110830 research outputs found
Sort by
Large-Language-Model-Aided Assistive Robot for Single-Operator Bimanual Teleoperation: Introduction and Validation of a Flexible Assistance System
Get PDFBimanual teleoperation tasks are highly demanding for human operators, requiring the simultaneous control of two robotic arms while managing complex coordination and cognitive load. Current approaches to this challenge often rely on rigid control schemes or task-specific automations that do not adapt well to dynamic environments or varied operator needs. This article presents a novel large language model (LLM)-aided bimanual teleoperation assistant (BTLA) that helps operators control dual-arm robots through an intuitive voice command interface and variable autonomy. The BTLA system enables a hybrid control paradigm by combining natural language interaction for an assistive robot arm with direct teleoperation of the dominant robotic arm. Our system implements six core manipulation skills with varying autonomy, ranging from direct mirroring to autonomous object manipulation. The BTLA leverages the LLM to interpret natural language commands and select an appropriate assistance mode based on task requirements and operator preferences. Experimental validation on bimanual object manipulation tasks demonstrates that the BTLA system yields a 240.8% increase in success rate over solo teleoperation and a 69.9% increase over dyadic teleoperation, while significantly reducing operator mental workload. In addition, we validate our approach on a physical dual-arm UR3e robot system, achieving a 90% success rate on challenging soft bottle handling and box transportation tasks
Variance Preserving Spectral Subsampling
No full textGenerating statistically faithful short-duration gamma-ray spectra from a single long measurement is essential in nuclear safeguards, supporting tasks such as algorithm development and machine-learning applications, especially when list-mode data are unavailable. Existing subsampling methods often distort the statistical characteristics of genuine short-duration measurements, leading to biased or unreliable analytical outcomes and thereby undermining downstream tasks. In this work, we compare five subsampling approaches using a benchmark set of 156 genuine replicate spectra collected with a high-purity germanium detector. We evaluate each method with respect to run-to-run variance, channel-to-channel variance, and preservation of total counts (losslessness). Across a wide range of subsampling ratios, only binomial subsampling without replacement consistently reproduces the statistical properties of genuine short-duration spectra, maintaining proper dispersion even in sparse spectral regions and perfectly preserving total counts. These results provide a mathematically principled and practically validated framework for generating synthetically shortened spectra when true short-duration measurements are unavailable
Parametric Phi-Divergence-Based Distributionally Robust Optimization for Insurance Pricing
No full textThis paper investigates the use of φ -divergence-based distributionally robust optimization (φ -DRO) for offline insurance pricing. To this end, we introduce a parametric formulation of DRO where the uncertainty follows a known parametric model. We interpret φ -divergence-based DRO as the optimization of a risk functional over the objective distribution, offering an intuitive understanding of its behaviour. Applying this framework to a real-world-inspired insurance pricing problem, we find that the obtained robust policies tend to be overly conservative and yield limited performance gains under distributional shifts. These findings are validated through experiments in both the insurance pricing domain and a synthetic pricing environment. Our results suggest that while φ -DRO provides theoretical robustness, its practical benefits in offline pricing may be limited in certain problems
Observational ozone data over the global oceans and polar regions : The TOAR-II Oceans data set version 2024
No full textEnhancing Oral Presentation Skills Through Generative AI
Get PDFThis empirical study investigates the impact of generative AI on students' oral presentation skills within a university language degree program. Participants in this study were divided into two groups based on their choice of using generative AI: the experimental group (EG), who opted to use AI tools for support, and the control group (CG), who prepared independently without AI assistance. Quantitative data from pre- and post-tests revealed no significant difference in pre-test scores. Both groups demonstrated improvements in the post-test, with slightly greater gains observed in the EG. Notably, frequent AI users outperformed those who used AI sparingly. Thematic analysis of qualitative data from students’ reflective accounts and informal interviews highlighted generative AI’s role in supporting content development, including creating structured outlines, generating background visuals, providing relevant information, and suggesting idiomatic expressions. EG students reported that using generative AI encouraged critical thinking when selecting appropriate materials, evaluating AI-generated content, and refining prompts. However, challenges were also unveiled, including a lack of natural flow in AI-generated speech and occasional contextual or historical inaccuracies. The primary reason cited by CG students for not using AI was the concern over potential plagiarism. These findings underscore the importance of prompt refinement and the necessity for targeted AI literacy training. Students perceived generative AI as beneficial for oral language practice and presentation skill development. The study highlights the value of experiential learning approaches that encourage critical engagement with AI-generated content. Further research could explore the long-term effects of integrating generative AI into language education and identify best practices for its effective use
Subjective and objective memory in a community-derived sample of people with epilepsy : Evidence from the crimes and four doors tests
