OmniAcc: Personalized Accessibility Assistant Using Generative AI

2025-09-08

AI Summary

To address navigation barriers faced by mobility-impaired individuals in urban environments due to insufficient accessibility information, this study proposes a generative-AI–based personalized accessible navigation framework. Our method integrates GPT-4, high-resolution satellite imagery, and OpenStreetMap data, leveraging zero-shot learning and customized prompt engineering to automatically detect and structurally verify critical accessibility infrastructure—including ramps and crosswalks—enabling real-time path planning and dynamic accessibility assessment. On the crosswalk detection task, our approach achieves 97.5% accuracy, significantly enhancing independent mobility for wheelchair users and providing scalable, data-driven support for urban accessibility evaluation and planning. The core innovation lies in fine-grained accessibility semantic understanding and cross-modal reasoning without requiring annotated training data.

Technology Category

• Humans and AI: AI for Accessibility
• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Diffusion Models for Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Individuals with ambulatory disabilities often encounter significant barriers when navigating urban environments due to the lack of accessible information and tools. This paper presents OmniAcc, an AI-powered interactive navigation system that utilizes GPT-4, satellite imagery, and OpenStreetMap data to identify, classify, and map wheelchair-accessible features such as ramps and crosswalks in the built environment. OmniAcc offers personalized route planning, real-time hands-free navigation, and instant query responses regarding physical accessibility. By using zero-shot learning and customized prompts, the system ensures precise detection of accessibility features, while supporting validation through structured workflows. This paper introduces OmniAcc and explores its potential to assist urban planners and mobility-aid users, demonstrated through a case study on crosswalk detection. With a crosswalk detection accuracy of 97.5%, OmniAcc highlights the transformative potential of AI in improving navigation and fostering more inclusive urban spaces.

Problem

• Identifies wheelchair-accessible urban features using AI
• Provides personalized navigation for mobility-impaired individuals
• Enhances accessibility information with real-time query responses

Innovation

• Uses GPT-4 and satellite imagery for accessibility mapping
• Provides personalized wheelchair-accessible route planning
• Employs zero-shot learning for precise feature detection