Gait Guidance using Mixed Reality for Gait Improvement

Abstract

In Japan, the aging demographic is in an increasing need of support with their gait improvement and independence to prevent falling and injuries resulting in hospitalization. However, existing mixed reality methods are limited because that they cannot be implemented with a headset. Moreover, they use large-sized avatar computer graphics (CG) that easily obstructs the field of view. This study developed a gait improvement system based on footprint CG for only-headset application that does not readily block the view. The user was instructed to walk while stepping on the 3DCG footprints superimposed on the surrounding images by the system. Using motion capture, we verified whether the system influenced the user's gait (stride length, toe angle, and knee height) under 11 display conditions with varying stride lengths, angles, and heights of the footprints. The analysis results revealed that the user's stride length by up to 8.80%, toe angle by up to 2.44°, and knee height by up to 3.76 cm varied with the footprint stride length, angle, and height, respectively. Overall, the results suggested that the proposed can control the user's gait. Further refinement of the conditions and a large amount of subject data are needed to more accurately clarify the footprint presentation conditions in which the user's gait will improve subconsciously while using the system.