Publications

Road safety remains a critical challenge, with human errors such as drowsiness and distraction as contributing factors. Physiological indicators such as Heart Rate (HR) and Respiratory Rate (RR) provide insights into understanding the driver state but are difficult to acquire in real-world driving scenarios due to the constantly changing driver pose and complex ambient environment.

To tackle this challenge, a noncontact cardiorespiratory monitoring system using Frequency Modulated Continuous Wave (FMCW) radar is proposed. Unlike optical or acoustic systems, the use of radar can minimize interference from the ambient environment. However, its relatively low spatial resolution, driver motion, and variability in human Radar Cross Section (RCS) present significant signal processing challenges, especially for target tracking.

To address these issues, a novel target tracking algorithm and a reference-free Signal Quality Index (SQI) based on phase continuity are proposed, aimed at significantly improving the consistency and reliability of target tracking and overall signal quality. A performance evaluation using a driving simulator and synchronized Electrocardiography (ECG) recordings for ground truth demonstrates that our proposed method achieves a mean absolute error (MAE) of 1.65 beats per minute (BPM) for HR, 0.79 respirations per minute (RPM) for RR, and a relative accuracy of 97.83 % and 95.06 %, respectively.

These results highlight the potential of the proposed system to enhance driver monitoring by providing accurate and continuous HR and RR measurement, contributing to the early detection of adverse driver states.

Traditional direct status measurement methods for assessing lithium-ion batteries, such as Coulomb counting and open-circuit voltage (OCV), provide limited information into internal mechanical changes in operando. Ultrasound nondestructive evaluation (NDE) offers a complementary approach by probing the mechanical response of cells.

This study investigates ultrasound monitoring of state of charge (SoC) and degradation in lithium-polymer (LiPo) pouch cells under controlled cycling and temperature variations. Time of flight (ToF) and signal amplitude (SA) were extracted from through-transmission waveforms via cross-correlation and electrochemical data during CCCV protocols. ToF showed closed-loop behaviour and stronger correlations with SoC than SA. Repeated 0.25C cycling induced capacity fade and ultrasonic shifts, though successive temperature calibrations partially masked these changes. Tests at 15°C, 25°C, and 35°C revealed ToF shifts of up to ∼0.7 µs relative to the 25°C baseline. Therefore, a temperature correction factor normalised ToF to this reference, unifying the ToF–SoC trend at 95% state of health (SoH) and clarifying degradation signatures.

These results show the importance of accounting for temperature effects in the ultrasonic signatures in support of accurately monitoring degradation.

The study examines workplace travel policies’ role in cutting commuting emissions. Using the LSEG Environmental, Social and Governance database (previously known as Refinitiv), we analyze a sample of 2,932 organizations employing over 86 million people across 73 countries to identity predictors of (a) workplace travel policies, and (b) commuting emission reductions.

Drawing on political and organizational science literatures, we examine the roles of employee involvement and trust in reducing travel emissions. Sector characteristics strongly influence policy adoption—professional services firms are six times more likely than manufacturing firms to implement transportation policies (OR = 5.98, p < 0.001).

While these policies significantly correlate with emissions reductions, the effect size is modest (Cohen's d = 0.225, R2 = 0.0765). Notably, trust in employers emerges as a significant predictor of emissions reductions (β = -0.122, p < 0.05), while traditional employee involvement structures show limited effectiveness. 

These findings extend beyond local case studies, suggesting successful emissions reduction depends on both policy design and organizational context.

This study presents the first demonstration of the use of X-ray diffraction (XRD) to quantify the radial or transverse deformation in Hexcel IM7 PolyAcryloNitrile (PAN)-based carbon fibres at temperatures as low as 200 K (-70 °C).

The Coefficient of Thermal Expansion (CTE) is a critical design parameter that needs to be precisely quantified for the next generation of carbon fibre-based Liquid Hydrogen (LH₂) storage tanks for net-zero aviation. This variable quantitatively describes the thermal mismatch between the fibre and the resin that is the driver for microcracking and tank leakage. However, quantification of the CTE of the fibres is experimentally challenging. The results provide unique insights, indicating that the microscopic transverse CTE of the fibre (α₂₂) is equal to 26.2 × 10^-6 K^-1 and is governed by van der Waals forces, similar to those in the basal c-axis (out-of-plane) direction of graphite and the radial direction of multi-wall carbon nanotubes.

Taking into account the microcrack-induced relaxation effect reported in polycrystalline graphite, the macroscopic fibre transverse CTE was determined to be 7.86 × 10^-6 K^-1. XRD data were also collected on Hexcel IM7/8552 Uni-directional (UD) and Quasi-isotropic (QI) composite laminates to investigate the influence of the interaction of the resin matrix with the fibre lattice and the stacking sequence on the development of thermal fibre lattice strain.

In the UD laminate, the presence of resin induces an additional transverse strain in the fibres as a result of resin contraction during cooling, leading to the development of a compressive strain in the fibre direction. This behaviour was found to be in good agreement with numerical simulations, with a 13 % error at the lowest measured temperature.

In contrast, the fibres in the QI configuration were reinforced in the transverse direction, effectively mitigating the influence of resin contraction. These CTE values, insights, and resulting models are essential for multi-scale modelling, design and certification of carbon fibre composite LH₂ tanks that are required to achieve net-zero aviation.

