Characterising wave properties based on measurement data using a machine-learning model
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for estimating wave properties of a body of water. A computer-implemented system obtains measurement data for a duration of time from an inertial measurement unit (IMU) onboard an underwater device, generates model input data based on at least the measurement data obtained at the plurality of time points, and processes the model input data to generate model output data indicating one or more wave properties using a machine-learning model. The system further determines, based on at least the one or more wave properties, whether the device is safe to be deployed.
1. A computer-implemented method comprising:
obtaining, by an underwater camera device, inertial measurement unit (IMU) data that reflects motion of the underwater camera device over a period of time;
generating, by the underwater camera device, particular model input data based at least on the IMU data;
providing, by the underwater camera device, the particular model input data to a machine learning model that is trained to output estimated wave properties based on model input data;
receiving, by the underwater camera device, a particular estimated wave property; and
determining, by the underwater camera device, a future device position for the underwater camera device after receiving the particular estimated wave property.
2. The method of claim 1 , wherein generating the particular model input data comprises transforming movement or positional parameters of the IMU data from time-domain data to frequency-domain data.
3. The method of claim 1 , wherein determining a future position for the underwater camera device comprises determining to move to a stowed position for the underwater camera device.
4. The method of claim 1 , wherein the particular estimated wave property comprises an estimated value on a Douglas sea scale.
5. The method of claim 1 , wherein determining a future device position comprises determining to move to a predetermined distance from a cage structure.
6. The method of claim 1 , wherein determining a future device position comprises determining to move to a center of a cage structure.
7. The method of claim 1 , wherein determining a future device position comprises determining that the particular estimated wave property satisfies a threshold.
8. A system comprising:
one or more computers; and
one or more storage devices storing instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
obtaining, by an underwater camera device, inertial measurement unit (IMU) data that reflects motion of the underwater camera device over a period of time;
generating, by the underwater camera device, particular model input data based at least on the IMU data;
providing, by the underwater camera device, the particular model input data to a machine learning model that is trained to output estimated wave properties based on model input data;
receiving, by the underwater camera device, a particular estimated wave property; and
determining, by the underwater camera device, a future device position for the underwater camera device after receiving the particular estimated wave property.
9. The system of claim 8 , wherein generating the particular model input data comprises transforming movement or positional parameters of the IMU data from time-domain data to frequency-domain data.
10. The system of claim 8 , wherein determining a future position for the underwater camera device comprises determining to move to a stowed position for the underwater camera device.
11. The system of claim 8 , wherein the particular estimated wave property comprises an estimated value on a Douglas sea scale.
12. The system of claim 8 , wherein determining a future device position comprises determining to move to a predetermined distance from a cage structure.
13. The system of claim 8 , wherein determining a future device position comprises determining to move to a center of a cage structure.
14. The system of claim 8 , wherein determining a future device position comprises determining that the particular estimated wave property satisfies a threshold.
15. A non-transitory computer-readable storage medium that stores instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:
obtaining, by an underwater camera device, inertial measurement unit (IMU) data that reflects motion of the underwater camera device over a period of time;
generating, by the underwater camera device, particular model input data based at least on the IMU data;
providing, by the underwater camera device, the particular model input data to a machine learning model that is trained to output estimated wave properties based on model input data;
receiving, by the underwater camera device, a particular estimated wave property; and
determining, by the underwater camera device, a future device position for the underwater camera device after receiving the particular estimated wave property.
16. The medium of claim 15 , wherein generating the particular model input data comprises transforming movement or positional parameters of the IMU data from time-domain data to frequency-domain data.
17. The medium of claim 15 , wherein determining a future position for the underwater camera device comprises determining to move to a stowed position for the underwater camera device.
18. The medium of claim 15 , wherein the particular estimated wave property comprises an estimated value on a Douglas sea scale.
19. The medium of claim 15 , wherein determining a future device position comprises determining to move to a predetermined distance from a cage structure.
20. The medium of claim 15 , wherein determining a future device position comprises determining to move to a center of a cage structure.