Adequação de um framework para suportar realidade virtual em dispositivos móveis

Abstract

In face of the technological innovations provided by the new paradigm of Mobile Computing, there is, increasingly, an environment to develop software in this area. Major difficulties are detected in the development of friendly interfaces for devices with small screens (Abreu 2005). Pressman (Pressman 2006) considers the user interface design as a theme that has become very important with the increasing use of computers. Besides, he states that most of the Human Computer Interfaces (IHC) is performed by visual means, and there is a definitive tendency towards pictorial communication (graphic) design in HCI. The author also recommends that the visual interface layout must be based on a real world metaphor (Pressman 1995). This theme extends as well to the use of mobile devices, as its software development grows. One way to achieve this metaphor of the real world is by using Virtual Reality (VR) techniques. This technology is an advanced graphical interface between a user and a computer system (Kirner 2004). VR can also be extended to mobile computing, providing greater usability to the software developed for them. This work has researched different software development platforms for mobile devices, investigating their functionalities and its suitability for the development of Virtual Reality applications. During this investigation, the need for a better standardization of architectures that support VR in mobile device has been recognized, mainly when portability matters are considered. Thus, a standard 3D scene (JRS 184 - M3G) of a development platform (J2ME) has been selected to evaluate the limitations of the referred portability. To achieve such a goal, this work proposes an architecture that, through \"lowest common denominator\" for these scenes, allows the development of Virtual Reality applications for mobile devices without the need of lowlevel programming at selected platform. To evaluate the suitability of this proposal, these needs have been embedded into a library and, subsequently, attached to a framework, called LWUIT, expanding it in a new framework (RV_LWUIT), keeping the same portability pattern, originally proposed. As a result, it is possible, in both cases, a significant reduction of code and a consequent reduction in time spent to put on the market (time to market) with VR applications for these devices.