20505 research outputs found
Sort by
Goal oriented trajectory prediction conditioned on reachable road context
Get PDFTrajectory prediction of surrounding traffic agents is crucial for autonomous vehicles to perform collision-free and efficient planning at urban intersections. Despite interactions with neighbour objects, road layout information plays an essential role in improving prediction accuracy and enhancing the interpretability of prediction models. However, exploring reachable areas and effectively leveraging these contextual clues in predictions remain challenging. In this work, a goal-oriented trajectory prediction framework is proposed to integrate valuable road layout information. The framework leverages sparse and non-uniform map elements to represent moving intentions. For effective exploration of relevant map elements, a constrained breadth-first search is proposed, enabling simultaneous and efficient exploration across lateral and longitudinal directions by incorporating behavioural constraints. The attention mechanism and a dynamic mask are combined to focus on the most relevant map element features and predict corresponding goal points, facilitating the final trajectory prediction. This progressive narrowing of the inference space enhances both the accuracy and interpretability of the prediction model. Experimental results on the Intersection Drone Dataset and Roundabout Drone Dataset demonstrate that the proposed model achieves a 73.1% accuracy in predicting the most likely map elements, with an average displacement error below 0.75 m and a final displacement error around 1.95 m with a prediction horizon of 4 s.This work was supported in part by the China Scholarship Council under Grant No.202108690001 and the Graduate Research and Innovation Projects of Jiangsu Province under Grant KYCX21_3334.Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineerin
MOSOF with NDCI: a cross-subsystem evaluation of an aircraft for an airline case scenario
Get PDFThis article belongs to the Special Issue Sensor Data-Driven Fault Diagnosis TechniquesDesigning cost-effective, reliable diagnostic sensor suites for complex assets remains challenging due to conflicting objectives across stakeholders. A holistic framework that integrates the Normalised Diagnostic Contribution Index (NDCI)—which scores sensors by separation power, severity sensitivity, and uniqueness—with a Multi-Objective Sensor Optimisation Framework (MOSOF) is presented. Using a high-fidelity virtual aircraft model coupling engine, fuel, electrical power system (EPS), and environmental control system (ECS), NDCI against minimum Redundancy-maximum Relevance (mRMR) is benchmarked under a rigorous nested cross-validation protocol. Across subsystems, NDCI yields more compact suites and higher diagnostic accuracy, notably for engine (88.6% vs. 69.0%) and ECS (67.7% vs. 52.0%). Then, a multi-objective optimisation reflecting an airline use-case (diagnostic performance, cost, reliability, and benefit-to-cost) is executed, identifying a practical Pareto-optimal ‘knee’ solution comprising 12–14 sensors. The recommended suite delivers a normalised performance of ≈0.69 at ≈USD36k with ≈145 kh MTBF, balancing the cross-subsystem information value with implementation constraints. The NDCI-MOSOF workflow provides a transparent, reproducible pathway from raw multi-sensor data to stakeholder-aware design decisions, and constitutes transferable evidence for model-based safety and certification processes in Integrated Vehicle Health Management (IVHM). The limitations (simulation bias, cost/MTBF estimates), validation on rigs or in-service fleets, and extensions to prognostics objectives are discussed.Sensor
Female veteran transition: exploring gendered power relations, discipline and decision making
Get PDFThis paper investigates how military life affects the decision making of female RAF veterans during transition into the civilian world. By adopting a longitudinal approach and analyzing interviews conducted over 3 years this study explores the relationship between how normalized gendered militarized behaviors and idealized military aspirational identities affect choices made during transition out of the military. Disruptions to this idealized identity brought on by the decision to leave the military trigger recognition of and adjustments to the sense of self. Female veterans' consistency of self is supported through this disruption by drawing on deeply valued aspects of aspirational identity while rejecting others. The study has contributed to theory by demonstrating how gendered discipline and power combine to influence female veterans’ ways of being and support aspects of aspirational identities as they transition into civilian life and face disruptions caused by differences in gendered expectations within military and wider society. By design, this research focused on a limited number of female veterans' experiences as they left the RAF, allowing a hitherto underexplored veteran cohort to be better understood.Gender, Work & Organizatio
Metallic glasses — Versatile radiation-tolerant materials for nuclear fusion applications
Get PDFNuclear fusion (NF) imposes unprecedented requirements on materials involved. Metallic glasses (MGs) offer an impressive set of properties that hold promise to overcome related challenges. These properties range from high corrosion resistance over high mechanical strength to high radiation tolerance including possible self-healing of irradiation-induced structural changes. Their high compositional flexibility allows MGs to be designed for optimal use in various areas of NF devices. Here we provide an introduction as to how these unique properties and related manufacturing processes can be exploited for a multitude of applications in NF. An outline of a development roadmap to expedite efforts in this direction is given.Cranfield University 75th Anniversary Research Fellowship schemeCranfield Impact Acceleration schemeCranfield University Global Research Fund schemeFusion Engineering and Desig
