Ship motion predictor

• US 20050102072A1
• Filed: 01/30/2003
• Published: 05/12/2005
• Est. Priority Date: 02/16/2002
• Status: Active Grant

• Current Assignee: BAE Systems Plc
• Original Assignee: BAE Systems Plc
• Inventors: Deakin, Anthony Michael
• Granted Patent: US 7,174,242 B2
• Time in Patent Office: Days
• US Class Current: 701/21
• CPC Class Codes: B63B 39/14 for indicating inclination or duration of roll
• CPC Class Codes: B63B 49/00 Arrangements of nautical instruments or navigational aids nautical measuring instruments G01C; radio navigation, analogous arrangements using other waves G01S
• CPC Class Codes: G01S 13/86 Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
• CPC Class Codes: G01S 2013/0272 Multifunction radar
• CPC Class Codes: G01S 7/4004 of parts of a radar system
• CPC Class Codes: H01Q 1/18 Means for stabilising antennas on an unstable platform

Abstract

There is disclosed a predictive filter for predicting the future position of a look direction of a rotating directional antenna mounted on a vessel. Prediction of a look direction of the antenna at any instant is made from input of ship'"'"'s compass data including roll data, pitch data and heading data, as well as data describing a rotational position of the antenna, output from an antenna synchro device. Smoothing of the digital roll, pitch and heading data and antenna position data is followed by application of a moving time window, filter which removes any offset from the data within of the moving time window to produce data having a zero mean. The zero mean data is input into an auto-regression algorithm, the output of which is combined with mean roll, pitch, heading, position and antenna velocity data to obtain predicted positions of the antenna relative to the vessel, and a predicted motion of the vessel. The combined vessel motion prediction and antenna position and velocity prediction can be used to establish a predictive future look direction of the antenna for a period of between 0.1 and 5 seconds into the future.

Claims

1. A method of predicting an orientation of a vessel, said method comprising the steps of:

inputting roll data describing a roll of said vessel;

inputting pitch data describing a pitch of said vessel;

inputting heading data describing a heading of said vessel;

applying a travelling mean window to said roll data, pitch data and heading data to obtain a time moving average roll data, time moving average pitch data; and

time moving average heading data;

filtering any offset data of said time moving average roll data, time moving average pitch data and time moving average heading data, such that said time moving average roll data, pitch data and heading data has a zero mean value;

applying an auto-regression process to said zero mean time moving average roll data, pitch data and heading data;

predicting angles of movement of major three dimensional X, Y and Z axes of said vessel from an output of said auto-regression process, and said mean roll data, pitch data and heading data; and

storing a result of said predicted angles as a set of rotation matrix data.

2. The method as claimed in claim 1, further comprising the steps of:

inputting an antenna position data describing a rotational position of an antenna;

inputting an antenna velocity data describing a rotation rate of said antenna; and

smoothing said antenna position data and antenna velocity data by processing said antenna position data and said antenna velocity data through a polynomial filter.

3. The method as claimed in claim 2, wherein said polynomial filter uses a time window in the range 0.5 to 2 seconds.

4. The method as claimed in claim 2, comprising the step of applying a travelling mean filter to said antenna position data and said antenna velocity data.

5. The method as claimed in claim 2, comprising the step of applying a travelling mean filter to said antenna position data and said antenna velocity data, by computing a mean value of said data within a time window in the range 25 to 100 seconds.

6. The method as claimed in claim 2, comprising generating a predicted antenna velocity data by applying a Marple algorithm to said antenna velocity data.

7. The method as claimed in claim 1, wherein said step of prediction of angles of movement comprises the steps of:

starting from a time and angle of said vessel'"'"'s previous fore-aft datum, computing a position of said vessel with a linear law using a data window mean velocity; and

adding a correction calculated by integrating an auto-regression predicted velocity differential over a same time interval as said data window.

8. The method as claimed in claim 1, further comprising the step of:

applying a three point Lagrange interpolation.

9. A method of predicting a true orientation of an antenna, said antenna rotatably mounted on a vehicle, said method comprising the steps of:

inputting a stream of antenna rotation data describing a current measurement of a rotational position and rotational velocity of said antenna with respect to said vehicle;

applying a moving time window sample to said antenna rotation data to obtain a time moving average antenna rotation data;

filtering any offset data of said time moving average antenna rotation data, such that said time moving average antenna rotation data has a zero mean value;

applying auto-regression to said zero mean time moving average antenna rotation data;

predicting a mean antenna rotation data;

predicting an orientation of major three dimensional X, Y and Z axes of said vehicle; and

storing a result of said predicted vehicle orientation and predicted antenna rotation as a set of rotation matrix data, said rotation matrix data describing a predicted antenna position in relation to a set of geographical axes fixed relative to the earth.

