Pedestrian Navigation System for Visually Impaired People Using HoloLens and RFID

Abstract

This research proposes a wearable system to help visually impaired people navigate while avoiding obstructions. Visually impaired people require a highly accurate navigation system to obtain reliable positioning. Thus, we offer a positioning method using radio frequency identification (RFID) and the quasi-zenith satellite system (QZSS) to improve positioning accuracy. Our navigation system needs to also recognize the surrounding environment and the direction users are facing, and it must carefully help navigate around obstacles. We use Microsoft HoloLens to detect the direction the user is facing and to locate a walkable path that avoids obstruction. HoloLens has multiple cameras and a depth sensor for positioning and mapping. The system navigates by combining the global area path (via RFID and QZSS) and the local area path (via HoloLens). In this paper, we describe system construction and approach, and we present results of demonstrations.