Kauri Classroom

Project Overview

In this project, I collaborated with the biological sciences faculty at Victoria University to develop a virtual reality experience showcasing the impact of kauri dieback disease. As a solo developer, I was responsible for all the coding and in engine creation, while also working closely with microbiologist Dr. Monica Gerth and biochemist Wayne Patrick to bring their vision to life. The project was aimed at schools in New Zealand, where a portable shipping container classroom and lab was deployed. The container provided a hands-on learning environment for kids to conduct soil experiments, observe local kauri trees, and use Oculus Quest headsets to explore the disease's effects on microbes and root structures in virtual reality.

The user flow was designed with an interaction mechanic that was easy to teach and versatile enough to handle all the interactions throughout the experience. The first interaction is a simple "point and click" mechanic, demonstrated in a stress-free and easily understood manner. In case of inactivity, timeouts are in place to ensure users progress through the experience, which is crucial for high throughput and provides a defined maximum time for the experience. It was important to have a accurate representation of the environment when viewing the trees from above. To achieve this, the sky and landscape needed to reflect the Waipoua forest accurately. To get the correct star positions, I used the "Ultra Dynamic Sky" Unreal plugin, which allowed me to input the latitude, longitude, date, and time for a real-life star layout. However, this was not compatible with the Oculus Quest, so I captured 360-degree images of the sky, one for night and one for sunrise, and blended between them to create the illusion of a night to day transition.

For the landscape, I utilized image and height data from Google Earth, which was then mapped to a mesh. To optimize performance, I implemented a distance-based level of detail (LOD) system that reduces the number of vertices as the user moves further away from the play area.

One of the most popular features was the ability to see through the ground and observe the roots. This was achieved by using a global material parameter collection that accepted a world position vector, creating a sphere mask around it to make anything inside it invisible. To enhance the effect, I added a glowing outline along the edge of the mask.

To visualize the mycelial mat, I employed the use of a Niagara particle system. I created leader particles that emitted in a noisy, displaced pattern, which in turn spawned trail particles to form ribbons. At random intervals, the leader particles also triggered an event that spawned another type of particle, which was a 3D model of an oospore. Additionally, I integrated a commercially available microbe pack to add vibrant color to the visualization.