Enhanced Real-Time Camera Pose Estimation Using ArUco Markers in ROS-Based Robotic Systems

Accurate camera pose estimation is crucial for various applications in robotics and computer vision, enabling precise navigation and interaction with the environment. This paper presents a straightforward approach to camera pose estimation and calibration using ArUco markers. Different Aruco’s is used for getting the camera pose and calibration is performed with a single ArUco marker to determine the camera's intrinsic parameters. To enhance the accuracy of pose estimates, bundle adjustment is applied, minimizing re-projection errors through optimization. Additionally, Networkx is used to build a graph, visualizing the spatial relationships between detected markers. The entire system was tested in the Stonefish ROS simulator and Real Environment, validating the calibration and pose estimation processes. The results demonstrate the effectiveness of our methods, offering a reliable framework for applications that require precise camera pose estimation in robotic systems.