James Riordan

This work is focussed on sputtered lead cadmium selenide (Pb _1−x Cd _x Se) thin films as the active material within high-speed room temperature mid-infrared photovoltaic (PV) detectors for use in non-dispersive infrared gas sensors. Variation of the Cd content provides a means of controlling material bandgap and hence wavelength response. Traditional methods of Pb _1−x Cd _x Se deposition are thermal evaporation or chemical bath techniques which do not provide required stable deposition for accurate control of Cd content and hence wavelength response. This paper describes pulsed direct current (DC) sputter deposition of Pb _1−x Cd _x Se, from a sintered pre-mixed sputter target as an alternative deposition technique. Sputtered Pb _1−x Cd _x Se material characterisation is described identifying optimum Cd content as x = 0.1 for maximised response at a wavelength of 3.3 μm, matching use as a detector for methane gas detection. Work has focussed on characterising pulsed DC sputter deposition of the 3.3 μm optimised Pb _0.9 Cd _0.1 Se (n-type) and post deposition oxygen doping by thermal diffusion to create p-type Pb _0.9 Cd _0.1 Se: O thereby maximising PV detector photo-response at a specific wavelength of 3.3 μm. PV detector design, associated photolithographic mask designs/ lift-off process sequence, sputter deposition of Pb _1−x Cd _x Se and Ti interdigitated electrodes for PV detector fabrication is described. The resulting 1 × 1 mm ² PV detector device exhibited a peak detectivity D* of 1 × 10 ¹⁰ , 4 × 10 ⁹ and 3 × 10 ⁹ cm∙Hz ^1/2 ∙W ⁻¹ at 233 K, 298 K and 313 K respectively under 500 K blackbody illumination.

Hyperspectral image (HSI) classification is vital for environmental monitoring, land cover mapping, and precision agriculture, but its effectiveness is often constrained by the scarcity of labeled samples and high spectral similarity among classes. To address these challenges, we propose CrossCapsViT, a hybrid classification framework that integrates Capsule Networks (CapsNets) and Vision Transformers (ViTs) through a Cross-Attention Fusion (CAF) mechanism and a cross-layer adaptive fusion module, enabling richer and more discriminative spectral–spatial feature learning. To further improve efficiency in data scarce scenarios, we embed an Actor–Critic reinforcement learning based active learning (RAL) strategy that jointly leverages accuracy, uncertainty, and diversity in the reward structure, guiding the selection of the most informative samples while reducing labeling effort. Experiments conducted on four benchmark datasets (KSC, PU, HU2013, and Salinas) and a custom UAV based saltmarsh dataset (Derrymore, collected with a Pika-L sensor) demonstrate that CrossCapsViT with RAL consistently outperforms CapsViT and other baseline models in terms of classification accuracy, robustness, and generalizability. The proposed framework achieves up to 25% improvement in class-level accuracy on challenging vegetation classes, while reducing dependence on large annotated datasets, highlighting its potential for practical deployment in real-world ecological monitoring and remote sensing applications.

Accurate classification of salt marsh vegetation is vital for conservation efforts and environmental monitoring, particularly given the critical role these ecosystems play as carbon sinks. Understanding and quantifying the extent and types of habitats present in Ireland is essential to support national biodiversity goals and climate action plans. Unmanned Aerial Vehicles (UAVs) equipped with optical sensors offer a powerful means of mapping vegetation in these areas. However, many current studies rely on single-sensor approaches, which can constrain the accuracy of classification and limit our understanding of complex habitat dynamics. This study evaluates the integration of Red-Green-Blue (RGB), Multispectral Imaging (MSI), and Hyperspectral Imaging (HSI) to improve species classification compared to using individual sensors. UAV surveys were conducted with RGB, MSI, and HSI sensors, and the collected data were classified using Random Forest (RF), Spectral Angle Mapper (SAM), and Support Vector Machine (SVM) algorithms. The classification performance was assessed using Overall Accuracy (OA), Kappa Coefficient (k), Producer’s Accuracy (PA), and User’s Accuracy (UA), for both individual sensor datasets and the fused dataset generated via band stacking. The multi-camera approach achieved a 97% classification accuracy, surpassing the highest accuracy obtained by a single sensor (HSI, 92%). This demonstrates that data fusion and band reduction techniques improve species differentiation, particularly for vegetation with overlapping spectral signatures. The results suggest that multi-sensor UAV systems offer a cost-effective and efficient approach to ecosystem monitoring, biodiversity assessment, and conservation planning.