No full textSubjective self-reports of difficulties with memory are relatively common in people with epilepsy, though these do not always align with performance on objective memory tasks. The current study gathered qualitative and quantitative subjective reports of memory function in a group of people with epilepsy who were recruited via the charity Epilepsy Action, along with controls. Participants also carried out one of two recently developed experimental tasks (Crimes or Four Doors) that provide objective measures of long-term memory and forgetting, along with an additional verbal learning and recall task, each of which assess retention over a one-week period. Relative to controls, people with epilepsy reported memory problems across all the subjective measures, while also showing more objective forgetting on Crimes and Four Doors. When combining the epilepsy and control samples, subjective forgetting and memory satisfaction correlated with objective delayed recall and forgetting. Within the epilepsy sample, delayed recall correlated with subjectively experienced forgetting. These findings provide new evidence for subjective and objective memory difficulties in epilepsy and indicate the need for development of appropriate tools to detect atypical forgetting in this population
Suicide risk assessment and management : guidelines and experiences
No full textThe current NHS long-term plan places an emphasis on the reduction of suicide, however suicide rates in England do not appear to be reducing. This thesis considers reasons for this, which may relate to inconsistent and incoherent published guidance, as well as psychosocial factors affecting clinicians who undertake suicide risk assessment and management, and discusses how these issues may be addressed. Section one presents a systematic policy review of guidelines and policy influencing suicide prevention in healthcare settings in England, considering the extent to which these could be considered consistent and coherent. 21 papers were located using systematic searches of three academic databases and several online archives. These were scrutinised using quality appraisal, thematic analysis of content and the generation of key areas of inconsistency. Overall quality of documentation was considered to be low. Key inconsistencies related to how guidance regarding conduct of individualised risk assessment and management, may be incompatible with other practical considerations and the context within which guidance is followed. Section two presents a qualitative study investigating the effects of exposure to service user attempted or completed suicide, on professionals’ subsequent conduct of suicide risk assessment and management. 10 mental health professionals were interviewed regarding their experiences of service user attempted or completed suicide and subsequent work with suicide risk, with data analysed using reflexive thematic analysis. The study was informed by Perceptual Control Theory, which influenced the interview technique as well as analysis and discussion. Key findings related to how professionals were able to resolve difficulties through reflective methods and support, and conduct effective risk assessment and management going forward. A critical appraisal is presented in section three. This discussed key issues from sections one and two, including methodological considerations, strengths and limitations, relevance to clinical psychology and areas of future research
A millimeter-wave end-fire dual-polarized array antenna with symmetrical radiation patterns and high isolation
Get PDFThis paper presents a millimeter-wave end-fire dual-polarized (DP) array antenna with symmetrical radiation patterns and high isolation. The DP radiation element is formed by integrating a quasi-Yagi antenna (providing horizontal polarization) into a pyramidal horn antenna (providing vertical polarization), resulting in a DP radiation element with a symmetrical radiation aperture. To efficiently feed the DP element while maintaining high isolation, a mode-composite full-corporate-feed network is employed, comprising substrate-integrated waveguide supporting the TE10 mode and substrate-integrated coaxial line supporting the TEM mode. This design eliminates the need for additional transition structures, achieving excellent mode isolation and a reduced substrate layer number. A 1 × 4-element DP array prototype operating at 26.5–29.5 GHz using low temperature co-fired ceramic technology was designed, fabricated, and measured. The test results indicate that the prototype achieves an average gain exceeding 10 dBi for both polarizations within the operating band. Thanks to the symmetrical DP radiation element and mode-composite full-corporate-feed network, symmetrical radiation patterns for both polarizations are observed in both the horizontal and vertical planes, along with a high cross-polarization discrimination of 22 dB and polarization port isolation of 35 dB
Geotemporal Design : Creating new imaginaries of urban nature through environmental light and sound
Get PDFThis paper introduces Geotemporal Design, a critical and speculative framework that foregrounds the interplay between environmental light, sound, and urban nature. While urban environments are often thought of as anthropogenic environments, this research contextualises urban life within broader ecological and planetary rhythms. Through an interdisciplinary approach combining cultural geography, environmental humanities, and speculative design, the research captures temporal patterns of human, biological, geographical, and technological activities using environmental sensors and creative visualizations. Drawing on projects such as Sonorous Landscapes (Griffiths, 2024), the paper explores how capturing diurnal, circalunar, and seasonal variations in urban light and sound can create new imaginaries of urban nature. Speculative artifacts are used to move from human-centered urban narratives to more-than-human ecological narratives, fostering increased environmental awareness and stewardship among communities and urban designers. Findings from participatory workshops demonstrate the effectiveness of these methods in making ecological rhythms legible and facilitating an atunement to biodiversity and urban ecologies. Geotemporal Design proposes an innovative pathway toward more socially and ecologically integrated urban futures, suggesting new methodological directions for environmental engagement, urban design, and policy making