Person identification is a pivotal challenge in the security domain, with important and impactful applications such as identifying crime suspects and locating missing persons. One convenient person identification method is gait identification, where individuals are identified by their unique walking style. However, traditional methods of gait identification are often affected by variations in appearance and occlusion. This work introduces a novel and robust spatiotemporal kinematics‐informed non‐invasive gait identification (STONI‐GID) method that uses human pose estimation, occlusion state estimation and deep machine learning. Furthermore, unlike some existing methods, we demonstrate that our method remains unaffected by everyday appearance changes, environment, or viewing angle. Our approach achieved identification accuracy of up to 98.66% when evaluated using our primary dataset of 65 diverse participants in real‐world environments. Moreover, the model outperformed existing methods during cross‐dataset validation on the large Southampton dataset and the Gait Recognition Image and Depth Dataset (GRIDDS), achieving identification accuracies of 97.68% and 99.12%, respectively. Our findings will particularly advance the research frontiers of real‐world gait identification and impact interdisciplinary areas of security and healthcare applications.

This paper presents a sign language translation system developed to enhance accessibility for lesser-supported sign languages, with a specific focus on Indian Sign Language. Given that India has the world's largest deaf population, the communication gap between sign language users and non-users poses a significant barrier. The objective of this research is to examine existing support systems for Indian Sign Language and to build a system capable of detecting, recognizing, and translating signed words through the use of connected cameras. To achieve this goal, image classification using transfer learning techniques was applied to a small dataset of Indian Sign Language gestures, resulting in an average recognition accuracy rate of 83%. Over 10,000 images across 12 different gestures were collected, and the training/validation process reached peak testing accuracy levels of 95%. The final model was deployed into a web-based application specifically designed for use in areas with limited access to advanced technology, aiming to reduce communication barriers and improve inclusivity for the deaf and hard-of-hearing community.

This paper presents a semi-automated feedback system designed to support programming education by generating personalised feedback using rule-based analysis and Natural Language Generation (NLG). The centralised system, accessible across platforms via a web browser within institutions, evaluates Python code using Abstract Syntax Trees (AST), checking for syntax, indentation, comments, and required constructs. It also performs plagiarism checks and tracks student performance with data visualisation tools. Built with Django, the system ensures accuracy, consistency, and instructor oversight. Unlike existing tools, the system combines AST-based analysis, instructor-defined rules, and rule-based NLG in a modular and open-source platform designed for flexibility and scalability. Results show improved feedback efficiency and instructional value.

This project presents a low-cost, end-to-end autonomous driving prototype built on an RC car platform powered by a Raspberry Pi 4. The system uses a front-facing USB camera and an HC-SR04 ultrasonic sensor for perception, combining visual and distance data through late-fusion logic to interpret traffic lights and avoid nearby obstacles. A custom dataset of traffic light images was used to fine-tune an SSDMobileNetV2$(320 \times 320)$model using TensorFlow, achieving a mean average precision of 0.812 on COCO metrics. The model was quantized to INT8, converted to ONNX format, and deployed with ONNX Runtime for efficient inference. Realtime tests showed an average latency of 286 ms and stable performance under thermal limits. The car reliably stopped on red or amber lights and moved forward only on confirmed green signals with a clear path. The results validate real-time autonomous navigation on low-cost embedded hardware and provide a reproducible baseline for future edge$A I$and robotics research.

The increasing prevalence of motor impairments due to stroke, spinal cord injuries, and neurodegenerative conditions has underscored the urgent need for accessible, noninvasive rehabilitation technologies. This study presents a computational framework for a brain-computer interface (BCI) system that analyses electroencephalography (EEG) signals with machine learning (ML) to simulate the control of a robotic hand for upper-limb rehabilitation. Leveraging a publicly available dataset for motor imagery (MI) and a simulated approach for steady-state visual evoked potential (SSVEP) data, the system extracts user intentions through ERD/ERS and frequency-domain features, which are then classified using models such as linear discriminant analysis (LDA), support vector machines (SVM), and convolutional neural networks (CNN). These outputs are translated into simulated commands for a servo-actuated artificial hand. Offline evaluation demonstrated high classification accuracy, with SVM achieving up to 92.3% on motor imagery (MI) tasks and 94.1% on SSVEP tasks. The simulated end-to-end latency per command was estimated at 1100-1250 ms for SSVEP and 2150 ms for MI. The results confirm the feasibility of the machine learning pipeline for a portable, low-cost, ML-powered BCI system and provide a robust, ethically straightforward foundation for future hardware integration. This research contributes to the development of adaptable, user-centred assistive technologies with the potential for home-based deployment and clinical integration.

