A floating structure for transport is presented, formed by a train arrangement of rotary bodies of revolution that reduces the drag of same during sailing, the train arrangement of rotary bodies being formed by a front body, intermediate bodies and a rear body that have rotation synchronized with the speed of travel of the structure, the intermediate bodies of revolution being connected together by longitudinal rotation shafts by connections secured to an upper platform, while the longitudinal rotation shafts of the front body and the rear body are connected to the rotation shafts of adjacent bodies by hinges, which are pivotably connected to an end of draft-adjustor, pivotably connected at their other ends to the upper platform, the longitudinal rotation shafts being disposed perpendicular to the structure's travel direction and associated with actuators. The rotary bodies are separated by a distance of approximately 5% or less of their diameter.

The invention relates to a method for constructing and/or manufacturing a water sports device (2) which has a modular structure comprising a floating body (4). The modules can be connected together via interfaces and are connected during operation. In particular, the invention relates to a method for constructing and/or manufacturing a foilboard or driver propulsion vehicle, wherein a server device is provided, and a program-controlled input interface is provided for user-defined inputs on a terminal (5), in particular a mobile terminal, which is arranged at a distance from the server device in particular. The modules are imaged in a computer program of the server device and/or of the terminal (5), and at least one outer contour of the floating body (4) of the water sports device (2) can be, in particular, freely defined by the user. On the basis of the outer contour of the floating body (4) defined in the program, automated manufacturing information is produced, and the floating body (4) manufactured according to the production information can be combined with another module, in particular multiple other modules, to produce the water sports device (2). The invention also

A system assists the driving of a ship and is configured to estimate the structural loads of the ship due to the direct wave excitation, and structural loads of the ship due to the whipping effect caused by the wave slamming. The system includes at least one reference sensor adapted to provide an indication of a motion or stress magnitude at a predetermined point of the ship structure, and is further configured to calculate an estimate of the magnitude at the predetermined point in the ship structure, compare the indication of magnitude with the estimate of the magnitude so as to determine an offset value, and correct the estimates of the structural loads and/or the estimate of the magnitude on the basis of the offset value.

A method of monitoring accelerations on a vessel includes measuring acceleration on the vessel using one or more sensors. The one or more sensors are communicatively coupled to a computing unit. Real-time acceleration information representative of an acceleration on the vessel based at least in part on the measured acceleration from the one or more sensors is generated. Acceleration prediction information representative of predicted wave slam using the computing unit is generated. Using the acceleration prediction information, automatic control of trim, steering, or throttle controls of the vessel is performed in a fashion computed to reduce the effects of the predicted wave slam.

A method of monitoring accelerations on a vessel includes measuring acceleration on the vessel using one or more sensors. The one or more sensors are communicatively coupled to a computing unit. Real-time acceleration information representative of an acceleration on the vessel based at least in part on the measured acceleration from the one or more sensors is generated. Acceleration prediction information representative of predicted wave slam using the computing unit is generated. Using the acceleration prediction information, automatic control of trim, steering, or throttle controls of the vessel is performed in a fashion computed to reduce the effects of the predicted wave slam.

A low-order vessel propulsion power prediction method may be performed to determine factors, including power demand parameters, used in configuring a propulsion system for a marine vessel. The low-order method may receive stability data and vessel operation profile data, in addition to computational fluid dynamics simulation results to determine predicted vessel power profiles. The predicted vessel power profiles may be used to configure a powertrain system model for the marine vessel.

A low-order vessel propulsion power prediction method may be performed to determine factors, including power demand parameters, used in configuring a propulsion system for a marine vessel. The low-order method may receive stability data and vessel operation profile data, in addition to computational fluid dynamics simulation results to determine predicted vessel power profiles. The predicted vessel power profiles may be used to configure a powertrain system model for the marine vessel.

The present invention provides a system and method of side-stepping the need to retrain neural network model after initially trained using a simulator by comparing real-world data to data predicted by the simulator for the same inputs, and developing a mapping correlation that adjusts real world data toward the simulation data. Thus, the decision logic developed in the simulation-trained model is preserved and continues to operate in an altered reality. A threshold metric of similarity can be initially provided into the mapping algorithm, which automatically adjusts real world data to adjusted data corresponding to the simulation data for operating the neural network model when the metric of similarity between the real world data and the simulation data exceeds the threshold metric. Updated learning can continue as desired, working in the background as conditions are monitored.

The present invention provides a system and method of side-stepping the need to retrain neural network model after initially trained using a simulator by comparing real-world data to data predicted by the simulator for the same inputs, and developing a mapping correlation that adjusts real world data toward the simulation data. Thus, the decision logic developed in the simulation-trained model is preserved and continues to operate in an altered reality. A threshold metric of similarity can be initially provided into the mapping algorithm, which automatically adjusts real world data to adjusted data corresponding to the simulation data for operating the neural network model when the metric of similarity between the real world data and the simulation data exceeds the threshold metric. Updated learning can continue as desired, working in the background as conditions are monitored.

A method of monitoring accelerations on a vessel includes measuring acceleration on the vessel using one or more sensors. The one or more sensors are communicatively coupled to a computing unit. Real-time acceleration information representative of an acceleration on the vessel based at least in part on the measured acceleration from the one or more sensors is generated. Acceleration prediction information representative of predicted wave slam using the computing unit is generated. Using the acceleration prediction information, automatic control of trim, steering, or throttle controls of the vessel is performed in a fashion computed to reduce the effects of the predicted wave slam.