A ROS-based distributed multi-robot localization and orientation strategy for heterogeneous robots

Abstract The problem of estimating and tracking the location orientation a mobile robot by another in heterogeneous distributed multi-robots is studied this paper. We propose multi-robot localization strategy (DMLS) that Robotic Operating System (ROS) based. It consists an algorithm fuses data diverse sensors from 2 robots are not connected within their transform trees to localize measure relative position orientation. method exploits robust detection Convolutional Neural Networks (CNN) accurate measurements local costmap. composed two parts: part measurement part. Localization done optimization alignment calibration CNN output with costmap individual robot. collaborative fusing sensor align synchronize frames both costmaps. To illustrate performance strategy, proposed compared conventional object measuring uses computer vision QR codes. results show while maintaining degree simplicity efficiency costs. paper also presents various application experiments laboratory simulation environments. By using method, collaborate achieve collective intelligence individuals, which increases team performance.