Depth camera SLAM on a low-cost WiFi mapping robot

Abstract

Radio-Frequency fingerprinting is an interesting solution for indoor localization. It exploits existing telecommunication infrastructure, such as WiFi routers, along with a database of signal strengths at different locations, but requires manually collecting signal measurements along with precise position information. To automatically build signal maps, we use an autonomous, self-localizing, low-cost mobile robotic platform. Our robot relies on the Kinect depth camera that is limited by a narrow field of view and short range. Our two-stage localization architecture first performs real-time obstacle-avoidance-based navigation and visual-based odometry correction for bearing angles. It then uses RGB-D images for Simultaneous Localization and Mapping. We compare the applicability of 6-degrees-of-freedom RGB-D SLAM, and of particle filtering 2D SLAM algorithms and present novel ideas for loop closures. Finally, we demonstrate the use of the robot for WiFi localization in an office space.