20505 research outputs found
Sort by
Hierarchical coverage path planning for a multi-modal robot exploring disconnected regions
Get PDFWe present a near-optimal Coverage Path Planning (CPP) approach for multi-modal robots in complex environments with multiple arbitrarily shaped disconnected regions. The problem arises in ground coverage missions conducted in unstructured terrains, such as planetary exploration or search and rescue missions, where safe regions are disconnected by areas of high slope that a robot with a single locomotion modality cannot traverse. A multi-modal robot can switch between different locomotion modalities (e.g. driving and flying) to safely navigate these challenging environments while ensuring complete coverage. The proposed method identifies both traversable and non-traversable areas based on Digital Elevation Model (DEM) meshes. The problem is then formulated as a hierarchical Traveling Salesman Problem (TSP) and solved using Mixed-Integer Linear Programming (MILP). The proposed approach is evaluated on 500 randomly generated maps and four real-world simulation scenarios constructed from Martian terrain DEM data.26th Annual Conference, TAROS 202
Toward intuitive drift assist control: driver drift intention recognition using a data-based approach
Get PDFIn contrast to autonomous drifting where path-planning determines when and where the vehicle drifts, to support drift-assist control systems in the framework of Advanced Driving Assistance System (ADAS), the human driver’s intention needs to be recognised before the system intervenes with its assist functionality to facilitate drifting at the desired moment. We propose a method based on Bidirectional Long Short-Term Memory Network (Bi-LSTM) to interpret driver’s intention to start/exit drifting, utilising only basic driver inputs and vehicle state signals. Firstly, to comprehensively understand driver’s behaviour during drifting, we discuss the distinctive features in throttle and steering inputs and the corresponding vehicle acceleration signals during drift cornering and normal cornering, respectively. Next, two Bi-LSTM models are designed separately for the recognition of the ‘Intention to Start Drifting’ and the ‘Intention to Exit Drifting’. Then, these models are trained and evaluated through a data set that contains over 500 laps of driving collected from the racing simulator Assetto Corsa. To validate the proposed approach, test sets of different drivers, track layouts and car models are adopted. The proposed intention recognition models successfully reach an accuracy of over 90% in recognising the two concerned intentions and outperform other classification methods in comparison.This work was supported by the Rimac Technology.Vehicle System Dynamic
Random wavelet kernels for interpretable fault diagnosis in industrial systems
Get PDFDeep learning is a powerful method for fault diagnosis, but its "black-box" nature raises concerns in critical applications. This paper presents an interpretable, lightweight method combining random convolution kernel transformation (ROCKET) with wavelet kernels, which offer systematic time-frequency analysis and intuitive insights. Principal component analysis (PCA) is used to extract relevant patterns, forming a health indicator that guides maintenance decisions. A case study on linear actuator fault diagnosis demonstrates the method's balance of interpretability and computational efficiency, making it a valuable tool for reliable asset health monitoring in resource-limited settings.Engineering and Physical Sciences Research Council (EPSRC)The research was partially supported with EPSRC funding (EP/P027121/1).CIRP Annal
A standardized comparative framework for machine learning techniques in lithium-ion battery state of health estimation
Get PDFThe accurate estimation of lithium-ion battery State of Health (SOH) is essential for enhancing performance, safety, and lifecycle management in modern energy systems. While numerous individual studies have explored machine learning approaches for SOH prediction, a systematic comparative analysis using consistent experimental protocols and rigorous cross-validation remains limited. This study addresses this gap by presenting the first comprehensive comparison of three advanced machine learning models—Extreme Gradient Boosting (XGBoost), Random Forest, and Support Vector Machine (SVM)—using a standardized experimental framework with NASA battery datasets. Our novel contribution lies in implementing a unified training-testing protocol using battery #5 for training and batteries #6, #7, and #18 for validation, combined with systematic hyperparameter optimization through grid search and k-fold cross-validation. Key improvements include: (1) first standardized one-to-many validation protocol ensuring cross-battery generalization assessment that eliminates the data splitting limitations of previous comparative studies, (2) unified hyperparameter optimization methodology applied identically across all algorithms, eliminating the confounding effects of inconsistent parameter tuning that have biased previous comparisons, and (3) establishment of quantitative performance benchmarks providing evidence-based model selection criteria for practical Battery Management System (BMS) applications. The XGBoost model achieved superior performance with MAE of 0.016 and MSE of 0.000347, establishing empirical benchmarks for model selection in battery health diagnostics through our systematic comparative methodology. This work provides the first standardized comparative framework for SOH estimation, offering evidence-based guidance for BMS implementations and advancing the field toward more rigorous and replicable research practices in battery prognostics.Future Batterie
