Quantised small language models (SLMs) deployed at the edge are increasingly important for enabling high performance AI agents in interactive virtual worlds, where low latency operation is key for seamless user experience. This paper investigates the deployment of quantised SLMs at the edge to meet performance requirements and maintaining high response quality to support real time conversational agent interactions in an immersive virtual world. We developed a RAG based back-end AI system to support virtual agent conversational capabilities in a virtual world prototype, and experimented with two SLM families and three quantisation levels (8-bit, 4-bit, and 3-bit) to evaluate system performance through end-to-end latency, time-to-first-token, throughput, and memory headroom. We also assess response quality via expert evaluations of accuracy, relevance, safety, and human alignment against non-quantised baselines. Results show that moderate quantisation enables performant practical edge deployment with good response quality, demonstrating that edge-deployed quantised SLMs can support real-time narrative interactions and remain practically useful for virtual-world agents, providing evidence for deploying high performance conversational AI in immersive environments.

Conducting real-time wireless propagation simulation in interactive 3D environments is computationally demanding , and this challenge becomes more critical in Virtual Reality (VR), where stable and high frame rates are required for usability and comfort. This paper presents VRadioSim, a GPU-based wireless propagation simulator prototype that supports real-time visualisation of multipath rays and received-signal heatmaps in desktop and tethered VR modes. The prototype loads 3D scenes from GLB models and enables interactive placement of transmitter and receiver entities, visualisation of signal accumulation across scene surfaces, and point-based querying of received power values. A preliminary evaluation across varying scene complexities and ray counts indicates that the desktop version remains consistently interactive, and that tethered VR is currently most practical for lower complexity scenes. Signal sampling measurements show that received-power estimates are mostly consistent across different geometrical scene representations, suggesting stability of the current GPU-based propagation implementation. Overall, the findings demonstrate the practical feasibility of prototype for interactive desktop analysis and provide an initial basis for further optimisation and validation in immersive VR settings.

Digital Twin technology has been used in many different domains to integrate disparate and complex data sources into a single, comprehensible system. A less explored field is monitoring indoor air quality and building efficiency by providing real-time visualization and spatially contextualized data in commercial buildings. Managers of commercial buildings often have little or no training in indoor air quality, heating ventilation and air conditioning systems (HVAC), or building efficiency. This makes it difficult for them to evaluate environmental and efficiency data, if such data are available. This study evaluates the effectiveness of Digital Twins in 2D Desktop and 3D Virtual Reality to support non-expert user engagement, data interpretation, and spatial understanding for indoor air quality and building efficiency analysis. Sensors were installed in a mixed-use office/light industrial space, with a system infrastructure that collects and stores historical data readings. A comparative study was conducted between i) a traditional web-based dashboard (the control condition), ii) a 2D, and iii) a Virtual Reality Digital Twin using a within-subject research design involving non-expert participants, to determine the efficacy levels of user engagement with the data presentation mechanisms of each system, their spatial understanding, and their ability to conduct data interpretation. The findings demonstrate that VR Digital Twin supports increased spatial awareness and understanding, immersion, and data discovery, and the 2D Digital Twin system provides clarity and accessibility for trend analysis. The study contributes by i) validating the role of Digital Twins for Indoor Air Quality; ii) exploring user perceptions on different Digital Twin mediums in terms of user engagement; and iii) highlighting the potential of Digital Twins for public safety, particularly in indoor environments.

The Metaverse is significantly impacting the field of Digital Cultural Heritage, providing new opportunities for preserving, presenting, engaging, and interacting with historical and cultural artefacts through immersive virtual environments. This paper presents the development and implementation of an Intelligent Reality Virtual Museum prototype system for interactive digital cultural heritage experiences, within the context of a Cyber-Physical-Social System (CPSS) architectural framework for Metaverse applications development. The prototype integrates advanced Artificial Intelligence and Large Language Models in fusion with Digital Twins, Robotic and Gaming technologies to deliver immersive cultural heritage experiences that blend the real with the digital worlds. The paper presents the system architecture, system design, and development process, and the results of a user evaluation study assessing the usability and user engagement with the system during an interactive cultural heritage experience. The results revealed a usable system capable of offering an engaging and entertaining interactive experience to its users, highlighting the potential to be used for supporting and enhancing cultural heritage accessibility and engagement in our current digital age. System limitations and plans for future work are presented, focusing on refining and expanding the system capabilities towards the development of a new type of a Cyber-Physical-Social Eco-Society of systems for Metaverse applications to support and enhance the way users interact with and experience cultural heritage.

