Preliminary study for motion pose of inshore ships based on point cloud: Estimation of ship berthing angle

This paper presents a method to obtain inshore-ship angle based on LiDAR 3D point cloud data. When the inshore ship is berthing now, the compass equipment is usually used to get the angle information of the ship. This paper puts forward an idea of obtaining ship angle information through point cloud data by studying the characteristics of the ship hull and treating LiDAR signals when the ship is berthing. The point cloud data in the range of the ship hull was determined, and the angle information between ship and wharf was finally estimated through these point cloud data. In this paper, by establishing a 3D scene of the port, the point cloud data of ships berthing from multiple angles are simulated, and the algorithm is verified on these point cloud data. The results show that the maximum absolute error of the algorithm in this paper is 0.49 degrees, the root-mean-square error (RMSE) is 0.27 degrees, the mean absolute error (MAE) is 0.17 degrees, and our method is competitive with other existing inshore ship angle estimation methods. Therefore, the method in this paper can effectively obtain the angle information of the ship when it is berthing and help the ship keep the parallel berthing as much as possible so that the berthing operation of the ship is safer. data is at: https://drive.google.com/drive/folders/1msiQdONOXWWgtX2pS_4fAe6mBPlZ4g7k?usp=sharing code is at: https://github.com/serendipitylxd/Estimation-of-berthing-angle