'Watch Your Step': Precise Obstacle Detection and Navigation for Mobile Users Through Their Mobile Service

Abstract

Obstacle detection or navigation is useful for mobile users, especially for the visually impaired. In this paper, we propose a method for detecting obstacles precisely from mobile devices by using a depth camera. We demonstrate the system based on Google Project Tango. The system was developed by employing the Java SDK of a Google Tango mobile device. It mainly uses a built-in infrared (IR) sensor to collect data to build a perception of the surrounding environment and then constructs a matrix to record the distance information. Subsequently, the system collects data from the surrounding environment in real time and analyzes the data. If any potential risk of being exposed to injury exists, then the system delivers a warning message and indicates a safe path for the user. An experiment was designed to evaluate our system. On the basis of the experimental results, we discuss the implications for the design of the system.