Laser Rangefinder-Based Multi-Robot Cooperative Localization Using Cloud Computing

Abstract

Multi-Robot Cooperative Localization (MRCL) is a promising technology for applications, e.g., shipping, production, logistics, etc. However, MRCL for outdoor applications faces challenges in localization accuracy due to the low-precision baseline measurement between every two robots by Global Positioning System (GPS). In this paper, we focus on target localization using three mobile robots with ordinary cameras in the same horizontal space. Taking the cloud server as the basic computing platform, we propose a localization system using 3-robots with the laser rangefinder for the baseline measurement between robots. We also propose a trilateral positioning algorithm, which combines the advantages of both least squares matrix method and gradient descent method. The simulation results demonstrate that the proposed system can improve the accuracy by 80.4% compared with the traditional method.