El programa de Doctorado en Ingeniería y Tecnologías Industrial, Informática y Civil organiza un nuevo seminario de investigación interdisciplinar.
En esta ocasión, bajo el título "Self-adaptive and context-aware intelligent training systems in sensorised immersive virtual reality environments for occupational risk prevention", la docente e investigadora del Área de Lenguajes y Sistemas Informáticos de la Universidad de Burgos Gadea Lucas Pérez, abordará el desarrollo de un sistema de formación auto-adaptativo en Realidad Virtual Inmersiva (RVI) para la prevención de riesgos laborales en la operación de maquinaria de carga. La ponencia se impartirá en inglés.
La actividad tendrá lugar en la Sala de juntas 1 (edificio A1) de la Escuela Politécnica Superior de Río Vena, el jueves 10 de abril de 2025, a partir de las 12:00 h. Además, podrá seguirse en Microsoft Teams a través del enlace.
Resumen:
The primary goal of occupational risk prevention (ORP) is to provide workers with the necessary training to develop positive health and safety habits. This is particularly critical in the technical instruction required for operating new loading machinery, such as forklift trucks and telescopic handlers.
Previous research has demonstrated that adaptive learning—where problems, stimuli, or tasks are adjusted based on the learner’s performance—significantly improves training outcomes. Immersive Virtual Reality (IVR) offers the possibility of simulating complex situations; however, current implementations are often rigid, providing tailored solutions for specific scenarios that fail to accommodate diverse learning styles, including differences in experience, risk perception, and gender.
In this research, a self-adaptive IVR training system for machinery handling has been developed, designed to dynamically adjust to users’ individual needs in real time. The main challenges lie in integrating multiple data sources to accurately model the user's state in real time, identifying precise measurement systems that do not interfere with the learning process, and adapting the virtual environment in real time to ensure a truly responsive and adaptive experience. This presentation will discuss these challenges, as well as the methods used for user modeling, which rely on multimodal datasets gathered from participants performing various training exercises. Additionally, we will present an advance of the first results obtained, highlighting the system’s effectiveness and areas for further refinement.