CE-18.09

Title: 

Labyrinth VR Application

Author(s): 

Chandra Pradyumna Adusumilli

Oral Defence Date: 

Friday, July 13, 2018 - 10:20

Location: 

TH 434

Committee: 

Profs. Ilmi Yoon and Bill Hsu

Abstract: 

Virtual Reality shows strong potential to be used as diverse medical treatments. A VR, called Labyrinth, is a virtual environment that challenges and stimulates the hippocampus region of the brain (associated with Long Term Memory). Adaptive training with Labyrinth is intended to stimulate hippocampal plasticity and enhance accuracy and capacity for detailed memory. Labyrinth uses Unity Engine for the procedural generation, quick prototyping and Virtual Reality (VR) capabilities. Labyrinth is played in VR using the HTC Vive on the Virtuix Omni, an omnidirectional treadmill platform, or the Vive trackers for movement. Players will navigate the city/landscape maze, and will use the Vive controllers to interact with the game. Player movements and rotation of the trackers are captured and filtered to identify only the Y-Axis motion of the feet which is in turn reflected as forward translation movement in the maze. Physics colliders are attached to the player as well as the walls and floor of the maze and the collision event between the floor and the player allows tracking of the path taken by the player. Labyrinth, unlike generic applications which predefine all aspects of a scene, uses procedural generation to create the maze. This makes Labyrinth flexible while handling multiple levels and layouts from a single interface although making the development more complicated. Since the maze is procedural the pathing too needs to be generated programmatically. Labyrinth path finding algorithm creates a graph of all the walkable areas of the maze, thus identifying the shortest path along the interactable points using BFS. All the player navigation data is saved in a readable format to allow scientists to run further analysis.

Keywords: 

Virtual Reality, Unity, Vive, Procedural Generation, Game Development, Gamification, Pathfinding, C#, Neuroscience, Player Tracking