No abstract

The recent use of mobile devices and the widespread adoption of wireless computer networks have stimulated mobile and ubiquitous computing research. In this context, education is being considered one of the main application areas. The ability to accurately track the learner's position through a location system and the context awareness support allow the discovery of new pedagogical opportunities. Nowadays, there are several approaches to the organization of ubiquitous computing environments, even though the majority of them employ a centralized architecture. This approach brings some benefits, like easing the access to the learner profiles, but it also has some disadvantages intrinsic to the model, like scalability concerns and the physical separation between the learners and their profile data. This article proposes a collaboration model created for decentralized environments, capable of improving the learning process through the stimulus to the interaction between learners. This approach is modeled using agents, employing on a socio-interactionist model based on the "most capable pair" concept first introduced by Vygotsky.

The application of ubiquitous technologies in the improvement of education strategies is called Ubiquitous Learning. This article proposes the integration between two models dedicated to support ubiquitous learning environments, called Global and CoolEdu. CoolEdu is a generic collaboration model for decentralized environments. Global is an infrastructure designed to create ubiquitous learning environments. Global provides software agents that perform tasks common to ubiquitous learning processes. By extending these agents or adding new ones, a system can be specialized to support ubiquitous learning environments. The CoolEdu/Global integration created a collaborative and decentralized ubiquitous learning environment. The resulting environment was evaluated through a simulated scenario dedicated to explore its functionalities. The results were encouraging and showed the potential of deploying the environment in real situations.

No abstract

GLOBAL é uma infraestrutura descentralizada para ambientes de aprendizagem ubíqua, baseada em agentes de software. A partir da extensão dos seus agentes ou adição de novos, a infraestrutura pode ser especializada para a criação de ambientes de aprendizagem ubíqua. Este artigo contribui apresentando a modelagem da infraestrutura através de uma metodologia multiagentes. Nesse trabalho, GLOBAL foi completamente modelada usando a metodologia Prometheus. Além disso, o artigo discute como os agentes foram implementados e como foram integrados a um sistema dedicado à colaboração em ambientes de aprendizagem ubíqua.