In recent years, augmented reality has received a considerable lot of interest. Augmented reality application cases are expanding all the time. People may now readily experience augmented reality because to the widespread availability of smartphones. This enables scholars from all around the world to use it. The use of augmented reality-based interactive technologies in education is on the rise. In this study, we proposed a method that allows students to learn the human skeletal system and anatomy using their smartphones. The proposed method offers visualizing the human skeletal system, including bones, in 3D using augmented reality and also interacting with the 3D bones. The substantial finding of the study reflects on how students can benefit from this Augmented Reality-based interactive method. Qualitative assessment was accumulated from 159 medical students who have experience with the traditional human skeleton learning experience and also participated in the augmented reality-based human skeleton system.

This paper presents the development and evaluation of a simulated inventory management system using NetLogo, with a focus on demonstrating the potential of TurtleBot 4 robotics to optimize stock control in warehouse environments. Faced with technical challenges in using a physical TurtleBot 4, the project shifted to a simulation approach, which allowed for a detailed exploration of how agent-based models can improve inventory management processes. Drawing on research into SLAM algorithms, real-world business practices, and inventory management systems, the simulation replicates key warehouse functions, including the receiving, storing, and dispatching of goods. The project’s design includes a graphical user interface (GUI) that simulates wireless data transfer, enabling real-time interaction with the system. The artefact was tested successfully in various scenarios, highlighting the potential of robotics to enhance efficiency and streamline operations. The artefact overall performed as intended, providing valuable insights into the future of robotic integration in inventory management.

Sign languages are languages that utilize the visual-manual modality to convey meaning. These languages are expressed through manual articulations combined with non-manual elements. Sign languages are complete natural languages, possessing their own grammar and lexicon. They are primarily used by individuals who are Deaf or have hearing impairments. Sign languages are not universal and are not mutually intelligible with one another, though they do exhibit striking similarities among them. In the context of Britain, the most prevalent form of Sign Language is known as British Sign Language (BSL). BSL possesses its own distinct grammatical structure and syntax; as a language, it is neither dependent on nor closely related to spoken English. This paper offers a comprehensive review of sign language systems, including an exploration of related studies on British Sign Language, as well as an examination of the legal, social, and ethical considerations associated with these languages.

Cooperation to access wireless networks is a key approach towards optimizing the use of finite radio spectrum resources in overcrowded unlicensed bands and to help satisfy the expectations of wireless users in terms of high data rates and low latency. Although solutions that advocate this approach have been widely proposed in the literature, they still do not consider a number of aspects that can improve the performance of the users’ connections, such as the inclusion of (1) cooperation among network operators and (2) users’ quality requirements based on their applications. To fill this gap, in this paper we propose a centralized framework that is aimed at providing a “win‐win” cooperation among Wi‐Fi and cellular networks, which takes into account 5G technologies and users’ requirements in terms of Quality of Service (QoS). Moreover, the framework is supported by smart Radio Access Technology (RAT) selection mechanisms that orchestrate the connection of the clients to the networks. In particular, we discuss details on the design of the proposed framework, the motivation behind its implementation, the main novelties, its feasibility, and the main components. In order to demonstrate the benefits of our solution, we illustrate efficiency results achieved through the simulation of a smart RAT selection algorithm in a realistic scenario, which mimics the proposed “win‐win” cooperation between Wi‐Fi and cellular 5G networks, and we also discuss potential benefits for wireless and mobile network operators.