Adaptive redirected walking in a virtual world

Redirected walking enables the physical exploration of large virtual environments within the confines of more limited physical spaces. In this paper, we present a novel method for adaptively determining an appropriate mapping from a linear path in a larger virtual environment into a curved path in a smaller tracked space that simultaneously integrates the use of both rotational and translational offsets. We evaluate the method using a software simulator that redirects a user walking on random paths between 16 randomly selected target locations in a virtual environment that is twice as wide as the available tracked space. We find that our controller has a 77% success rate using an average rotational offset of 0.5 degrees per step and an average translational gain of 1.5.