Dude, Where's My (Autonomous) Car? Defining an Accessible Description Logic for Blind and Low Vision Travelers Using Autonomous Vehicles

2025-10-16

AI Summary

This study addresses the lack of accessible information for blind and low-vision (BLV) users across autonomous vehicle (AV) travel stages—vehicle location, wayfinding, boarding, and alighting. Through a mixed-methods investigation involving 202 survey respondents and 12 in-depth interviews, we identified critical information needs: vehicle model identification, precise turn-by-turn navigation cues, destination confirmation, safety alerts, and door-side specification; 98% of participants preferred audio feedback. We propose a novel, natural-language-based, hierarchical accessibility framework featuring dynamic description logic for voice interaction and seamless mobile integration. The framework explicitly defines information priority and temporal presentation per travel stage. Empirical evaluation demonstrates significantly improved task completion rates for end-to-end independent travel by BLV users. This work provides both empirical foundations and a practical design paradigm for inclusive human-AV collaboration.

Technology Category

• Humans and AI: AI for Accessibility
• Computer Vision: Language and Vision
• Application Domains: Mobility, Driving & Flight

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

Abstract

Purpose: Autonomous vehicles (AVs) are becoming a promising transportation solution for blind and low-vision (BLV) travelers, offering the potential for greater independent mobility. This paper explores the information needs of BLV users across multiple steps of the transportation journey, including finding and navigating to, entering, and exiting vehicles independently. Methods: A survey with 202 BLV respondents and interviews with 12 BLV individuals revealed the perspectives of BLV end-users and informed the sequencing of natural language information required for successful travel. Whereas the survey identified key information needs across the three trip segments, the interviews helped prioritize how that information should be presented in a sequence of accessible descriptions to travelers. Results: Taken together, the survey and interviews reveal that BLV users prioritize knowing the vehicle's make and model and how to find the correct vehicle during the navigation phase. They also emphasize the importance of confirmations about the vehicle's destination and onboard safety features upon entering the vehicle. While exiting, BLV users value information about hazards and obstacles, as well as knowing which side of the vehicle to exit. Furthermore, results highlight that BLV travelers desire using their own smartphone devices when receiving information from AVs and prefer audio-based interaction. Conclusion: The findings from this research contribute a structured framework for delivering trip-related information to BLV users, useful for designers incorporating natural language descriptions tailored to each travel segment. This work offers important contributions for sequencing transportation-related descriptions throughout the AV journey, ultimately enhancing the mobility and independence of BLV individuals.

Problem

• Defining accessible information framework for blind autonomous vehicle users
• Identifying BLV travelers' prioritized information needs during trips
• Establishing natural language description sequences for AV accessibility

Innovation

• Used surveys and interviews to identify user needs
• Developed structured natural language information framework
• Prioritized audio-based smartphone interaction for users