26013 research outputs found
Sort by
A Vision-Guided Deep Learning Framework for Dexterous Robotic Grasping Using Gaussian Processes and Transformers †
Get PDFRobotic manipulation of objects with diverse shapes, sizes, and properties, especially deformable ones, remains a significant challenge in automation, necessitating human-like dexterity through the integration of perception, learning, and control. This study enhances a previous framework combining YOLOv8 for object detection and LSTM networks for adaptive grasping by introducing Gaussian Processes (GPs) for robust grasp predictions and Transformer models for efficient multi-modal sensory data integration. A Random Forest classifier also selects optimal grasp configurations based on object-specific features like geometry and stability. The proposed grasping framework achieved a 95.6% grasp success rate using Transformer-based force modulation, surpassing LSTM (91.3%) and GP (91.3%) models. Evaluation of a diverse dataset showed significant improvements in grasp force modulation, adaptability, and robustness for two- and three-finger grasps. However, limitations were observed in five-finger grasps for certain objects, and some classification failures occurred in the vision system. Overall, this combination of vision-based detection and advanced learning techniques offers a scalable solution for flexible robotic manipulation
Opening the Dialogue on Death: Navigating the Journey of Bereavement, Grief and Trauma Within Probation Delivery
Get PDFPrivacy Strategies for Police Personnel: Co-designing a Self-Assessment Tool
Get PDFThis article presents a study aimed at co-designing technological capabilities to support the protection of UK police officers and staff in public-facing roles. Firstly, we examined the digital literacy of police personnel, their experiences with online harms, and their requirements for a tool to equip them with privacy strategies in their use of digital platforms. These findings informed the design and development of a Self-assessment Tool that enhances police personnel’s privacy awareness and literacy. We discuss the challenges encountered in tailoring the tool to their specific needs and present the final product. The tool is a critical component of a broader strategy to reduce online harms experienced by police personnel as public-facing professionals
Social Robot Assistant for Group Interactions with Secondary School Students: A Participatory Design Study
Get PDFThe application of social robots in group settings is an emerging area of research with the potential to transform numerous fields, particularly education. This paper explores the potential of social robots as assistants in collaborative group interactions among secondary school students through a Participatory Design (PD) study. This was achieved by conducting a focus group (10 participants, ages 11 to 15 years) that included discussions, robot interactions, and co-design activities. The findings reveal the students’ challenges in group interactions, and their perceptions of how robots could assist them. The first part of the focus group was a exploration and co-design stage to encourage participants to discuss about and interact with social robots for group collaboration. This stage highlighted some of the challenges students face during group work and how they believe a social robot could assist them. The second part of the focus group involved getting the participants to discuss and co-design robot behaviours for a specific group collaborative task. This revealed the participants emphasis on the robot behaviours being clear, specific and relevant for the task. These insights contribute to the design of effective social robots for group collaborative settings for this user group
Translation, Cross-Cultural Adaptation, and Validation of the Simplified Chinese Version of Sport Concussion Assessment Tool 6.
No full textObjective:
This study aimed to translate, culturally adapt, and validate the latest Sport Concussion Assessment Tool 6 (SCAT6) into simplified Chinese.
Design:
This study adopted a mixed design with both qualitative and quantitative methods. The quantitative part of the study adopted a repeated measures design.
Setting:
The translation and revision process were completed through online communications. The validation process was performed in person at the athletes' normal training facilities.
Participants:
Thirty-two healthcare professionals and twenty-one athletes were invited into the expert panel and end-user panel for the qualitative part of the study. Thirty-nine elite or collegiate athletes were included in the quantitative part of the study.
Independent Variables:
No independent variables were assumed in the translation and the expert appraisal processes. Administrators were the independent variables for the inter-rater study and the testing sessions for the test–retest study.
Main Outcome Measures:
Expert opinions were collected during the translation and appraisal process. Testing scores and time of completion were obtained from test–retest study and inter-rater examinations.
Results:
Revision and alterations of the Chinese SCAT6 and supplementary materials were made according to the reviews from experts and end users. The internal consistency and the inter-rater reliability of the simplified Chinese SCAT6 for all subscales were excellent [intraclass correlation (ICC) ranged from 0.83 to 0.999], the test–retest reliability for all subscales were moderate to excellent (ICC ranged from 0.51 to 0.92). The average time to complete was 20.2 ± 3.5 minutes.
Conclusion:
The simplified Chinese SCAT6 is a valid sport concussion assessment instrument for mainland China end users
Comparative potential of biogas production from the distillery, fruit and vegetable waste and their mixtures (digestion).
Get PDFBiogas is becoming increasingly important as a renewable energy source in the face of global warming and declining fossil fuel reserves. Biogas is produced by anaerobic digestion of organic materials which can be available from various wastes such as agro-industrial, human, fruit waste, distillery, animal waste and aquatic plants. This study deals particularly with the comparative potential of biogas production from distillery, fruit and vegetable waste and their mixtures (digestion). The materials used as feed in this research were distillery waste which is dark-colored liquid waste from Desta Alcohol and Liquor Factory Private Limited Company. Fruit and vegetable waste such as banana peels, papaya, mango, tomato, avocado, cabbage leaves, watermelon skin, and orange skin were collected from juice houses and fruit and vegetable wholesale markets in Mekelle City, and Cow manure used as a buffer solution, collected from Desta Alcohol and Liquor Factory PLC. Waste samples were characterized for total solids, volatile solids, pH, biochemical oxygen demand, and chemical oxygen demand according to established standards. Biogas was analyzed using a biogas analyzer, an ORSAT apparatus for CO2, and a TUTWILER apparatus for H2S. Finally, the %CH4 was calculated from 100 % by ignoring other gases. The maximum biogas production from all wastes was observed at 37 °C. Mixture (co-digestion) produced high biogas in litter (L): 6.95, 9.47 and 9.54 at 20 °C, 37 °C and 50 °C respectively. The maximum methane composition was observed from the co-digestion (M) in (%) 67, 70 and 70.3 at 20 °C, 37 °C and 50 °C respectively. Methane yield was calculated at both temperature and substrates (waste). Comparatively, maximum methane yield was observed at 37 °C for distillery waste, fruit vegetable waste and mixture(digestion); 0.032, 0.061 and 0.079 L per gram volatile solids digestion (LCH4/gVS) respectively
Document clustering with evolved multi-word search queries
Get PDFText clustering holds significant value across various domains due to its ability to identify patterns and group related information. Current approaches which rely heavily on a computed similarity measure between documents are often limited in accuracy and interpretability. We present a novel approach to the problem based on a set of evolved search queries. Clusters are formed as the set of documents matched by a single search query in the set of queries. The queries are optimized to maximize the number of documents returned and to minimize the overlap between clusters (documents returned by more than one query). Where queries contain more than one word they are interpreted disjunctively. We have found it useful to assign one word to be the root and constrain the query construction such that the set of documents returned by any additional query words intersect with the set returned by the root word. Not all documents in a collection are returned by any of the search queries in a set, so once the search query evolution is completed a second stage is performed whereby a KNN algorithm is applied to assign all unassigned documents to their nearest cluster. We describe the method and present results using 8 text datasets comparing effectiveness with well-known existing algorithms. We note that as well as achieving the highest accuracy on these datasets the search query format provides the qualitative benefits of being interpretable and modifiable whilst providing a causal explanation of cluster construction