questions about naturalness, electroweak baryogenesis and Dark Matter. In this paper we characterize the reach on LST scenarios in two ways. We extend experimental searches to cover specific gaps in the LST parameter space, showing for the first time that assuming a single decay channel one can exclude the region of $m_{\tilde{t}}<m_{top}$, which in its turn excludes electroweak baryogenesis in MSSM. Also, we explore the extent to which searches are weakened in a more generic scenario when more than one decay channel takes place, even after their combination. This study highlights the need for a more comprehensive exploration of the LST parameter space.

hep-ph↗
arXiv2015

Simplified Models for Dark Matter Searches at the LHC

This document outlines a set of simplified models for dark matter and its interactions with Standard Model particles. It is intended to summarize the main characteristics that these simplified models have when applied to dark matter searches at the LHC, and to provide a number of useful expressions for reference. The list of models includes both s-channel and t-channel scenarios. For s-channel, spin-0 and spin-1 mediation is discussed, and also realizations where the Higgs particle provides a portal between the dark and visible sectors. The guiding principles underpinning the proposed simplified models are spelled out, and some suggestions for implementation are presented.

hep-ph↗
arXiv2013

Light sfermion interplay in the 125 GeV MSSM Higgs production and decay at the LHC

We study the effects from light sfermions on the lightest Higgs boson production and decay at the Large Hadron Collider (LHC) within the Minimal Supersymmetric Standard Model (MSSM). We find that the scenario with light coloured sfermions -- stops or sbottoms -- has the potential to explain a non-universal alteration, as hinted by LHC data, of the gluon-gluon Fusion ($\mu_{ggF}$) with respect to the Vector Boson Fusion ($\mu_{VBF}$) event rates and, in particular, can predict $\frac{\mu_{VBF}}{\mu_{ggF}}>1$ for all Higgs boson decay channels in large areas of the parameter space. We also find that the scenario with a light stop/sbottom can be complemented by the scenario in which the total Higgs width, $\Gamma_{\rm tot}$, is reduced due to a suppressed Yukawa coupling $Y_b$. In this case, the reduction of the Higgs production rates in the $ggF$ process which occurs in the maximal mixing scenario is compensated by the reduction of the $H\to b\bar{b}$ partial decay width, the largest component of $\Gamma_{\rm tot}$. Furthermore, we highlight the fact that, in the light stop/sbottom scenario, event rates with the Higgs boson decaying to a $b\bar{b}$ final state are predicted to be essentially below unity. Therefore, during the future LHC runs, the measurement of $h\to b\bar{b}$ final states is a matter of special importance. Amongst all viable MSSM configurations that we study (including revisiting a light stau solution), we emphasise most the scenario with a light stop, as the latter is also motivated by Dark Matter and Electro-Weak baryogenesis. We also perform fits of the MSSM against the LHC data, emphasising the fact that in most cases these are better than for the SM.

hep-ph↗