Exploring the Needs, Preferences, and Concerns of Persons with Visual Impairments Regarding Autonomous Vehicles
Julian Brinkley, Earl W. Huff, Briana Posadas, Julia Woodward, Shaundra B. Daily, Juan E. Gilbert · 2020 · ACM Transactions on Accessible Computing · doi:10.1145/3372280
Summary
This paper presents two complementary studies exploring the needs, preferences, and concerns of people with visual impairments regarding autonomous vehicles—a technology that could fundamentally transform mobility for those who cannot drive conventional vehicles. Study 1 was a 39-question online survey distributed to 516 blind and visually impaired respondents across the United States. Study 2 consisted of eight focus groups with 38 participants (22 blind, 16 low vision, mean age 51.5) conducted over two days in north central Florida. The research addresses a significant gap: while autonomous vehicles are frequently promoted as transformative for blind users, most development has focused on the "driver" paradigm of sighted users rather than the "operator" paradigm that would apply to people with visual impairments. The authors argue that self-driving vehicle technology is largely not being developed with accessibility in mind, despite marketing efforts (like Google/Waymo's prominent use of a blind user in promotional materials) that suggest otherwise. Survey respondents were predominantly female (54%), well-educated (59% bachelor's degree or higher), and over 55% had been blind or visually impaired for their entire lives. Focus group participants had household incomes ranging from $15,000 to $99,000 and were grouped primarily by degree of vision loss and age. Both studies explored general opinions, anticipated benefits, concerns about implementation, preferences for human-machine interfaces, willingness to pay, and concerns about discriminatory legislation.
Key findings
Survey results showed overwhelming optimism: 88% of respondents had positive impressions of self-driving vehicles, with 80% confident the technology would reduce crashes and 79% expecting reduced crash severity. However, significant concerns existed: 93% worried about equipment failure, 93% about vehicle confusion in unexpected situations, and 88% about interactions with pedestrians and bicycles. Notably, 93% expressed interest in owning a self-driving vehicle, with average willingness to pay $6,346 extra—higher than sighted consumers in comparable studies. Focus groups revealed six major themes: self-driving vehicle concerns (223 mentions), potential benefits (151), licensing/training (121), human-machine interface (88), purchase considerations (74), and risk/trust (33). Independence emerged as the dominant benefit—47% specifically used the word "independence" and 97% responded positively to mobility benefits. Participants shared powerful personal stories about how inability to drive affected employment, healthcare access, diet, and social connections. Critical accessibility-specific concerns emerged that differ from sighted users: How do you park a vehicle you cannot see? How do you find your car in a parking lot? How do you verify you've arrived at the correct location? How do you orient yourself upon exiting? These "last 100 feet" challenges were viewed as nearly as complex as the driving itself. For human-machine interface, 71% anticipated speech interaction, but many preferred smartphone control over built-in touchscreens due to familiarity and existing accessibility features. Blind participants were "extremely resistant" to standard touchscreens. Concerning legislation, 94% of survey respondents and 55% of focus group participants expressed concern about laws that could prevent blind people from operating autonomous vehicles—particularly requirements for a licensed driver or ability to take manual control in emergencies.
Relevance
This research provides foundational evidence for the accessible design of autonomous vehicles, documenting needs that have been "largely unexplored and under investigated." For accessibility practitioners and AV developers, the findings highlight that simply removing the driving task is insufficient—comprehensive accessibility requires addressing the entire journey including parking guidance, location verification, situational awareness during transit, vehicle locating, and emergency assistance. The preference for smartphone-based control over built-in vehicle interfaces has significant design implications: blind users already have accessible devices they know well, and duplicating controls in the vehicle may be less valuable than robust smartphone integration. The resistance to touchscreens and concerns about voice recognition accuracy (based on Siri/Alexa experiences) suggest that interface reliability is paramount. The legal and regulatory findings are particularly actionable: 94% concerned about discriminatory laws indicates the disability community is actively watching AV legislation. The focus groups' concept of an "operator's license" (training without vision requirements) emerged as an acceptable compromise that could inform policy discussions. Perhaps most importantly, the finding that higher-educated respondents were less likely to believe their needs were being considered suggests a perception gap between marketing claims and actual development practices—a warning sign for manufacturers claiming accessibility commitment.
Tags: autonomous vehicles · self-driving cars · visual impairment · mobility · transportation accessibility · human-machine interface · user needs
Standards referenced: SAE J3016 · ADA