Assistive Guidance System Based on Online Path Structure Recognition for the Visually Impaired

Abstract

Independent mobility is essential for visually impaired individuals, yet existing mobility aids such as white canes and guide dogs have limitations in accessibility and effectiveness. To address this, we propose an assistive robotic guidance system that replicates key functionalities of guide dogs. Our system operates without pre-built maps and dynamically recognizes road structures using LiDAR, allowing the robot to navigate based on the user’s orientation commands. The system consists of a modular architecture with separate guidance and mobility modules, ensuring adaptability across various environments and robotic platforms. Real-world experiments using both wheeled and quadrupedal robots demonstrate high accuracy in path recognition and effective navigation based on user-provided directional inputs. The results validate the feasibility of our approach in providing accessible and reliable mobility support for the visually impaired.