VR gaming is rapidly growing in popularity, with puzzle-based games featuring optical illusions attracting gamers attention. Optical illusions are an emerging gameplay mechanic, challenging players' perceptual and cognitive abilities. While widely used in 2D/3D games, their integration into VR remains underexplored. This study investigates the use of optical illusion challenges in a VR game, focusing on players' workload demands, possible onset of VR sickness, and overall experience. A not-so-serious educational game featuring optical illusion challenges as core tasks for progression was developed for VR and PC, and a comparative evaluation was conducted. Results revealed that these challenges provided an enjoyable and immersive experience, with generally low workload demands. However, the complexity and disorienting nature of some illusions occasionally increased physical and mental effort, triggering VR sickness in some players. These findings offer valuable insights for designing VR games, highlighting the importance of balancing challenge and comfort when implementing such mechanics.

Over the years, the wireless communication industry and the research community investigated methods of creating accurate and efficient models for signal propagation. The recent advancements in wireless communications and its exponential usage through high mobility of numerous connected devices introduced challenges in simulating dynamic radio propagation. To address these challenges, specialized software have been developed, offering high-fidelity simulations. However, these solutions have expensive cost requirements and are largely dependent on offline computations, lacking flexibility and scalability. As a result, their wider use in scientific and industrial sectors is limited. In response to these limitations, this paper proposes an alternative solution leveraging the latest developments in Gaming technologies, GPU technology, and Virtual Reality through the concept of Digital Twins to develop a prototype for a deterministic channel simulator. The prototype utilise a game development engine, high-performance GPU, and commercial VR headsets to achieve a low cost, accessible and scalable method for visualizing wireless signal propagation in real time. This paper presents the work in progress, describing the system architecture, current state of development and intended functionalities.

This paper describes the Work in Progress (WiP) in the development of an educational Cyber-Physical-Social Eco-System (CPSeS) using XR, AI, and Robotic technologies to support immersive learning. It is presenting the framework of the proposed CPSeS used to create an intelligent educational virtual environment which seamlessly blends the real with virtual worlds and is influenced by its users, real and artificial agents, and elements. The paper is an updated report of the WiP presented at iLRN 2020 conference, focusing on system additions and improvements, its current development status, and future plans and aspirations.

This work-in-progress paper describes the development of a conceptually led Cyber-Physical-Social EcoSystem (CPSeS) that seamlessly blends the real with the digital worlds using VR, Robots and Social Networking technologies to support immersive learning. The proposed CPSeS can bring students together by merging the real with virtual social spaces, enabling them to interact with each other, and with real and digital environments through physical and virtual robots. The framework used to design this system provides opportunities for developing innovative interactive systems and educational approaches to support and enhance immersive learning.

The use of educational Multi-User Virtual Environments that provide synchronous interaction, interactive and social learning experiences have the potential to increase student engagement. Due to increased social and cognitive presence, the use of such environments can result in greater student engagement when compared to traditional asynchronous learning environments. In this work, we hypothesized that students' engagement in collaborative learning activities will increase if Transactive Memory System constructs are present. Thus, we employed the theory of TMS that emphasizes the importance of Specialization, Coordination and Credibility between members in a team. The results show that there is a significant correlation between the development of TMS and students' engagement.

This paper presents the preliminary findings of an empirical study that aims to evaluate the efficacy of 3D virtual worlds for synchronous distant learning activities. A pilot study was conducted to evaluate the prototype developed, the procedures and instruments of data collection, to prepare the study investigating whether virtual worlds can help students participate more effectively in learning activities. The theoretical framework is presented, together with the proposed prototype, method and results of the pilot study, its associated limitations and the future work that we intend to perform.