Towards a Sign-Based Indoor Navigation System for People with Visual Impairments

Area of Science:

• Assistive Technology
• Human-Computer Interaction
• Computer Vision

Background:

• Outdoor GPS navigation tools are insufficient for indoor environments.
• Existing indoor navigation solutions often require expensive infrastructure or detailed visual environment models.
• Visually impaired individuals face significant challenges with indoor wayfinding.

Purpose of the Study:

• To develop and assess a novel smartphone-based indoor navigation aid for the visually impaired.
• To create a system that does not rely on additional physical infrastructure.
• To evaluate the feasibility of a navigation system using inertial sensing, computer vision, and floor plan data.

Main Methods:

• Development of a smartphone application integrating inertial measurement units (IMUs) and a camera.
• Utilizing computer vision algorithms to interpret the environment and identify landmarks (e.g., signs).
• Combining sensor data with pre-existing digital floor plan information for indoor localization.

Main Results:

• The developed system demonstrated the feasibility of smartphone-based indoor navigation for visually impaired users.
• The approach successfully estimated user location using inertial sensing, computer vision, and floor plan data.
• A formative study with blind participants provided valuable insights for system improvement.

Conclusions:

• The proposed smartphone-based navigation aid offers a promising, infrastructure-free solution for indoor wayfinding for the visually impaired.
• Further development is needed to address areas for improvement identified in the formative study.
• This technology has the potential to significantly enhance independence and mobility for visually impaired travelers.