Autonomous driving economic car-following motion strategy based on adaptive rollout model-based policy optimization
Get PDFReinforcement learning (RL) is a powerful framework with significant potential to enhance autonomous driving (AD) performance. However, its trial-and-error nature presents significant hurdles in terms of safety, efficiency, and stability. To address these challenges, we propose an adaptive rollout model-based policy optimization (AR-MBPO) algorithm tailored for car-following motion planning in autonomous electric vehicles (AEVs). The algorithm improves overall performance by incorporating a error-aware ensemble environment model and leveraging branched rollouts for efficient sample collection and policy optimization. A key innovation of AR-MBPO is an adaptive rollout mechanism that dynamically adjusts based on predictive accuracy, mitigating the impact of model inaccuracies. Additionally, energy efficiency is explicitly integrated into the optimization process to minimize energy consumption. We evaluate AR-MBPO through AEV car-following simulations, where it demonstrates superior performance, including rapid convergence, and reduced reliance on real-world interactions. The method simultaneously optimizes safety, traffic efficiency, and energy efficiency through dynamic distance adjustments, as evidenced by a 0% collision rate in testing scenarios and an 8.2% energy consumption reduction compared to non-energy-aware baselines. The results suggest potential applications in AD systems for improved safety and energy efficiency.The work has been supported by the Start-up Fund, PolyU (Grant No. P0039179).IEEE Transactions on Transportation Electrificatio
Understanding the risk of underwater skimming of slow sand filters
No full textJefferson, Bruce - Associate SupervisorAbstract
This research presents an in-depth examination of underwater skimming (UWS)
for cleaning slow sand filters (SSF), as an alternative to the traditional dry
skimming (DS). The study explores the risks and advantages of UWS compared
to DS methods, particularly in terms of filtration performance, microbial removal
effectiveness, and operational efficiencies.
The research includes an evaluation of the current state of SSF technology,
including operational practices and associated risks faced by operators within the
potable water sector, setting the background for identifying the potential
improvements that UWS may offer. The study further analyses the risks of particle
penetration, head loss development and water quality in SSFs when employing
UWS, supported by data from pilot-scale experiments conducted to compare
UWS and DS. These experiments compare the effects of UWS against the
conventional methods, focusing on filtrate quality, recovery after skimming, and
system resilience. Additionally, the study investigates the effects of UWS on
dissolved oxygen consumption within the SSF, using a mass balance model to
predict the dynamic changes in oxygen levels that occur due to operational
disturbances caused by UWS. This aspect of the research highlights the
environmental implications of adopting UWS in water treatment practices.
One of the key findings from this research is that UWS filters exhibited less
disruption post-skimming, allowing for quicker recovery to microbial compliance
compared to DS filters. They also maintained more stable microbial water quality
immediately post-skim, as evidenced by consistent removals of coliform at
1.9±0.2 log, compared to DS where removals dropped to 1.1±0.4 log. While
surface agitation in a worst-case scenario UWS system may increase the
likelihood of particle penetration and breakthrough, deeper media depth
(>500mm) mitigates this, with particle breakthrough reducing from 2229 no/mL at
200 mm to 53 no/mL at 500mm. Maintaining a 'sweetening flow' during UWS
could considerably improve DO supply, and reduce the emergence of anoxic
conditions, while considerably reducing the recovery time caused by any flow
reductions or pauses. For example, recovery can increase from 10.1-10.9 bed
volumes when flow is completely paused, to less than one bed volume with a
continuous sweetening flow during skimming.
The study maintains that UWS not only addresses the operational challenges
posed by traditional methods but also enhances the efficacy and sustainability of
SSFs. Through this comprehensive analysis, the research provides substantial
evidence that UWS could be instrumental in advancing SSF operation, solidifying
its importance as a vital aspect of future research and application in the field. This
research contributes to knowledge by demonstrating that UWS enhances the
efficacy and sustainability of slow sand filtration systems, reducing operational
downtime and maintaining microbial water quality. The potential impact of this
research is significant, as it offers a method to increase production capacity from
existing SSF assets, potentially transforming slow sand filtration practices
globally.Engineering and Physical Sciences Research Council (EPSRC)STREAM EngD Programm
Real-time prediction of wear morphology and coefficient of friction using acoustic signals and deep neural networks in a tribological system
Get PDFPredicting real-time wear depth distribution and the coefficient of friction (COF) in tribological systems is challenging due to the dynamic and complex nature of surface interactions, particularly influenced by surface roughness. Traditional methods, relying on post-test measurements or oversimplified assumptions, fail to capture this dynamic behavior, limiting their utility for real-time monitoring. To address this, we developed a deep neural network (DNN) model by integrating experimental tribological testing and finite element method (FEM) simulations, using acoustic signals for non-invasive, real-time analysis. Experiments with brass pins (UNS C38500) of varying surface roughness (240, 800, and 1200 grit) sliding against a 304 stainless steel disc provided data to validate the FEM model and train the DNN. The DNN model predicted wear morphology with accuracy comparable to FEM simulations but at a lower computational cost, and the COF with relative errors below 10% compared to experimental measurements. This approach enables real-time monitoring of wear and friction, offering significant benefits for predictive maintenance and operational efficiency in industrial applications.Processe
