Comparative Analysis of Fractal and ArUco marker for Navigation and Landing of Drones

Abstract

The detection and pose estimation of markers are increasingly crucial across various domains, including human-robot interaction, UAV navigation, robotic grasping, and underwater navigation and many more. This paper evaluates the robustness of two prominent fiducial markers ArUco and Fractal marker, under diverse environmental conditions. These conditions include static and dynamic platforms, varying light intensity, occlusion, reflection, distortion, and detection through glass medium and many other situations considering the area of both markers identical. Detection rates for all conditions were calculated for each marker by analyzing the proportion of detected frames out of approximately 1800 total frames per condition. The study reveals that Fractal markers outperform ArUco markers. They demonstrate faster detection, greater resilience to occlusion, better resolution handling, and consistent performance across multiple scenarios. These findings suggest that Fractal marker could significantly enhance the reliability and efficiency of marker based detection systems in complex real-world applications. The dataset will be made accessible to the general public, facilitating additional investigation and verification of the conclusions reported in this work.