Digital and Industry 4.0 technologies in olive farming and industry: recent applications and future outlook
No full textVital to the economy, culture and landscape of many regions around the world, the olive sector faces significant challenges, including rising production costs, labour shortages, climate change impacts, water scarcity, quality control issues and market demands for transparency and authenticity. Digital and other Industry 4.0 technologies offer transformative potential to address these pressures. This article provides a comprehensive review of recent applications and future prospects of technologies such as the Internet of Things, Artificial Intelligence, Machine Learning, Robotics and Automation, Big Data Analytics, Advanced Sensing, Remote Sensing, Nanotechnology and Blockchain across the olive value chain, from cultivation to supply chain management. Using a literature review methodology to identify key application areas, it synthesises evidence on how these innovations increase resource efficiency, optimise farm management, automate labour-intensive tasks, improve pest and disease control, ensure product quality and authenticity, facilitate traceability and add value through by-product valorisation. Key benefits include improved yields, reduced environmental impact, enhanced quality control, fraud deterrence and increased consumer confidence. Future prospects include deeper integration of technologies, more sophisticated AI-driven decision support, advanced robotics, widespread adoption of rapid sensing techniques, development of circular economy models and nanotechnology applications, while recognising the need for safety assessments. Overcoming barriers related to cost, digital literacy, data interoperability and equitable access, especially for smallholder farmers, is critical. This review highlights the strategic importance of embracing digital transformation to strengthen the resilience, sustainability and competitiveness of the global olive industry.Smart Agricultural Technolog
Usability of agricultural drought vulnerability and resilience indicators in planning strategies for small farms: a principal component approach
Get PDFWater-related stresses and risks of droughts, exacerbated by climate change, have been extensively documented. These studies often rely on various indicators to monitor and forecast the impacts of droughts. However, current literature on the usability of these indicators for modelling drought risk and in decision-making processes is fragmented and lacks a clear, systematic, and methodological approach. Usability, in this context, refers to the relevance, accessibility, clarity, and practicality of indicators for guiding planning strategies. To address this knowledge gap, the Management of Disaster Risk and Societal Resilience (MADIS)1 project aims to collate and assess drought vulnerability and resilience indicators from existing literature to support decision-makers in improving policies related to agricultural droughts on small farms.
The MADIS project identified over 100 indicators, from which 36 were selected for further analysis. A global online survey using the Delphi technique was conducted, and the resulting data was used to perform a Principal Component Analysis (PCA). Findings revealed that these 36 indicators could be reduced and grouped up to ten principal components, each corresponding to a theme across five categories: relevancy, understanding, accessibility, objectivity, and temporal. This study, therefore, highlights the practical usability of these indicators for developing context-specific and efficient resilience strategies.
Indicators related to water management were found to be crucial and applicable across all five categories, as the availability, quality, and source of water are essential for monitoring and mitigating drought hazards. Conversely, indicators related to rural development and demographics, while quantifiable and collected at different temporal scales, were deemed less understandable and accessible by experts. Grouping indicators under common themes reduces the complexity of evaluating similar indicators and aids in selecting the most relevant ones for different contexts. This approach simplifies indicator selection and enables decision-makers to formulate resilience policies more efficiently and comprehensively.Engineering and Physical Sciences Research Council (EPSRC)This work was funded by the Engineering and Physical Science Research Council (EPSRC, United Kingdom) Grant no. EP/V006592/1 and National Science Foundation (NSF, United States) Grant no. 2039506 and Belmont Forum Project DR32019 - Management of Disaster Risk and Societal Resilience (Old project name: Theory of Change Observatory on Disaster Resilience–TOCO DR).Climate Service
A framework for sustainable construction project based on BIM environment
Get PDFThe global construction industry's significant resource consumption and
environmental impact underscore the urgency of sustainability. This research
emphasizes the intersection of sustainability and construction, focusing on
ecological, economic, and social considerations. It highlights Building Information
Modelling (BIM) as a key enabler of sustainability within the construction value
chain.
