Gain Compensation in Redirected Walking

Redirected Walking Techniques (RWTs) enable a user to immersively explore a virtual environment larger than the available physical space by real walking. RWTs are based on the use of gains (translational, rotational and curvature), which introduce a mismatch between the virtual and physical trajectories. When these gains are applied within certain thresholds, the "manipulation" is unnoticeable and immersion is maintained. Numerous research has been carried out to identify these thresholds and factors that affect them such as walking speed, environment structure or tasks involved. However, it has not been known whether users change their walking behavior when RWTs are applied and if this in turn influences ftheir perception thresholds. In this paper, we investigate the change in users' walking behavior, particularly their walking speed, when translational gains are applied. We call this behavior gain compensation. 17 subjects were invited to play a shopping game where they had to walk 50 straight segments to fetch the ingredients. During each segment, one of the five different translational gains (0.7, 0.9, 1.0, 1.2, 1.4) was randomly applied and users' walking speeds were measured. Results show that there is a negative correlation between walking speed and translational gain.