This work focuses on the analysis of Mobile Pervasive Augmented Reality Systems (MPARS), that is, Augmented Reality systems to be used when the user is in motion, namely in dynamic activities such as outdoor sports. In this context, the integration of functionalities such as a hands-free tool for voice recognition is an asset. However, presenting the “right information at the right time” remains the biggest challenge facing these systems. Considering the current context, a functional system needs to perform adequate filtering of the information to be provided to the user in real-time. As a first step towards specifying such a system in an outdoor sports setting, this work focused on the development of a specific technology adoption model for the use of pervasive augmented reality applications. Such a model proved beneficial to assess user preferences in these scenarios and develop the first MPARS architecture, based on the metrics found to integrate user preferences, as well as proposing the concept of meaningful feedback to avoid information overload. To correctly decide what information to provide at each time, it is necessary to automate various context-dependent tasks, which led to the proposal to use Machine Learning methods to assist in providing real-time meaningful feedback to the user, reducing unwanted information and improving Quality of Experience. The final step aimed to create a proof of concept for a specific MPARS, using some outdoor sports activities as use cases. This prototype is based on heuristics that combine indicators of technological adoption and customizable automation, based on the surrounding context and user preferences, so that useful/desired information at any given time can be presented in real-time on the application layout. After carrying out a set of experiments using feedback from volunteer users, it was possible to prove the viability of the proposed system, showing that the model enables the generation and use of data with predictive capacity for what information should be presented at what time, increasing the quality of the user experience.

--

Mobile,Augmented Reality,Technology Acceptance,Context-Awareness,Sensing,Automatic Learning.

• Ciências da Computação e da Informação - Ciências Naturais