Explainable reinforcement and causal learning for improving trust to 6G stakeholders
Get PDFFuture telecommunications will increasingly integrate AI capabilities into network infrastructures to deliver seamless and harmonized services closer to end-users. However, this progress also raises significant trust and safety concerns. The machine learning systems orchestrating these advanced services will widely rely on deep reinforcement learning (DRL) to process multi-modal requirements datasets and make semantically modulated decisions, introducing three major challenges: (1) First, we acknowledge that most explainable AI research is stakeholder-agnostic while, in reality, the explanations must cater for diverse telecommunications stakeholders, including network service providers, legal authorities, and end users, each with unique goals and operational practices; (2) Second, DRL lacks prior models or established frameworks to guide the creation of meaningful long-term explanations of the agent's behaviour in a goal-oriented RL task, and we introduce state-of-the-art approaches such as reward machine and sub-goal automata that can be universally represented and easily manipulated by logic programs and verifiably learned by inductive logic programming of answer set programs; (3) Third, most explainability approaches focus on correlation rather than causation, and we emphasise that understanding causal learning can further enhance 6G network optimisation. Together, in our judgement they form crucial enabling technologies for trustworthy services in 6G. This review offers a timely resource for academic researchers and industry practitioners by highlighting the methodological advancements needed for explainable DRL (X-DRL) in 6G. It identifies key stakeholder groups, maps their needs to X-DRL solutions, and presents case studies showcasing practical applications. By identifying and analysing these challenges in the context of 6G case studies, this work aims to inform future research, transform industry practices, and highlight unresolved gaps in this rapidly evolving field.Engineering and Physical Sciences Research Council (EPSRC)EPSRC CHEDDAR: Communications Hub for Empowering Distributed Cloud Computing Applications and Research, EP/X040518/1 and EP/Y037421/1IEEE Open Journal of the Communications Societ
ProSAir 2022.11 User Guide
Get PDFProSAir (PROpagation Of Shocks in AIR) is a finite volume, compressible fluid dynamics solver based on second order MUSCL-Hancock time integration and the AUSMDV hybrid flux vector/difference splitting developed by Wada and Liou. ProSAir’s primary use is for modelling air-blasts in and around structures and estimating the resultant structural loading. ProSAir is a successor to Air3d version 9 and version 2022.11 still uses the same numerical algorithm. ProSAir improves on Air3d by an improved user interface.
Key features are:
Easy to use CFD modelling of blast waves caused by detonation of a single high
explosive device.
Fast and robust due to use of the well established AUSMDV finite volume scheme
for accurately capturing shocks.
Second order accuracy achieved by its MUSCL-Hancock integration scheme.
Use of spherical and axi-symmetric calculation phases to reduce run times.
Support for running on multi-core/processor computers.
Windows XP/Vista/7/8/10 64-bit version supplied as standard; 32-bit version for
use on legacy systems is still available.
Graphical user interface allowing rapid building of geometric models made up of
simple geometrical components.
Graphical display of solution during execution.
Batch jobs via command line invocation.
Window failure model based on the SAFE/SSG Glazing hazard guide
Can provide history of pressure, velocity, density and temperature at tens of thousands
of target points around the domain.
Graphical display may be captured to create video presentations.
User guide with many examples including sample input and output files.
Restrictions:
Explosive charges are assumed ideal, spherical and with no afterburning.
No method to input geometry from CAD packages.
Flowfield itself cannot be post-processed
‘Like drinking water after being in the desert’: social-symbolic work of procurement professionals in driving sustainable procurement
Get PDFSustainable procurement aims to strategically embed environmental, social, and economic considerations into purchasing decisions to minimise negative impacts and generate positive outcomes across the entire supply chain (Meehan & Bryde, 2011). Recent studies have argued that procurement professionals can play a pivotal role in sustainable procurement by acting as change makers and implementers within their organisations (Luzzini et al., 2024). However, individual-level perspectives on their agency remain limited and under-documented (Villena, 2019; Walker et al., 2012). The study examines how procurement professionals use agency to change procurement functions and incorporate sustainability. It draws on the notion of social-symbolic work which refers to the ‘purposeful, reflexive efforts of individuals, collective actors and networks of actors to shape social symbolic objects’ (Lawrence & Phillips, 2019, p.31). Where a social symbolic object is considered ‘a combination of discursive, relational, and material elements that constitute a meaningful pattern in a social system’ (Lawrence & Phillips, 2019, p.24). In this study, we examine procurement as the social-symbolic object where the discursive elements encompass the language and narratives related to purchasing. The relational elements pertain to the networks of relationships between procurement professionals, suppliers, internal departments, and regulatory bodies. Lastly, the material elements include the tools, technologies, policies, and procedures that enable procurement activities.EurOMA 202