Geometric Mapping and Pose Estimation Techniques for Improving Safety and Efficiency in Offshore Crane Operations

• Rafael T. Guthes FURG
• Matheus M. Dos Santos FURG
• Rafael Lemos FURG
• Guilherme Azambuja FURG
• Juliana Dos Santos FURG
• Eduardo N. Borges FURG
• Paulo L. J. Drews Junior FURG
• Silvia S. C. Botelho FURG

Resumo

The safety and effectiveness in handling heavy loads during offshore crane operations are of paramount importance to prevent accidents and minimize financial losses. One of the primary challenges in offshore operations is ensuring the secure transportation of loads, as operators face difficulties in visualizing and anticipating the movement of the transported cargo. This paper presents the development of a 3D scenario equivalent to the laydown area in an offshore environment. The geometric mapping of loads within the operational zone was accomplished using two cameras, along with the application of computational techniques to estimate the pose in dynamic environments. By employing multiple-view geometry techniques, the proposed study aims to assist in load tracking, trajectory prediction, and collision anticipation during crane operations. With a strategic positioning of the cameras, comprehensive spatial mapping and precise estimation of object poses within the loading and unloading area are made possible. Furthermore, the study proposes a 3D representation of the offshore environment, providing a concise visual demonstration of the system’s capabilities and facilitating the evaluation of the feasibility and effectiveness of the proposed approach. We emphasize the key computational methods for geometric mapping of the pose of the transported load, aiming to enhance the safety and efficiency of offshore crane operations. This work serves as a foundation for future advancements in geometric mapping techniques and position estimation in similar challenging environments.

Palavras-chave: Cameras, Crane, Keypoint, Offshore, Pose