Construction, vital to economic and societal development, necessitates
sustainability as a core project objective. Efficient resource utilization, compliance
with evolving sustainability standards, and effective collaboration among
stakeholders are crucial. However, communication challenges often impede
shared understanding and data integration. BIM emerges as a digital solution,
unifying project phases, facilitating collaboration, and informed decision-making.
In response to the global sustainability mandate, construction projects worldwide
are adopting more effective approaches. BIM plays a pivotal role in enhancing
efficiency, performance, and productivity. This research addresses a gap by
presenting a framework to assess how building materials impact energy
consumption within the BIM environment.
The research aims to develop a comprehensive framework for promoting
sustainability in construction projects through BIM. This involves investigating
sustainable practices, assessing BIM's role in sustainability, selecting optimal
engineering calculations, and creating an integrated framework. The framework's
effectiveness will be evaluated through a hypothetical case study.
Key research questions include BIM's alignment with sustainability, expected
improvements in addressing sustainability issues, and the value of sharing
sustainability calculations within the construction value chain. The thesis
comprises seven chapters, including a literature review on sustainability and BIM,
a detailed research methodology, a hypothetical case study, analysis of
conduction heat transfer calculations, the development of a sustainable
construction framework, and discussions, conclusions, and future directions. This
research seeks to empower the construction industry with a practical framework
for embedding sustainability within the BIM environment, driving efficiency,
environmental responsibility, and societal well-being.PhD in Manufacturin
Fabrication and functional evaluation of nature-inspired anti-bacterial surfaces
Get PDFGiusca, Claudiu - Associate Supervisor
Kumar, Vinod - Associate SupervisorThe critical need to develop novel and efficient anti-bacterial strategies,
particularly for biomedical implants, serves as a significant motivation for this
research. The rise in antibiotic resistance continues to pose a threat to
healthcare, making it increasingly important to explore alternative approaches to
combat implant-associated surgical site infections (SSIs). These infections arise
from bacterial attachment and biofilm formation on the surface of implants and
medical devices, leading to costly treatments and high recurrence rates. The
present research was aimed at investigating nature-inspired anti-bacterial
surfaces through a rigorous fabrication and testing campaign.
The study investigated various nanofabrication techniques, such as femtosecond
laser ablation, deep reactive ion etching, focused ion beam lithography and
scanning probe lithography, for creating nature-inspired sub-micron features on
stainless steel and silicon surfaces. The biological response of bacteria (S.
aureus) and osteoblast-like cells (MG-63) was evaluated on these surfaces to
test antibacterial as well as osseointegration response of the surfaces. S. aureus
was chosen due to its high relevance to SSIs and its prevalence in infection and
MG-63 cells served as a model for examining the osteoblast behaviour in
laboratory studies.
The thesis established a novel scale-dependent relationship between surface
topography and biological functionality, characterised by the dominating surface
wavelength and fractal dimension. The anti-biofouling mechanism was influenced
by surface topography, characterised through the fractal dimension, and it was
consistently achieved and deemed more suitable for future applications due to
the high anti-bacterial efficiency achieved compared to the mechano-bactericidal
mechanism. High aspect ratio features (0.056-0.280 µm wavelength, 0.295-
0.765 µm height, 0.045-0.046 µm diameter) did not induce mechano-bactericidal
effects on S. aureus NCTC7791, indicating further research is needed. Moreover,
the thesis demonstrated a predominant attachment of S. aureus and MG-63 cells
on the crystalline silicon surfaces on the (111) orientation.
ii
For feature sizes below 1 µm, the fractal dimension positively correlated with the
anti-bacterial effect and MG-63 cell spreading. For sizes significantly larger than
bacterial size (> 2 µm), no correlation was found with the anti-bacterial effect, but
surface complexity positively correlated with MG-63 cell spreading. For feature
sizes comparable to MG-63 cell size (10-40 µm), cell spreading was inhibited.
