The Role of Ship Simulators and Maneuvering Models in Maritime Operations

Abstract

The concept of autonomy in the maritime domain is attractive from both the safety and efficiency points of view, as in other transportation domains. To investigate the payoffs of autonomy as automation levels increase on a ship’s bridge, a literature review is presented. The concept hence is broken-down into two sides: the technology and human operator side.

The technology side is limited to hydrodynamics, Guidance,Navigation, and Control (GNC).

The operator side is limited to experiences by ship and airplane operators facing increased automation in their workplaces.

The literature review is distilled into the main functional challenges to be confronted in the pursuit of maritime autonomy. First, the uncertainties stemming from ship dynamics and environmental loads largely affect the performance of GNC technologies.

Second, the alteration of human-machine interaction as automation increases introduces new sources of error. Human operators face serious challenges dealing with highly automated systems. The following were concluded:

The accuracy of ship maneuvering models in calm waters and operational conditions is crucial for the advancements of maritime autonomy, The focus on remote control and the capabilities of remote operators rather than full autonomy and the capabilities of machines, The use of ship simulators for enabling research in both maneuvering models and remote control of ships.

The application of ship simulators and accuracy of ship dynamics therein merited further investigation. A comprehensive study to learn about ship simulators and their application in research and industry was conducted. This study includes a literature review, interviews, and a case study with the Norwegian Coastal Administration to identify the role of maritime simulators. Ship simulators are no longer merely used for the purpose of nautical education and training. The trends show an expanding scope of ship simulator applications. For example, they are used in the development and testing of autonomous ship controllers, underwater operations planning, and pilot recruitment.

Simulators must demonstrate an appropriate level of physical, behavioral, functional, and visual realism to achieve satisfactory suitability according to applicationobjectives. Physical realism pertains to hardware and furniture settings that are comparable to the bridge of a real ship. behavioral realism is limited to the behavior of bridge equipment. Functional realism is concerned with the physics of a moving ship in water, whereas visual realism focuses on the resolution, size, and shape of the displays and content within. The standard for Maritime Simulator System (DNVGLST-0033) provides appropriate levels of physics and behavior realism, but does not recognize simulator applications other than for education and training. Moreover,the standard does not require objective assessments of ship dynamics, as in the flight simulation standard (CS-FSTD).

Thus, a minimum accuracy requirement is proposed to identify the level of functional fidelity in a simulator. In accuracy requirement level 1 (ARL 1) the ship ‘feels realistic’ while it maneuvers. In ARL 2, the ship maneuvers accurately in calm water.

In ARL 3, the ship maneuvers accurately in operational conditions. Every application, based on its objectives, holds a minimum ARL. For example, a simulator training of nautical students requires ARL 1, however, a simulator training of experienced professionals requires a higher accuracy level.

Maneuvering models are essential for the functional realism of a simulator. They are also used in modern ship navigation systems as an additional input that can predict current and future ship states. The ever-growing scope of simulator applications raises questions about the accuracy and accessibility of maneuvering models.

Maneuvering in calm water has been long investigated; this thesis uses the existing maneuverability standards of the International Maritime Organization (IMO) in calm water and extends them to investigate maneuvering in waves. Specifically, the author is addressing the accuracy of different maneuvering models and their suitability for various applications. The 355 meter-long Duisburg Test Case (DTC) benchmarking containership is considered herein for this study.

This study evaluates the accuracy of two different maneuvering models using an objective assessment. The first maneuvering model belongs to an industry-standard simulation provider. The second is a novel research simulation tool recently developed by the Marine Technology Department of the Norwegian University of Science and Technology (NTNU). Both models are compared against experiments performed earlier (in 2018 and 2019) in the Sintef Ocean Basin in Trondheim, Norway.

The results show the differences between simulators and experiments in both calm water and several wave conditions. Results show that both simulators perform well in calm water, however, they perform differently in waves. The novel model compares more adequately to experiments in waves, appreciating the impact of waves on ship speed while maneuvering and turning.

In summary, this study provides an overview on the functional challenges in autonomous vessels from both the technological and the operator’s perspectives. It also provides a comprehensive overview on ship simulator applications and their role in maritime autonomy. It introduces the accuracy concern and thus proposes a minimum accuracy level standard, the ARL. The ARL standard is an objective measure that identifies the functional realism of a simulator. Based on the application, a simulator with a known ARL is either deemed compliant or non-compliant. Finally, this study objectively evaluates the accuracy of two different maneuvering codes and discusses their suitability for various applications.