Comparison of two approaches for web-based 3D visualization of smart building sensor data

Abstract

This thesis presents a comparative study on two different approaches for visualizing sensor data collected from smart buildings on the web using 3D virtual environments. The sensor data is provided by sensors that are deployed in real buildings to measure several environmental parameters including temperature, humidity, air quality and air pressure. The first approach uses the three.js WebGL framework to create the 3D model of a smart apartment where sensor data is illustrated with point and wall visualizations. Point visualizations show sensor values at the real locations of the sensors using text, icons or a mixture of the two. Wall visualizations display sensor values inside panels placed on the interior walls of the apartment. The second approach uses the Unity game engine to create the 3D model of a 4-floored hospice where sensor data is illustrated with aforementioned point visualizations and floor visualizations, where the sensor values are shown on the floor around the location of the sensors in form of color or other effects. The two approaches are compared with respect to their technical performance in terms of rendering speed, model size and request size, and with respect to the relative advantages and disadvantages of the two development environments as experienced in this thesis.