Femtosecond laser ablation emerged as a promising technique for commercial
applications, while scanning probe lithography proved to be a cost-effective,
flexible tool for prototyping and research-scale investigations.
In conclusion, the development and evaluation of nature-inspired anti-bacterial
surfaces have revealed valuable insights into the scale-dependent relationship
between surface topography and biological functionality. The findings from this
research have the potential to improve the performance and safety of implantable
medical devices by reducing the risk of implant-associated SSIs, ultimately
benefiting patients and healthcare providers.PhD in Manufacturin
Normalised diagnostic contribution index (NDCI) integration to multi objective sensor optimisation framework (MOSOF)—An environmental control system case
Get PDFIn modern aerospace systems, effective sensor optimisation is essential for ensuring reliable diagnostics, efficient resource allocation, and proactive maintenance. This paper presents Normalised Diagnostic Contribution Index (NDCI) integration into the Multi-Objective Sensor Optimisation Framework (MOSOF) to address application-specific performance nuances. Building on previous work, the proposed approach leverages a multi-objective genetic algorithm to optimise key criteria, including performance, cost, reliability management, and compatibility. NDCI is derived from simulation data obtained via the Boeing 737-800 Environmental Control System (ECS) using the SESAC platform, where degradation level readings across four fault modes are analysed. The framework evaluates sensor performance from the perspectives of Original Equipment Manufacturers (OEM), Airlines, and Maintenance Repair Overhaul (MRO) organisations. Validation against the Minimum Redundancy Maximum Relevance (mRMR) method highlights the distinct advantage of NDCI by identifying an optimal set of three sensors compared to mRMR’s six-sensor solution, and MOSOF’s multi-objective insertion enhances sensor deployment for different stakeholders. This integration not only expands the feasible solution space for sensor-pair configurations but also emphasises diagnostic value over redundancy. Overall, the enhanced NDCI-MOSOF offers a scalable, multi-stakeholder approach for next-generation sensor optimisation and predictive maintenance in complex aerospace systems. The results demonstrate significant improvements in diagnostics efficiency for stakeholders.Republic of Turkey’s Ministry of National EducationSensor
Organic management in coffee: a systematic review of the environmental, economic and social benefits and trade-offs for farmers
Get PDFGlobal coffee production is expanding, contributing to environmental degradation, notably through extensive use of inorganic fertilizers. Volatile prices, climate change, rising input costs, and pressure to decrease carbon footprints represent key challenges for farmers. Regenerative soil management and the use of organic management as an alternative to conventional mineral fertilizers offer one potential solution to address these challenges. However, information is limited regarding the potential options available for farmers, and their potential environmental, economic, and social impacts. We undertook a systematic review of the literature to assess the benefits and trade-offs from adopting different organic management approaches following PRISMA guidelines. We identified 43 peer-reviewed articles, predominantly focusing on agroforestry, plant-derived additions, soil management or animal manure to improve livelihoods and environment. Research priorities differ by region and there is a skew toward researching the environmental impacts of regenerative techniques. Our synthesis demonstrates multiple potential environmental benefits to organic management, but increasing economic risks and trade-offs for farmers, particularly in transitioning to organic management. We also highlight the social barriers facing farmers, from education to access to knowledge networks to support implementation. These challenges must be addressed to support any future sustainable transitions to organic management in coffee.(Natural Environment Research Council)This work was supported by the Natural Environmental Research Council (NE/X001687/1; NE/X001687/2; NE/X001679/1), the Douglas Bomford Trust, and the Jack Wright Memorial Trust.Agroecology and Sustainable Food System