34028 research outputs found
Sort by
Plant2Sim3D: From Real Plants to Simulated Worlds via Video-Driven 3D Mesh Generation
In agri-robotics, the absence of realistic and varied simulation settings, primarily due to difficulties in modelling the intricate visual and geometric structures of plants, limits the efficacy of training and assessing robotic systems. Current methodologies often rely on manually crafted or overly simplistic 3D models that fail to adequately represent the real-world diversity and complexity. To address this gap, we introduce \textit{Plant2Sim3D}, a video-driven pipeline that reconstructs simulation-ready, high-fidelity 3D plant meshes using neural rendering, interactive segmentation, and procedural world generation. To assess the fidelity of our reconstructions, we evaluate their geometric similarity to ground-truth scans, demonstrating close alignment in surface accuracy and coverage. These low distances and high coverage indicate that the reconstructed meshes preserve both structural accuracy and surface detail, validating their suitability for realistic and reliable use in simulated robotic environments.</p
Fog Bound
Fog Bound is a flash fiction composition of 500 words; it is a story about children venturing out to play a game of football in the fog. All is fine until one of them kicks the ball a little too far.</p
Effect of wheat bread with elevated amylose on postprandial glycaemic response: a randomised crossover trial delivered remotely using continuous glucose monitoring
This study measured the effect of white bread made with a starch branching enzyme II (sbeII) mutant wheat flour with elevated amylose on postprandial glycaemic response compared to an isoglucidic white bread made with a wild-type (WT) control flour. A randomised, double-blind, placebo-controlled two-period crossover trial was conducted to measure glycaemic responses after consuming sbeII and WT bread rolls, in duplicates. The study, comprising 26 healthy adult participants (≥18 years of age; BMI ≥ 18 and ≤30 kg m−2 ; HbA1C sbeII bread than the WT (p = 0.006). Effects on satiety and palatability were evaluated using online questionnaires; there was no difference between products in their overall effects on satiety, however more participants preferred the WT bread compared to the sbeII bread, which had a slightly harder and less resilient texture when measured instrumentally. Future studies should investigate the dose-dependent effects of foods with increased amylose on glycaemic responses to determine whether higher levels of amylose could yield greater metabolic benefits, while maintaining palatability and consumer acceptance.</p
Analysis of Adaptive Passive Technologies for the Robot Traversal in Unstructured Environments
This paper evaluates adaptive passive technologies for improving robotic locomotion in unstructured environments, focusing specifically on the Passively-Transformable Single-Part Wheel (PaTS-Wheel) and our novel variant, the “DogBone-Wheel.” We assessed these wheel designs in authentic woodland environments, measuring their energy efficiency, vibration profiles, and traversal capabilities in comparison to traditional wheels. Our findings demonstrate that the PaTS-Wheel achieved a superior obstacle-climbing ability (70% of wheel diameter vs. 25% for standard wheels), while maintaining remarkably consistent energy consumption across varied terrain types—only a 4% variation compared to 35.7% for standard wheels. The wheel exhibits predictable mechanical behaviour and balanced vibration characteristics, though with 39.8% higher baseline energy consumption than traditional wheels. However, practical limitations, including debris entrapment between flexure components and the need for adequate robot ground clearance, constrain real-world performance. The DogBone-Wheel’s modular construction offered manufacturing flexibility, but it reduced traction due to the use of PLA components. These results validate the potential of passive transformation mechanisms to enhance robotic mobility in challenging terrain with consistent energy performance and minimal active control.</p
Re-imagining @ourdaysofgold_film: Follower Experience, Polyvocality, and Distributed Autofiction
Our Days of Gold (ODOG) is an ongoing, durational artwork staged on Instagram at @ourdaysofgold_film since April 2017. Over its eight-year duration, the work has accumulated new layers of memory and interpretation shaped by followers’ responses, shifting platform aesthetics, and changes to Instagram’s visual logic, including the disappearance of the square grid in January 2025. Alongside creative contributions, the project draws on a survey conducted with long-term followers, tracing how experiences of viewing, remembering, and interpreting the work unfold over time. This co-authored exposition includes videos, screen-recorded navigations, and writing produced by followers whose contributions reveal a form of polyvocality: multiple interpretive threads and associations that remediate the archive while shaping its evolving narrative. Within this distributed process, ODOG engages autofiction not as a singular self-narration but as a collective mode of authorship, emerging through dispersed readings, layered memories, and networked resonances. At the same time, the project foregrounds the precarity of social media archives, where redesigns, algorithmic shifts, and potential platform loss constantly reshape how the work circulates and persists. Drawing on debates around remediation and digital preservation, ODOG tests how meaning, memory, and narrative can be sustained within unstable infrastructures while acknowledging their continual transformation.</p
