Flee to the Tree

flee cover.PNG

Flee to the Tree is a VR experience where we explored the navigation of vertical space in room scale VR.  A player started as a small creature in a forest being chased by a woodpecker.  Their only way to escape is to crawl through a hole in the tree next to them.  Once inside, the player can throw an acorn at the bird, which is now poking its beak through a small hole in the trunk.  After the bird flees, the player must climb up a ladder onto a series of platforms.  Using foot trackers, we had players cross these round platforms.  The foot trackers allowed them to actually see their feet, making the experience more realistic.  We experimented with using haptic feedback in the real world to make a player feel as though they were actually stepping onto these platforms but found that the tracking system did not calibrate to the real world accurately enough for our system to work. Once a player crossed the platforms, they climbed another ladder to the top of the tree where they could throw acorns at the bird flying around the tree.  Once the bird was hit, it flew away, and the player has finished the experience.

I served as the main developer for this project.  The game was developed for the HTC Vive and made with Unity and Steam VR.  I was in charge of building the scene in unity, making basic animations, scripting, and setting up the hardware (the foot trackers).  We built a base level prototype in five weeks to do user testing, which lead to several adjustments in both modelling and tracking systems.  After another five weeks of experimenting with ladder climbing and the platform system, we presented the project at a demo day to several Georgia Tech professors and industry employees.

Video demo of the project


User Testing

User testing for this project began by testing basic interactions and motivations.  Our experience was supposed to simply begin, and a user needed to know what to do without prompting.  The experience needed to be streamlined from beginning until end.  For us, this meant that the woodpecker needed to be scary enough to convince a user to flee and that the hole needed to be big enough to seem like an option for escape.  When a user made it to the top, they needed to already understand that the acorns at the top could be used to throw at the woodpecker to chase him away.

Our first round of user testing looked at these key interactions (fleeing and throwing).  We ran user testing through pre-interviews, talk alouds, and post-interviews.  The pre-interviews were used to determine what a user’s experience with VR was, since that influenced the experience for them.  For this testing, we found that the bird needed to be scarier, the whole bigger, and we needed to include a ‘tutorial’ for throwing acorns.  This was accomplished by having the bird poke its beak through a small hole in the tree and by placing a bird next to the player in the base of the tree.  Users quickly discovered they could pick up the acorn, throw it at the woodpecker, and then the woodpecker would flee.

Our next round of user testing focused on the platforms we used in the real world.  After carefully mapping out the room to match the virtual area, we began running talk aloud user tests, followed by surveys, to see how people reacted to the addition of something beneath their foot.  We found that the HTC Vive tracking system V1 was simply not accurate enough to maintain the same calibration round after round and thus we could not predict where the platforms would be in the actual room.  However, when the virtual space did align to the actual room and a user was unsuspecting of the platforms, the haptic feedback added to the overall sensation of being on a floating platform.

Future Work

For this project, potential future work would revolve around the floating platforms.  Using the newer tracking systems might allow for more accurate collaboration and thus, more research into the use of haptic feedback during walking.  This would allow us to raise or lower some of the platforms and see if various degrees of haptic feedback correlated with the height of the platform influenced a user’s sensation of moving higher or lower while actually staying on flat ground.  Other work would simply include expanding the experience and potential conflict with the woodpecker.

Sammi HudockGame Design