10. The method as claimed in claim 9, comprising smoothing said antenna rotation data by processing said antenna rotation data through a polynomial filter.

11. The method as claimed in claim 10, wherein said polynomial filter uses a time window in the range 0.5 to 2 seconds.

12. The method as claimed in claim 9, wherein said step of applying a moving time window sample to said antenna rotation data comprises:

computing a mean value of said data within a time window in the range 25 to 100 seconds.

13. The method as claimed in claim 9, comprising predicting an antenna velocity data by applying a Marple algorithm to said antenna velocity data.

14. An apparatus for predicting an orientation of a vessel, said apparatus comprising:

a roll data generator for generating roll data describing a roll of said vessel;

a pitch data generator for generating pitch data describing a pitch of said vessel;

a heading data generator for generating heading data describing a heading of said vessel;

a sampler for applying a moving time window sample to said roll data, pitch data and heading data to obtain a time moving average roll data, pitch data and heading data;

at least one filter for filtering any offset data of said time moving average roll data, pitch data and heading data, such that said time moving average roll data, pitch data and heading data has a zero mean value;

at least one auto-regressor for applying auto-regression to said zero mean time moving average roll data, pitch data and heading data;

at least one angle predictor for predicting angles of movement of major X, Y and Z three dimensional axis of said vessel from an output of said auto-regression, and a mean roll data, pitch data and heading data; and

a memory device for storing a result of said predicted angles as a set of rotation matrix data.

15. An apparatus for predicting an orientation of a rotating antenna mounted on a vessel, said apparatus comprising:

an antenna position data generator for generating an antenna rotation position data;

a sampler for applying a moving time window sample to said antenna position data to obtain a time moving average antenna position data;

at least one filter for filtering any offset data of said time moving average antenna position data, such that said time moving average antenna position data has a zero mean value;

at least one auto-regressor for applying auto-regression to said zero mean time moving average antenna position data;

at least one predictor for predicting a mean antenna position data; and

a memory device for storing a result of said predicted mean antenna position data as a set of rotation matrix data.

16. An apparatus for predicting an orientation of an antenna said apparatus comprising:

an antenna velocity data generator for generating an antenna velocity data;

a sampler for applying a moving time window sample to said antenna velocity data to obtain a time moving average antenna velocity data;

at least one filter for filtering any offset data of said time moving average antenna velocity data such that said time moving average antenna velocity data has a zero mean value;

at least one auto-regressor for applying auto-regression to said zero mean time moving average antenna velocity data;

at least one predictor for predicting a mean antenna velocity data; and

a memory for storing a result of said predicted mean antenna velocity data as a set of rotation matrix data.

17. An apparatus for predicting an orientation of a vessel, said apparatus comprising:

a roll data generator for generating roll data describing a roll of said vessel;

a pitch data generator for generating pitch data describing a pitch of said vessel;

a heading data generator for generating heading data describing a heading of said vessel;

a sampler for applying a moving time window sample to said roll data, pitch data and heading data to obtain a time moving average roll data, pitch data and heading data;

a filter for filtering any offset data of said time moving average roll data, pitch data and heading data, such that said time moving average roll data, pitch data and heading data has a zero mean value;

an auto regressor for applying auto-regression to said zero mean time moving average roll data, pitch data and heading data;

an angle predictor for predicting angles of movement of major X, Y and Z three dimensional axes of said vessel from an output of said auto-regression, and a mean roll data, pitch data and heading data;

a memory device for storing a result of said predicted angles as a set of rotation matrix data;

an antenna rotation data generator for generating an antenna rotation data;

a sampler for applying a moving time window sample to said antenna rotation data to obtain a time moving average antenna rotation data;

at least one filter for filtering any offset data of said time moving average antenna rotation data such that said time moving average antenna rotation data has a zero mean value;

at least one auto-regressor for applying auto-regression to said zero mean time moving average antenna rotation data;

at least one predictor for predicting a mean antenna rotation data; and

a memory device for storing a result of said predicted mean antenna rotation data as a set of rotation matrix data.