Mapping and localization of cooperative robots by ROS and SLAM in unknown working area

Abstract

In this paper, we developed a control system hardware based on ROS and mapping and localization for two cooperative robots' self-driving and working in an unknown area. We applied the SLAM (Simultaneous Localization and Mapping) technology to recognize the robots' positions and environment conditions in the unknown area. And we also developed an UI system with C# windows programming to communicate between robots, and connected this UI to ROS. In order to improve the traditional architecture of ROS system and to use no outside sensors for self-driving and controlling of multi-robots, we designed a new hybrid architecture that only one PC served the task of master and the embedded system of Odroid-U3 and sensors were installed inside of two robots for communicating to master. By improving a control system with Linux, Windows, PC, and embedded board, the location of robot could be estimated successfully without expensive sensors and space markers. So, we expect that controlling of multiple robots in unknown working area would be possible easily without complex path planning of former studies.