Systems and methods for sea state prediction and autonomous navigation in accordance with embodiments of the invention are disclosed. One embodiment of the invention includes a method of predicting a future sea state including generating a sequence of at least two 3D images of a sea surface using at least two image sensors, detecting peaks and troughs in the 3D images using a processor, identifying at least one wavefront in each 3D image based upon the detected peaks and troughs using the processor, characterizing at least one propagating wave based upon the propagation of wavefronts detected in the sequence of 3D images using the processor, and predicting a future sea state using at least one propagating wave characterizing the propagation of wavefronts in the sequence of 3D images using the processor. Another embodiment includes a method of autonomous vessel navigation based upon a predicted sea state and target location.

A hull behavior control system for controlling behavior of a hull of a marine vessel includes a memory and at least one controller coupled to the memory. The at least one controller is configured or programmed to control a steering that changes the traveling direction of the marine vessel, obtain a water surface shape around the marine vessel, estimate movement of a wave based on the water surface shape, and when it is determined that the hull rides the wave whose movement has been estimated, control the steering so as to reduce an influence of the wave on the hull.

A computer implemented method and apparatus for a marine vessel data system, the method comprising: receiving data from at least one sensor configured to measure vibration and operationally arranged to the marine vessel to provide time-domain reference sensor data; maintaining the time-domain reference sensor data within a data storage system; generating a Fast Fourier Transform (FFT) on the time-domain reference sensor data to provide a plurality of reference spectra files in frequency-domain, wherein each reference spectra file comprises spectra data defined by amplitude information and frequency information, and each spectra file is associated with condition information determined based on collection of the time-domain reference sensor data; normalizing each reference spectra file by converting the frequency information to order information using the condition information to provide normalized reference spectra files; and training a convolutional autoencoder type of neural network using the normalized reference spectra files.

A computer implemented method and apparatus for a marine vessel data system, the method comprising: receiving data from at least one sensor configured to measure vibration and operationally arranged to the marine vessel to provide time-domain reference sensor data; maintaining the time-domain reference sensor data within a data storage system; generating a Fast Fourier Transform (FFT) on the time-domain reference sensor data to provide a plurality of reference spectra files in frequency-domain, wherein each reference spectra file comprises spectra data defined by amplitude information and frequency information, and each spectra file is associated with condition information determined based on collection of the time-domain reference sensor data; normalizing each reference spectra file by converting the frequency information to order information using the condition information to provide normalized reference spectra files; and training a convolutional autoencoder type of neural network using the normalized reference spectra files.

A computer implemented method and apparatus for a marine vessel data system, the method comprising: receiving data from at least one sensor configured to measure vibration and operationally arranged to the marine vessel to provide time-domain reference sensor data; maintaining the time-domain reference sensor data within a data storage system; generating a Fast Fourier Transform (FFT) on the time-domain reference sensor data to provide a plurality of reference spectra files in frequency-domain, wherein each reference spectra file comprises spectra data defined by amplitude information and frequency information, and each spectra file is associated with condition information determined based on collection of the time-domain reference sensor data; normalizing each reference spectra file by converting the frequency information to order information using the condition information to provide normalized reference spectra files; and training a convolutional autoencoder type of neural network using the normalized reference spectra files.

A method for controlling a towing train including a ship and at least one tug acting on the ship, including the steps of: providing a data model, which includes fixed data of the ship and of the at least one tug as well as variable environmental data; determining the current course, the thrust vector, and the inertial force of the ship and specifying a desired travel direction of the ship with subsequent calculation of the correction force vector and correction torque required to achieve the desired travel direction; calculating the required positions, orientations, and drive settings of the at least one acting tug using an algorithm that accesses the data model and generating control commands for the at least one tug such that the sum of all the force vectors and torques of the at least one acting tug corresponds to the required correction force vector and correction torque; transmitting the generated control commands to at least one acting tug and monitoring the completion of the control commands; and conducting an evaluation of the produced correction force vector and correction torque after completion of the control commands and generating and storing correction values in the data model when deviations are detected between the produced correction force vector and the required correction force vector and/or between the produced correction torque and the required correction torque and then repeating certain steps.

A method for controlling a towing train including a ship and at least one tug acting on the ship, including the steps of: providing a data model, which includes fixed data of the ship and of the at least one tug as well as variable environmental data; determining the current course, the thrust vector, and the inertial force of the ship and specifying a desired travel direction of the ship with subsequent calculation of the correction force vector and correction torque required to achieve the desired travel direction; calculating the required positions, orientations, and drive settings of the at least one acting tug using an algorithm that accesses the data model and generating control commands for the at least one tug such that the sum of all the force vectors and torques of the at least one acting tug corresponds to the required correction force vector and correction torque; transmitting the generated control commands to at least one acting tug and monitoring the completion of the control commands; and conducting an evaluation of the produced correction force vector and correction torque after completion of the control commands and generating and storing correction values in the data model when deviations are detected between the produced correction force vector and the required correction force vector and/or between the produced correction torque and the required correction torque and then repeating certain steps.

The invention relates to a method (200) for determining the sloshing of a liquid loading of a vessel, said method (200) comprising:

a step (206) of determination of a monomodal excitation as a function of a multimodal excitation to which said vessel is subjected, the multimodal excitation comprising a state of the swell and a wind sea state, wherein the state of the swell comprises a direction of the swell and the wind sea state comprises a wind sea direction, and wherein the monomodal excitation exhibits a total direction equivalent to one out of the direction of the swell and the wind sea direction closest to a direction at right angles to the longitudinal axis of the vessel,

a step (208) of determination of a datum relating to the sloshing of said loading as a function of the monomodal excitation.

The invention relates to a method (200) for determining the sloshing of a liquid loading of a vessel, said method (200) comprising:

a step (206) of determination of a monomodal excitation as a function of a multimodal excitation to which said vessel is subjected, the multimodal excitation comprising a state of the swell and a wind sea state, wherein the state of the swell comprises a direction of the swell and the wind sea state comprises a wind sea direction, and wherein the monomodal excitation exhibits a total direction equivalent to one out of the direction of the swell and the wind sea direction closest to a direction at right angles to the longitudinal axis of the vessel,

a step (208) of determination of a datum relating to the sloshing of said loading as a function of the monomodal excitation.

A method for the reduction of ship fuel consumption through the optimisation of vessel draft, speed and trim using historical vessel data. Historical global, online data, is collected for multiple vessel operating parameters associated with its previous voyages. After initial filtering and cleaning of the gathered data, a process of analysing the data to determine the optimum draft, speed and trim for the vessels' given speed is described. The determined optimum draft, speed and trim values are then presented to the Captain or an automatic draft and trim optimisation system for the current draft and trim to be adjusted. This application therefore discloses a method for analysing historical vessel data to provide advice on optimum draft, trim and speed. A method for predicting the achievable fuel savings and recording the fuel savings achieved is also disclosed.