Location Based Multiplayer Mobile Game in Indoor Environments

Abstract

Recent advances in Augmented Reality (AR) technology and its potential has made lots of headlines in recent years. Availability of exciting and usable applications such as: Pokémon Go and IKEA Place have really helped increasing the awareness about the technology. Since smart phones are capable of supporting AR, both consumers and corporations have become even more interested in applications that provide AR experience. Such applications could further improve user experience and provide better services if they could be localized. GPS has become the de-facto standard for outdoor localization. However, it cannot work in indoor environment because it requires a direct line of sight. Although, there are many techniques for indoor localization, none of them could be considered the best alternative. There is a need to identify the best solution for using with AR-based applications. Also, it is very important to know whether user experience degrades in multi-user applications when compared to single user application.

This thesis outlines the techniques used for indoor localization and communication and aims to present image-based localization as the most suitable technique to be used with AR for mobile applications. In this thesis, we developed a location-based multiplayer mobile game. The game used AR to combine virtual objects and real environment to generate game scene and used image-based localization technique for localizing the player in the real environment. The game was developed using event-driven programming approach and used publish-subscribe message exchange pattern over the WebSocket connection for exchanging messages. The game was able to localize players in indoor environment with acceptable level of accuracy for AR experience