Dual Active Marker Modules for Enhanced 6D Pose Estimation

초록

This study proposes the development and application of dual active marker tracking modules for enhanced 6D pose estimation, aiming to overcome the limitations of traditional marker-based pose estimation methods. Traditional approaches, which rely on the installation of multiple 2D markers or volumetric markers, often encounter issues like marker loss, damage, and recognition difficulties from certain angles. To address these challenges, our system incorporates two synchronized marker tracking modules, each equipped with an embedded computer, a 2-axis actuator, an augmented reality (AR) marker, and a camera. These modules actively track each other's markers, ensuring reliable performance even in large workspaces. By aggregating data from both modules and employing a Kalman filter (KF), our system achieves precise estimation of the relative positions and orientations between the modules. Through the integration of dual active marker tracking modules, our approach not only enhances pose estimation accuracy but also expands the possibilities for innovative applications in robotics, virtual reality, and AR applications.