Pose Estimation for Bypass Line Clamp in Complex Operating Scenario

The robot necessitates autonomous identification of the pose (orientation and position) of the working tools during the establishment of the bypass system. Therefore, this paper proposes a novel marker system based on CylinderTag to estimate the pose of bypass line clamps. During marker generation, a method for generating 3D information tailored to cylindrical targets is proposed, improving the convenience and speed of marker deployment. For marker detection, an OPTICS (ordering points to identify the clustering structure) clustering-based marker detector is proposed. Experimental results demonstrate that this novel detector improve the success rate of pose estimation of bypass line clamp using CylinderTag in complex scenarios and exhibits stronger robustness to lighting variations.