Vision-Based Automated Guided Vehicle

Abstract

Guiding technology plays a crucial role in the domain of automated guided vehicles (AGVs) to ensure accurate and efficient navigation. This paper introduces a vision-based guiding method that utilizes an innovative artificial landmark called the ArUco code, implemented on a Raspberry Pi single-board computer. Traditional guiding systems based on colored or reflective tape provide road information but lack precise positional data. In contrast, the proposed method leverages the ArUco code, serving as a comprehensive guide that furnishes both position and navigation information for the AGV. By capturing images of the environment using an onboard camera, the AGV's vision system processes the images to derive essential road parameters. The ArUco code, strategically placed along the pathway, acts as a reliable artificial landmark. An advanced localization algorithm, utilizing the ArUco code, accurately determines the AGV's position in real-time. This combined guidance and localization approach significantly enhances the AGV's navigational capabilities. The implementation of the proposed method is based on the Raspberry Pi platform, employing the OpenCV library and programmed in Python. To validate the efficacy of the system, an experiment was conducted using a four-wheeled mobile robot. The experimental results demonstrate the successful integration of the vision-based guiding method with the ArUco code, showcasing precise positioning and navigation for the AGV. This research contributes to the field of AGV guidance by offering a solution that combines vision-based guiding with the utilization of ArUco codes. The proposed method provides a cost-effective and adaptable approach for achieving accurate and reliable navigation in diverse industrial environments. Industries seeking to optimize their AGV systems can benefit from this novel guiding technology, improving productivity and operational efficiency.