IEEE Trans Vis Comput Graph. 2018 Oct;24(10):2663-2674. doi: 10.1109/TVCG.2017.2771520. Epub 2017 Nov 10.
Understanding real walking in virtual environments (VEs) is important for immersive experiences, allowing users to move through VEs in the most natural way. Previous studies have shown that basic implementations of real walking in virtual spaces, in which head-tracked movements are mapped isometrically to a VE, are not estimated as entirely natural. Instead, users estimate a virtual walking velocity as more natural when it is slightly increased compared to the user's physical locomotion. However, these findings have been reported in most cases only for young persons, e.g., students, whereas older adults are clearly underrepresented in such studies. Recently, virtual reality (VR) has received significant public and media attention. Therefore, it appears reasonable to assume that people at different ages will have access to VR, and might use this technology more and more in application scenarios such as rehabilitation or training. To better understand how people at different ages walk and perceive locomotion in VR, we have performed a study to investigate the effects of (non-)isometric mappings between physical movements and virtual motions in the VE on the walking biomechanics across generations, i.e., younger and older adults. Three primary domains (pace, base of support and phase) of spatio-temporal parameters were identified to evaluate gait performance. The results show that the older adults walked very similar in the real and VE in the pace and phasic domains, which differs from results found in younger adults. In contrast, the results indicate differences in terms of base of support domain parameters for both groups while walking within a VE and the real world. For non-isometric mappings, we found in both younger and older adults an increased divergence of gait parameters in all domains correlating with the up- or down-scaled velocity of visual self-motion feedback. The results provide important insights into the design of future VR applications for older adults in domains ranging from medicine and psychology to rehabilitation.
理解虚拟现实环境（VE）中的真实行走对于沉浸式体验非常重要，它使使用者能够以最自然的方式在 VE 中移动。先前的研究表明，在虚拟空间中基本实现真实行走，即将头部跟踪的运动等距映射到 VE 中，并不被完全视为自然。相反，当虚拟行走速度比使用者的身体运动稍微增加时，使用者会估计这种虚拟行走速度更自然。然而，这些发现大多仅在年轻人（例如学生）的情况下报告，而在这些研究中，老年人的代表性明显不足。最近，虚拟现实（VR）引起了公众和媒体的广泛关注。因此，可以合理地假设不同年龄段的人都可以使用 VR，并且在康复或培训等应用场景中，他们可能会越来越多地使用这项技术。为了更好地了解不同年龄段的人在 VR 中行走和感知运动的方式，我们进行了一项研究，以调查物理运动与 VE 中虚拟运动之间的（非）等距映射对代际行走生物力学的影响，即年轻人和老年人。确定了三个主要的时空参数领域（步幅、支撑基础和相位）来评估步态表现。结果表明，老年人在真实环境和 VE 中的步幅和相位领域中行走方式非常相似，这与年轻人的结果不同。相比之下，在 VE 和现实世界中行走时，两组的支撑基础领域参数都存在差异。对于非等距映射，我们发现，在年轻人和老年人中，所有领域的步态参数都随着视觉自我运动反馈的缩放速度增加而增加，与实际的行走速度增加而增加而增加。这些结果为未来为老年人设计从医学和心理学到康复的各个领域的 VR 应用提供了重要的见解。
