Indoor navigation with smartphone-based visual SLAM and Bluetooth-connected wheel-robot

Abstract

A smartphone-based positioning system suitable for indoor robot application is developed in this research by integrating image and other sensors on a smartphone with wheel odometer feedback from a Bluetooth-connected robot vehicle. WiFi signal, inertial sensors, and CMOS sensor in the smartphone were considered to provide measurements for positioning calculations. The real-time performances of SIFT, SURF, and ORB image feature detection and tracking algorithms are compared in the smartphone and ORB was chosen for implementation. Experimental result of Visual SLAM positioning is presented together with state convergence analysis.