Room Mapping and Robot Localization Using RGB-D Camera and IMU Sensor

Room mapping and localization of mobile robots is an important and fundamental problem in robotics research. Environmental perception and identification of the robot's position and orientation to a global coordinate are basic prerequisites for autonomous navigation on mobile robots. In situations where autonomous mobile robots must operate in an unknown environment, localization and mapping are required simultaneously. Thus, simultaneous localization and mapping (SLAM) on mobile robots in unknown environments has become a topic that is widely discussed due to its important theoretical value and possible application. The SLAM is considered the key to realizing the autonomous mobile robot. This also applies to domestic service robots (DSR) that operate in a dynamic environment. In order to accomplish its duties in this dynamic environment, DSR must be able to recognize the environment and the various changes that occur around it. In this paper, we develop the domestic service robots (DSR) and implement SLAM that is based on visual information captured from the RGB-D (i.e., color and depth) camera combined with an inertial measurement unit (IMU). To test the developed systems, we have conducted several experiments. As the preliminary study, we achieved good results.