Analyzing the Impact of Spatio-Temporal Sensor Resolution on Player Experience in Augmented Reality Games

Along with automating everyday tasks of human life, smartphones have become one of the most popular devices to play video games on due to their interactivity. Smartphones are embedded with various sensors which enhance their ability to adopt new new interaction techniques for video games. These integrated sen- sors, such as motion sensors or location sensors, make the device able to adopt new interaction techniques that enhance usability. However, despite their mobility and embedded sensor capacity, smartphones are limited in processing power and display area compared to desktop computer consoles. When it comes to evaluat- ing Player Experience (PX), players might not have as compelling an experience because the rich graphics environments that a desktop computer can provide are absent on a smartphone. A plausible alternative in this regard can be substituting the virtual game world with a real world game board, perceived through the device camera by rendering the digital artifacts over the camera view. This technology is widely known as Augmented Reality (AR). Smartphone sensors (e.g. GPS, accelerometer, gyro-meter, compass) have enhanced the capability for deploying Augmented Reality technology. AR has been applied to a large number of smartphone games including shooters, casual games, or puzzles. Because AR play environments are viewed through the camera, rendering the digital artifacts consistently and accurately is crucial because the digital characters need to move with respect to sensed orientation, then the accelerometer and gyroscope need to provide su ciently accurate and precise readings to make the game playable. In particular, determining the pose of the camera in space is vital as the appropriate angle to view the rendered digital characters are determined by the pose of the camera. This defines how well the players will be able interact with the digital game characters. Depending in the Quality of Service (QoS) of these sensors, the Player Experience (PX) may vary as the rendering of digital characters are affected by noisy sensors causing a loss of registration. Confronting such problem while developing AR games is di cult in general as it requires creating wide variety of game types, narratives, input modalities as well as user-testing. Moreover, current AR games developers do not have any specific guidelines for developing AR games, and concrete guidelines outlining the tradeoffs between QoS and PX for different genres and interaction techniques are required. My dissertation provides a complete view (a taxonomy) of the spatio-temporal sensor resolution depen- dency of the existing AR games. Four user experiments have been conducted and one experiment is proposed to validate the taxonomy and demonstrate the differential impact of sensor noise on gameplay of different genres of AR games in different aspect of PX. This analysis is performed in the context of a novel instru- mentation technology, which allows the controlled manipulation of QoS on position and orientation sensors. The experimental outcome demonstrated how the QoS of input sensor noise impacts the PX differently while playing AR game of different genre and the key elements creating this differential impact are - the input modality, narrative and game mechanics. Later, concrete guidelines are derived to regulate the sensor QoS as complete set of instructions to develop different genres or AR games