Hybrid software architecture with peripheral computing for adaptive VR driving training systems with biometric feedback
Gespeichert in:
| Titel: | Hybrid software architecture with peripheral computing for adaptive VR driving training systems with biometric feedback |
|---|---|
| Autoren: | Artur Maliuha |
| Quelle: | Авіаційно-космічна техніка та технологія, Vol 0, Iss 5, Pp 96-111 (2025) |
| Verlagsinformationen: | National Aerospace University «Kharkiv Aviation Institute», 2025. |
| Publikationsjahr: | 2025 |
| Bestand: | LCC:Motor vehicles. Aeronautics. Astronautics |
| Schlagwörter: | адаптивна віртуальна реальність, біометричний зворотний зв'язок, навчання водіїв, периферійні обчислення, гібридна архітектура, оптимізація затримки, машинне навчання, мікросервіси, масштабованість, класифікація напружень, Motor vehicles. Aeronautics. Astronautics, TL1-4050 |
| Beschreibung: | The subject of the article is the methodology for developing a hybrid three-tier software architecture for adaptive VR driving training systems with real-time biometric feedback, which ensures the distribution of functional components between the client, peripheral and microservice levels. The goal is to develop an architecture that provides a balance between minimal latency of biometric data processing for a comfortable VR experience, reduced costs, and the possibility of gradual scaling. Tasks to be solved: performing a comparative analysis of existing architectural approaches according to the criteria of latency, cost and scalability; developing a hybrid software architecture model; experimental verification of the proposed architecture; determining the scope of application of the hybrid architecture. The methods used include architectural design methods, synthesis and decomposition methods. The following results were obtained. A hybrid architecture was developed with a client layer for rendering and collecting biometric data, a peripheral layer for classifying user state and adapting training scenarios, and a microservice layer for asynchronous traffic simulation and data analytics with possible scalability. Experimental verification demonstrated a reduction in the latency of biometric data processing. The scope of application of the developed hybrid model was determined. Conclusions. The results of the study confirmed the relevance of a hybrid approach to architectural design with the placement of computationally intensive components on a peripheral server in the local network of the educational institution instead of a centralized cloud server to eliminate global network delays, and also provided an opportunity to justify the choice of software architecture taking into account the number of users, budget constraints, and latency requirements. The scientific novelty of the results obtained lies in the creation of a hybrid architecture model that involves the distribution of functional components of the system by levels based on constraints on processing speed, implementation cost, and the degree of independence of auxiliary calculations, which makes it possible to increase the efficiency of resource use in training using virtual reality, as well as to increase the efficiency of training by adapting to the current state of the learner. |
| Publikationsart: | article |
| Dateibeschreibung: | electronic resource |
| Sprache: | English Ukrainian |
| ISSN: | 1727-7337 2663-2217 |
| Relation: | http://nti.khai.edu/ojs/index.php/aktt/article/view/3149; https://doaj.org/toc/1727-7337; https://doaj.org/toc/2663-2217 |
| DOI: | 10.32620/aktt.2025.5.09 |
| Zugangs-URL: | https://doaj.org/article/d77fc281669d4a348940fe478d8cd1d4 |
| Dokumentencode: | edsdoj.77fc281669d4a348940fe478d8cd1d4 |
| Datenbank: | Directory of Open Access Journals |
Nájsť tento článok vo Web of Science | Abstract: | The subject of the article is the methodology for developing a hybrid three-tier software architecture for adaptive VR driving training systems with real-time biometric feedback, which ensures the distribution of functional components between the client, peripheral and microservice levels. The goal is to develop an architecture that provides a balance between minimal latency of biometric data processing for a comfortable VR experience, reduced costs, and the possibility of gradual scaling. Tasks to be solved: performing a comparative analysis of existing architectural approaches according to the criteria of latency, cost and scalability; developing a hybrid software architecture model; experimental verification of the proposed architecture; determining the scope of application of the hybrid architecture. The methods used include architectural design methods, synthesis and decomposition methods. The following results were obtained. A hybrid architecture was developed with a client layer for rendering and collecting biometric data, a peripheral layer for classifying user state and adapting training scenarios, and a microservice layer for asynchronous traffic simulation and data analytics with possible scalability. Experimental verification demonstrated a reduction in the latency of biometric data processing. The scope of application of the developed hybrid model was determined. Conclusions. The results of the study confirmed the relevance of a hybrid approach to architectural design with the placement of computationally intensive components on a peripheral server in the local network of the educational institution instead of a centralized cloud server to eliminate global network delays, and also provided an opportunity to justify the choice of software architecture taking into account the number of users, budget constraints, and latency requirements. The scientific novelty of the results obtained lies in the creation of a hybrid architecture model that involves the distribution of functional components of the system by levels based on constraints on processing speed, implementation cost, and the degree of independence of auxiliary calculations, which makes it possible to increase the efficiency of resource use in training using virtual reality, as well as to increase the efficiency of training by adapting to the current state of the learner. |
|---|---|
| ISSN: | 17277337 26632217 |
| DOI: | 10.32620/aktt.2025.5.09 |