Unmanned aerial vehicle (UAV) state estimation is fundamental across applications like robot navigation, autonomous driving, virtual reality (VR), and augmented reality (AR). This research highlights the critical role of robust state estimation in ensuring safe and efficient autonomous UAV navigation, particularly in challenging environments. We propose a deep learning-based adaptive sensor fusion framework for UAV state estimation, integrating multi-sensor data from stereo cameras, an IMU, two 3D LiDAR’s, and GPS. The framework dynamically adjusts fusion weights in real time using a long short-term memory (LSTM) model, enhancing robustness under diverse conditions such as illumination changes, structureless environments, degraded GPS signals, or complete signal loss where traditional single-sensor SLAM methods often fail. Validated on an in-house integrated UAV platform and evaluated against high-precision RTK ground truth, the algorithm incorporates deep learning-predicted fusion weights into an optimization-based odometry pipeline. The system delivers robust, consistent, and accurate state estimation, outperforming state-of-the-art techniques. Experimental results demonstrate its adaptability and effectiveness across challenging scenarios, showcasing significant advancements in UAV autonomy and reliability through the synergistic integration of deep learning and sensor fusion.

The integration of drones into shared airspace for beyond visual line of sight (BVLOS) operations presents significant challenges but holds transformative potential for sectors like transportation, construction, energy and defence. A prerequisite for this integration is equipping drones with enhanced situational awareness to ensure collision avoidance and safe operations. Current approaches mainly target single object detection or classification, or simpler sensing outputs that offer limited perceptual understanding and lack the rapid end-to-end processing needed to convert sensor data into safety-critical insights. In contrast, our study leverages radar technology for novel end-to-end semantic segmentation of aerial point clouds to simultaneously identify multiple collision hazards. By adapting and optimizing the PointNet architecture and integrating aerial domain insights, our framework distinguishes five distinct classes: mobile targets like drones (DJI M300 and DJI Mini) and airplanes (Ikarus C42), and static returns (ground and infrastructure) which results in enhanced situational awareness for drones. To our knowledge, this is the first approach addressing simultaneous identification of multiple collision threats in an aerial setting, achieving a robust 94% accuracy. This work highlights the potential of radar technology to advance situational awareness in drones, facilitating safe and efficient BVLOS operations.

Future autonomous Unmanned Aerial Vehicles (UAV) missions will take place in highly cluttered urban environments. As a result, the UAV must be able to autonomously evaluate risks and react to unforeseen hazards. The current regulatory framework for missions implements SORA guidelines for hazard detection, but its application to air-to-air collision is limited. This research defined a rigorous verification and validation framework (V&V) for digital twins for use in future autonomous UAV missions. The researchers designed a sentry mission for a UAV to evaluate its capacity to detect small uncooperative flying objects. A digital twin of the DJI M300 vision system was built using a game engine and a V&V framework was developed to assure the quality of results in both virtual and real-world scenarios. The results showed the capability of the digital twin to identify vulnerabilities and worst-case scenarios in UAV mission operations, and how it can assist remote pilots in identifying air-to-air collision hazards. Furthermore, the probability of air-to-air collision was calculated for three sentry patterns, and the results were validated in the field. This research demonstrated the capability to identify vulnerabilities and worst-case scenarios in UAV mission operations. We present how the digital twin of an operational theatre can be exploited to assist remote pilots with the identification of air-to-air collision hazards of small uncooperative objects. Furthermore, we discuss how these results can be used to enhance current SORA-based risk assessment practices.

The safe operation of drone swarms beyond visual line of sight requires multiple safeguards to mitigate the risk of collision between drones flying in hyper localised scenarios. Cooperative navigation and flight coordination strategies that rely on pre-planned trajectories and require constant network connectivity are brittle to failure. Drone embedded sense and detect offers a comprehensive mode of separation between drones for deconfliction and collision avoidance. This paper presents the first airborne LiDAR based solution for drone-swarm detection and localisation using 3D deep learning. It adapts and embeds the PointPillars deep learning neural network on the drone. To collect training data of close-quarter multi drone operations and safety critical scenarios, a scenario Digital Twin is used to augment real datasets with high fidelity synthetic data. The method has been validated in real-world tests. The trained model achieves over 80% recall and 96% precision when tested on real datasets. By incorporating a detection-by-tracking algorithm the system can reliably monitor the separation distance of multiple drones in challenging environments.

Services that exploit Unmanned Aircraft Systems (UAS) are poised to revolutionise the service industry with a projected value of 71 BUSD by the end of the decade. A key enabler of this technology is the unlocking of autonomous Beyond Visual Line of Sight (BVLOS) operations. BVLOS operations will depend on a robust Detect and Avoid (D&A) capability. Yet, currently in the UK and EU, BVLOS operations are only allowed in specific cases and scenarios. As a result, the technological landscape for the development of robust D&A faces limitations, and there is little market incentive for development. Furthermore, while automated BVLOS is a future technology, a strong D&A capability is of importance now for all types of UAS operations. As the remote pilot has to deal with information overload from the controller device and the environment. These high-risk UAS operations are becoming more common. In this paper, we discuss the current legal framework in the UK making comparisons to EU countries. We make the case that even when an operation abides by the current framework the remote pilot is exposed to several legal liabilities. We review the roadmaps for UAS adoption (including certification processes for UAS-based products) and highlight that for software-intensive systems, key steps are missing to assure the quality of the product. Finally, we build on these findings to set forwards a path to complement future certification processes to enable autonomous based UAS operations to share the airspace with remotely piloted operations.