Navigation training for persons with visual disability through multisensory assistive technology (Preprint)

Abstract

Visual impairment (VI) is a growing problem for many middle-aged and older adults. Conventional mobility aids, such as white canes and guide dogs, bring key limitations that have prompted growing interest in electronic travel aids (ETAs). Despite remarkable progress, current ETAs lack empirical evidence and realistic testing environments, and often focus on the substitution or augmentation of a single sense.

Our effort has two main objectives. First, we establish a novel virtual reality (VR) environment to test the efficacy of ETAs in complex urban environments for a simulated VI. Second, we evaluate the impact of haptic and audio feedback, individually and combined, on navigation performance, movement behavior, and perception of participants. Through this study, we aim to address gaps in order to advance pragmatic development of assistive technologies for persons with VI.

The VR platform was designed to resemble a subway station environment provided with the most common challenges faced by persons with VI during navigation. This environment was used to test our multisensory, assistive technology (AT)-integrated VR on 72 healthy subjects performing an obstacle avoidance task while experiencing symptoms of VI. Each participant performed the task four times: once with haptic feedback, once with audio feedback, once with both feedback types, and once without any feedback. Data analysis involved metrics such as completion time, number of collisions, head and body orientation, and walking trajectory length and smoothness. A two-way repeated measures ANOVA was conducted for each metric to assess the effectiveness and interaction of the feedback modalities. We also conducted a descriptive statistical analysis on participants’ answers to a questionnaire administered after the tests, assessing their experience and preference for feedback modality.

Results from our study showed that haptic feedback significantly reduced collisions (P=.03), and the variability of the pitch angle of the head (P=.02). Audio feedback improved trajectory smoothness (P=.03) and mitigated the increase in the trajectory length from haptic feedback alone (P=.04). Participants reported a high level of engagement during the experiment (52/72, 76%), found the experiment interesting (42/72, 58%), and preferred combined feedback modalities (29/72, 40%).

AT is crucial for individuals with visual impairments, yet they often lack user-centered designs. To bridge this gap, research should prioritize consumer-oriented methodologies, testing devices in a staged manner with progression towards more ecologically valid, realistic settings, ensuring safety. Our multisensory, AT-integrated VR system takes a holistic approach, offering a first step towards enhancing users' spatial awareness. Our approach promotes safer mobility and holds potential for applications in medical treatment, training, and rehabilitation. Continuous technological advancements provide opportunities to refine such devices, significantly improving independence and quality of life for those with VI.