Wearable Interfaces for Orientation and Wayfinding
David A. Ross, Bruce B. Blasch · 2000 · Proceedings of the Fourth International ACM Conference on Assistive Technologies (Assets '00) · doi:10.1145/354324.354380
Summary
This paper from the Atlanta VA Rehabilitation R&D Center presents the development and evaluation of three wearable orientation interfaces designed to help people with severe visual impairment maintain spatial orientation while navigating, with street crossing as the critical test scenario. The authors note that the approximately 11.4 million visually impaired Americans are an increasingly aging population with high rates of co-morbidities including hearing loss, reduced sensation (peripheral neuropathy), cognitive decline, and diminished interest in learning new skills — factors that make interface design particularly challenging. Three interfaces were developed on a wearable computer platform: a stereophonic sonic guide ("Sonic Carrot") that produces a spatially positioned bell-like tone indicating the direction of a target relative to the user's head or body orientation, updated 30 times per second; a speech interface that announces the target's direction using either clock-face descriptions or degrees relative to the user's heading every two seconds; and a shoulder-tapping interface using a 3x3 grid (later reduced to three contact speakers) mounted under the shirt across the top of the shoulders that produces tapping sensations indicating directional information. Orientation input was provided by a digital compass mounted on either the shoulder (body-referenced) or in a hat (head-referenced), with a pedestrian signal detector designed to receive directional data from an accessible pedestrian signal system.
Key findings
Testing with 15 subjects aged 62 to 80 (average 69), who ranged from totally blind for over 40 years to recently partially sighted with acuity of 20/500, revealed that no single interface was best for all users — performance varied widely by individual and mode of operation. However, the shoulder-tapping interface was found to be the most universally usable across the diverse participant group. A very significant improvement in walking pace was achieved when subjects used their best interface/mode, and average veering was reduced to 31% of baseline, though this was not statistically significant given that average baseline veering was around 10 feet. When subjects used their best interface, hesitations dropped to zero, confused episodes dropped to zero, and out-of-crosswalk incidents decreased. Subject preferences were divided: five chose the tapping interface, five chose speech, and four chose the Sonic Carrot. Subjective feedback highlighted important design considerations: the Sonic Carrot was praised for not causing the user to "overcorrect" but some found it hard to localize; speech was valued for being easy and natural to respond to but headphones made it hard to hear traffic sounds; and the tapping interface was liked for leaving hands and ears free but needed to be more adjustable. Four subjects and two who volunteered the suggestion independently recommended a combined tapping/speech interface as ideal.
Relevance
This research addresses a safety-critical accessibility need — independent street crossing for blind and visually impaired pedestrians — through rigorous user-centered evaluation with actual members of the target population. The key insight that the aging blind population presents dramatically different needs from younger blind adults, due to co-morbidities affecting hearing, sensation, cognition, and motivation, remains highly relevant as the population continues to age. The finding that no single modality works for everyone, leading to the recommendation of a combined tapping/speech interface, exemplifies the principle that accessible navigation systems must be multimodal and personalizable. For modern practitioners, this work foreshadows current research on haptic navigation aids in smartphones and wearables, and the emphasis on not blocking environmental audio awareness (traffic sounds, echoes) remains a critical design constraint for any wearable navigation system. The participatory testing approach with older adults who have varying combinations of disabilities provides a model for inclusive evaluation methodology.
Tags: wayfinding · orientation and mobility · wearable technology · blindness and low vision · haptic feedback · spatial audio · navigation · pedestrian infrastructure · assistive technology · universal design