Autonomous Mobile Robot Design and Testing for Data Center Monitoring Mission

Abstract

This paper presents a cost-effective autonomous mobile robot (AMR) designed for remote data center monitoring, motivated by the need to enhance operational efficiency and reduce human intervention in critical infrastructure. The proposed model integrates a TurtleBot Kobuki base with a novel SRF10-based virtual bumper system, RPLIDAR scanner, and GoPro camera, enabling autonomous navigation, obstacle avoidance, and high-resolution photo capture in data center corridors. Developed through a university-industrial collaboration, the prototype leverages ROS-compatible subroutines to minimize development costs. Laboratory and real-world tests validate the AMR’s performance, achieving 92% navigation success and 94.2% gap navigation accuracy, while identifying limitations like positioning errors and low-clearance obstacle detection. Proposed upgrades, including a faster LIDAR and enhanced computing power, aim to meet industrial standards. This study establishes a foundation for scalable, automated monitoring solutions across data centers and similar sectors.