Women, leadership and governance: an examination of the sport and physical activity sector
A report written in collaboration with the Active Partnerships National Organisation. </p
A Retrieval Augmented Generation Approach for Planning on General Purpose Service Robots
The introduction of Large Language Models (LLMs) in the past few years has disrupted the fields of natural language understanding and embodied artificial intelligence. LLMs have previously been used to generate plans from human instructions, enabling a physical agent (robot) to complete abstract long-form tasks without direct programming, when provided with a list of capabilities. On the other hand, Retrieval Augmented Generation (RAG) extends the capabilities of LLMs by querying an easily-updated knowledge base ‘on the fly’, outside of its own parameters, to generate a response. This work is one of the pioneers in exploring the usability of a RAG-based approach to generate robot plans, with RAG used to inform about the capabilities of the robot, and the context in which it is operating, instead of being stored them in the LLM itself. The usability of this approach was evaluated in the context of general purpose service robots (GPSR). Experimental results revealed that the proposed approach outperforms by 15% state-of-the-art LLM-based techniques for robot planning in terms of generating valid plans from ‘implicit’ queries where a GPSR is expected to understand and perform an action based on environmental context, without being explicitly instructed on every detail. Our approach runs locally on with low computational requirements enabling real-time robot operations.</p
Impact of Social Media on Mental Health of young people of Asian background
Introduction: Given the increasing prevalence of social media, concerns about theimpact of social media on mental well-being have intensified. However, research evidence on this topic is unclear in ethnic minority groups particularly those from Asian background. This study aimed to explore the impact of social media on the mental health of young people from Asian background.Methods: As a starting point, a scoping review was conducted which identified a research question for further exploration. The next step involved using qualitative methods to explore the views of young people from Asian background, aged 18-24 years, about the impact of social media on their mental health.Due to time, financial and other constraints relating to this student project, a decision was made, to recruit a purposive sample of young people from Asian background living in the UK for the qualitative study.For the recruitment of participants, a flyer was posted on social media (including Facebook, X and Instagram). Upon reading the flyer, participants expressed interest in the study by contacting the researcher to complete a consent form and to arrange a date for interview. The interviews were done online via Microsoft Teams using semi-structure questions.The interviews were audio recorded, transcribed verbatim and analysed using a thematic method. Data collection stopped when saturation was reached after interviewing and analysing data from 15 participants.Results: The first part of this study, a scoping review, included 23 studies that met the inclusion criteria. The review found limited qualitative evidence relating to young people from Asian background and the impact of social media. That is, of the 23 studies, only 2 were qualitative, resulting in the need to conduct a qualitative study in this area.Participants in the qualitative study expressed feelings of loneliness and low self-worth, particularly when comparing themselves too thers on social media platforms. The pressure to conform to online standards and the tendency to engage in social comparison emerged as significant factors contributing to mental distress among young people. Despite these negative aspects, participants identified positive elements of social media, such as maintaining connections with friends and family, accessing support networks, and finding communities with shared interests, thoughts, opinions and access to social and emotional support. This underscores the complex role of social media in the lives of young people, functioning as both a source of connection and a potential risk to serious mental health.Conclusion: While social media can facilitate connections, it also poses significant mental health challenges, particularly due to the unique cultural and ethnic experiences of young people from Asian background. These findings emphasise the need for targeted support for young people in navigating their social media experiences more positively.</p
Experience of social harms among female sex workers following HIV self-test distribution in Malawi: results of a cohort study
