MoCArU: Low-Cost Wireless Portable Robot Localization System Using IoT.

Localization is crucial for various automation systems to provide awareness of the robot's position and orientation. Additionally, a localization system that offers portability, flexibility, and low computational and economic costs is required by a variety of robotics applications. However, no existing system can offer all the aforementioned features suitable for motion capture tasks involving ground swarm robots. In this study, we propose MoCArU, a novel Motion Capture system based on odometry and ArUco, robustly recognized through image data with low computational cost. We have evaluated the system's performance by comparing it with the ground truth trajectory and adopting different numbers of cameras. The results show that MoCArU can achieve a root mean square error of 0.1345±0.0065 m using ten cameras. Our findings add to previous knowledge by presenting a robust and cost-effective alternative to existing localization methods. Here, we show that MoCArU's use of lightweight camera stands and wireless communication ensures ease of installation, portability, and low computational cost, making it suitable for tracking swarm ground robot systems. We anticipate this system to be used in various applications, such as robot position control, navigation, and obstacle avoidance control. Overall, MoCArU provides a reliable and cost-effective solution for the real-time localization of robots, so its wider applicability in various environments is a significant advantage in robotics. An open-source implementation of MoCArU, as well as its related details, is open for public use at https://www.rm.is.tohoku.ac.jp/MoCArU.