Robot localization method using multi-PTS cameras and Wi-Fi receiver

Abstract

This paper presents a location method for the mobile robots moving in the indoor environment by using the multi- projected texture stereo (PTS) cameras and a Wi-Fi Receiver. The proposed method is divided into three processes. First, a modulation of the multi-PTS cameras is registering three PTS cameras to measure the wide 3D depth and color images corresponding to the objects surrounding the robot. Second, the measurements of the Wi-Fi receiver and the multi-PTS cameras can be used for building a radio map and a gird-based map. Last, the localization method consisting of a fingerprint and a particle filter algorithms can find the broad and the exact positions of the robot sequentially by matching the real-time measurements of sensors and the maps. To verify the feasibility of the proposed method, we performed the experiment with a real robot in an indoor office environment. And we confirmed that the proposed localization method can be applied to the actual robot practically. Finally, we remark the conclusion and plan the future works to enhance the proposed localization method.