In Malawi, female sex workers (FSW) have high HIV incidence and regular testing is suggested. HIV self-testing (HIVST) is a safe and acceptable alternative to standard testing services. This study assessed; whether social harms were more likely to be reported after HIVST distribution to FSW by peer distributors than after facility-based HIV testing and whether FSW regretted HIVST use or experienced associated relationship problems.Peer HIVST distributors, who were FSW, were recruited in Blantyre district, Malawi between February and July 2017. Among HIVST recipients a prospective cohort was recruited. Interviews were conducted at baseline and at end-line, 3 months later. Participants completed daily sexual activity diaries. End-line data were analysed using logistic regression to assess whether regret or relationship problems were associated with HIVST use. Sexual activity data were analysed using Generalised Estimating Equations to assess whether HIVST use was temporally associated with an increase in social harms.Results: Of 265 FSW recruited and offered HIVST, 131 completed both interviews. Of these, 31/131(23.7%) reported initial regret after HIVST use, this reduced to 23/131(17.6%) at the 3-month follow-up. Relationship problems were reported by 12/131(9.2%). Regret about HIVST use was less commonly reported in those aged 26–35 years compared to those aged 16–25 years (OR immediate regret—0.40 95% CI 0.16–1.01) (OR current regret—0.22 95% CI 0.07 – 0.71) and was not associated with the HIVST result. There was limited evidence that reports of verbal abuse perpetrated by clients in the week following HIVST use were greater than when there was no testing in the preceding week. There was no evidence for increases in any other social harms. There was some evidence of coercion to test, most commonly initiated by the peer distributor.Conclusions: Little evidence was found that the peer distribution model was associated with increased levels of social harms, however programmes aimed at reaching FSW need to carefully consider possible unintended consequences of their service delivery approaches, including the potential for peer distributors to coerce individuals to test or disclose their test results and alternative distribution models may need to be considered.</p
Hey athlete, you need to cut weight: weight loss guidance practices and perceptions of Chinese combat sports coaches
Weight loss (WL) before competition is a widespread practice in combat sports (CS). Among the various factors influencing athletes’ WL practice, coaches are consistently identified as the most influential in terms of guiding athletes’ WL practices.However, little is known about coaches’ guidance practices and perceptions of WL. Therefore, the aim of this study is to examine the WL guidance practices and perceptions of Chinese CS coaches.Background: Weight loss (WL) before competition is a widespread practice in combat sports (CS). Among the various factors influencing athletes’ WL practice, coaches are consistently identified as the most influential in terms of guiding athletes’ WL practices. However, little is known about coaches’ guidance practices and perceptions of WL. Therefore, the aim of this study is to examine the WL guidance practices and perceptions of Chinese CS coaches.Methods: The study employed an observational cross-sectional approach and adopted convenience sampling. The questionnaire consisted of 22 questions and featured four sections: (i) general information, (ii) personal experience, (iii) WL guidance practice, and (iv) perception about WL. It employed different response formats, including (i) 6 short-answer (fill-in-the-blank) questions, (ii) 15 single-choice questions consisting of 4 dichotomous items, 7 multiple-category items, and 4 Likert-type scale items, and (iii) 1 multiple-choice question. In total, 135 coaches completed the questionnaire and responses from 122 coaches (115 males, 7 females, 33.2 ± 7.1 years old) representing 10 CS modalities were included in the analysis.Results: A high proportion of coaches reported being aware of their athletes’ body mass (95%), as well as having previously required their athletes to undergo WL before competitions (96%), and having personally guided them through the WL process (94%). The primary basis for coaches’ guidance on WL was their own experience as former athletes (81%). The most recommended WL method by coaches was increasing exercise (57%). The frequency of consultation with nutritionists/doctors did not differ significantly across coaching certification level (p = 0.286), perception of the impact of WL on health (p = 0.676), and perception of the impact of WL on performance (p = 0.751). Coaches considered 18.1 ± 3.5 years as the appropriate age for athletes to initiate WL, and 6.0% ± 2.7% of body mass as the appropriate highest WL magnitude. Typically, they advised athletes to begin WL 41.4 ± 17.5 days before competition. Most coaches perceived themselves as having “some influence” (50%) over their athletes’ WL practices. The primary reason coaches (n = 117) encouraged WL was to “compete against lighter opponents to enhance winning potential” (69%). Most coaches believed that WL negatively impacts health (41%) and optimizes performance (49%). Regarding fairness in competition, most coaches believed that WL does not create an unfair competition (72%). The age at which coaches recommended athletes to begin WL differed significantly only in relation to sports discipline (p 0.05).Conclusions: Coaches are the primary drivers of pre-competition WL practices in CS, actively requiring athletes to reduce weight to compete in lower categories. While many coaches recognize potential health risks, a substantial proportion perceive WL as neutral or beneficial, and their recommendations are often based on personal experience rather than scientific evidence. Coaches’ perceptions influence the risk level of methods they recommend.</p