This paper demonstrates a new approach that exploits both lattice strain mapping via Wide Angle X-ray Scattering (WAXS) and Digital Volume Correlation (DVC) of Computed Tomography (CT) to understand the material response at different length scales in Carbon Fibre Reinforced Polymers (CFRPs) under in-situ loading, a phenomenon of substantial importance for the modelling, design, and certification of composite structures. WAXS gives insight into fibre lattice strain, while DVC provides sub-laminate response in the CFRP. A detailed numerical simulation was also developed to compare with these novel experimental methods.

This approach is the first demonstration that the strain within the crystalline regions of the fibre is distinct from the sub-laminate behaviour, with up to 80 % and 36 % differences in the longitudinal and transverse directions, respectively, as a result of the complex microstructure of the fibres.

An improved understanding of composite behaviour is fundamental to understanding how strain accommodation leads to structural failure, providing routes to refine part rejection criteria and reduce the environmental impact of this increasingly widespread material class.

Verification, validation, and testing of complex cyber-physical systems, such as autonomous vehicles, have traditionally relied on document-intensive processes and physical prototypes. However, accelerated development timelines increasingly challenge the feasibility of these approaches. Model-based systems engineering (MBSE) offers a model-centric alternative that enables new forms of system analysis. 

This article presents a graph-based methodology for evaluating system modelling language system models using functional flow block diagrams and interface definitions. By transforming system functions into a functional dependence graph and applying community detection techniques, we derive three structural metrics: system complexity, system modularity, and system test effort.

The methodology is applied to three case studies, an academic autonomous mobile robot and two industrial automotive propulsion systems, to explore its potential utility. Across these examples, the graph-based approach yielded improvements in all three metrics compared to traditional subject-matter-expert-based partitioning.

The findings suggest that graph-based evaluation may support early-stage architectural reasoning and model refactoring in MBSE workflows, with further validation needed to confirm its broader applicability and practical impact.

Numerous studies have detected associations between poor housing quality and increased risk for mental illness. However, it currently remains unclear in associations between poor housing quality and increased risk for women’s mental illness which housing quality indicators drive this association and hence which specific indicators should be prioritised in housing quality assessments or improvements.

In a sample of up to 9,669 pregnant women, we used a network analysis to investigate cross-sectional associations between poor housing quality indicators (e.g., house size, facilities, leaks or condensation/mould, decorations, and feelings towards the home) and depressive symptoms (assessed at age 28).

All 36 edges showed non-zero associations, whereby when considering all poor housing quality indicators ‘feelings-towards-the-home’ had the strongest association with depressive symptoms, and ‘feelings-towards-the-home’, in turn, was most strongly associated with house problems, size, and facilities.

Our findings highlight the importance of using multiple (or composite) person-centred measures of housing quality in the context of maternal mental health.

In-wheel electric machines are a desirable form of propulsion for lightweight electric vehicles, however the torque capability and speed range is often compromised due to the fixed gear ratio and volume/mass constraints from packaging the motor within the wheel of the vehicle. Actively changing the winding configuration of an electric machine during operation can be leveraged to enhance the torque capability, speed range, and efficiency of electric machines; the characteristic performance of which can be likened to that of a mechanical gearbox.

Reducing the flux linkage of a permanent magnet synchronous machine through winding reconfiguration from star to delta can significantly improve high speed efficiency, with a 32.78% reduction in total loss without detriment to low speed torque capability. This translates to a potential 5.43% reduction in total energy consumption over a real-world drive cycle of a solar powered electric vehicle. Improving drive cycle efficiencies for lightweight electric vehicles highlights a more sustainable transportation option for climate change mitigation strategy.

Accurate flux and inductance estimation is crucial for high-fidelity modelling and emulation of interior permanent magnet synchronous machines (IPMSMs).

This paper presents a systematic workflow for extracting d-and q-axis flux and inductance look-up tables (LUTs) from full-factorial experimental data using voltage equations derived from the IPMSM equivalent circuit model. The workflow begins with data acquisition from an IPMSM testbed, capturing current, voltage, speed, torque, and temperature across a wide operating range. Using the IPMSM voltage equations, the d- and q-axis flux linkages and inductances are computed while accounting for temperature-dependent resistance variations.

Gaussian Process Regression (GPR) is then employed to interpolate and extrapolate flux values over an extended operating range, ensuring accurate LUT generation. The final flux and inductance LUTs are formatted for direct integration into electric machine emulators, enabling real-time validation and optimisation of electric drive control strategies.

Experimental validation confirms the accuracy and reliability of the proposed approach, demonstrating its potential for hardware-in-the-loop (HIL) testing, virtual prototyping, and control system development in electrified powertrains.

The control of circulating currents in Modular Multilevel Converters (MMC) is desired for its benefits to stability of the system as well as reducing peak arm currents. In this paper, a simplified method for the control of circulating currents is implemented that is based on a simple scaling of the Sub module (SM) duty cycle reference, itself also based on the ratio of desired vs. actual capacitor voltage. This approach maintains transparency for a conventional 3-phase reference input, in terms of the required adjustments to maintain the MMC balanced and Circulating Currents under control, when using Phase Shifted Carrier Modulation (PS-PWM) as this scheme allows for direct manipulation of individual SM duty cycle values. The use of filtered samples of the voltages of the SM capacitors is shown to offer a strong simplification of the control of circulating currents when using PS-PWM.