Pose Estimation of Coil Workpieces by Automated Overhead Cranes Using an Improved Point Pair Features Algorithm.

Area of Science:

• Robotics and Automation
• Computer Vision
• Logistics and Supply Chain Management

Background:

• Automating port operations like warehousing and transshipment is crucial for efficiency.
• Traditional methods for coil workpiece pose estimation have limitations.

Purpose of the Study:

• To develop an automated system for crane operations using 3D point cloud algorithms.
• To improve the pose estimation accuracy and robustness for coil workpieces.

Main Methods:

• A novel point cloud registration algorithm combining Point Pair Feature (PPF) and Generalized Iterative Closest Point (GICP) algorithms.
• Utilizing 3D point cloud data for precise pose estimation of coil workpieces.

Main Results:

• The proposed algorithm demonstrated superior performance compared to traditional methods.
• Achieved average pose estimation errors of 1.1%–1.2% for one to three coils.
• Processing times ranged from 3.4 to 4.7 seconds, meeting practical requirements.

Conclusions:

• The novel algorithm enhances the robustness and applicability of point cloud registration.
• The developed method effectively supports automated crane operations for coil handling in ports.