Data fusion architecture

Patent metadata

Field	Value
Publication number	US-9644959-B2
Application number	US-201313921418-A
Country	US
Kind code	B2
Filing date	Jun 19, 2013
Priority date	Jun 22, 2012
Publication date	May 9, 2017
Grant date	May 9, 2017

Abstract

A data fusion architecture with a plurality of sensors, optionally position measuring equipment (PMEs), is described. Each sensor supplies measurement data x 1 , x 2 . . . x M and is associated with accuracy data H 1 , H 2 . . . H M indicative of the accuracy of the supplied measurement data. Sub-processing units derives first estimates sf 1 , sf 2 . . . sf M and second estimates Hn 1 , Hn 2 . . . Hn M of the variability of the measurement data supplied by the respective sensor. The first estimates are derived by processing the measurement data x 1 , x 2 . . . x M and the second estimates are derived by processing the accuracy data H 1 , H 2 . . . H M . The first and second estimates are combined in a multiplier to derive overall estimates σ 1 , σ 2 . . . σ M of the variability of the measurement data supplied by the respective sensor. Data fusion means such as a Kalman filter combines the measurement data supplied by each sensor and the overall estimates σ 1 , σ 2 . . . σ M for each sensor.

First claim

What is claimed is:

1. Architecture comprising:
a plurality of sensors, each sensor supplying measurement data and being associated with accuracy data indicative of the accuracy of the supplied measurement data;
each sensor being associated with means for deriving first estimates of the variability of the measurement data supplied by the respective sensor by processing the measurement data supplied by the respective sensor, means for deriving second estimates of the variability of the measurement data supplied by the respective sensor by processing the accuracy data associated with the respective sensor, and means for combining the first and second estimates to derive overall estimates of the variability of the measurement data supplied by the respective sensor;
means for removing measurements which deviate from other measurements of the measurement data prior to deriving the first estimates of the variability of the measurement data supplied by the respective sensor;
means for combining the measurement data supplied by each sensor and the overall estimates for each sensor.

2. The architecture of claim 1 , wherein the plurality of sensors includes a combination of different types of sensors and/or different types of measurement data formats.

3. The architecture of claim 1 , wherein the means for deriving the first estimates derives estimates of the variability of the measurement data in the horizontal plane.

4. The architecture of claim 3 , wherein the first estimates are the standard deviation of the measurement data supplied by the respective sensor in two axes, or the size and orientation of an error ellipse for the measurement data.

5. The architecture of claim 1 , wherein each sensor is further associated with means for deriving a filtered version of the first estimates.

6. The architecture of claim 5 , wherein the means for deriving a filtered version of the first estimates is a first low-pass filter.

7. The architecture of claim 6 , wherein the means for deriving the second estimates includes a second low-pass filter for deriving a filtered version of the accuracy data associated with the respective sensor and means for dividing the accuracy data by the filtered version of the accuracy data, the first and second low-pass filters having substantially the same filter characteristics.

8. The architecture of claim 1 , wherein the means for deriving the second estimates includes means for deriving a filtered version of the accuracy data supplied by the respective sensor and means for dividing the accuracy data by the filtered version of the accuracy data.

9. The architecture of claim 8 , wherein the second estimates are a normalised, filtered version of the accuracy data supplied by the respective sensor.

10. The architecture of claim 8 , wherein the means for deriving a filtered version of the accuracy data is a second low-pass filter.

11. The architecture of claim 1 , wherein the means for combining the measurement data supplied by each sensor and the overall estimates for each sensor is a Kalman filter.

12. The architecture of claim 1 , wherein the means for combining the measurement data supplied by each sensor and the overall estimates for each sensor uses a weighted average technique.

13. The architecture of claim 1 , wherein the sensors are position measuring equipment.

14. A dynamic positioning system for a marine vessel using architecture comprising:
a plurality of sensors, each sensor supplying measurement data and being associated with accuracy data indicative of the accuracy of the supplied measurement data;
each sensor being associated with means for deriving first estimates of the variability of the measurement data supplied by the respective sensor by processing the measurement data supplied by the respective sensor, means for deriving second estimates of the variability of the measurement data supplied by the respective sensor by processing the accuracy data associated with the respective sensor, and means for combining the first and second estimates to derive overall estimates of the variability of the measurement data supplied by the respective sensor;
means for removing the effects of low frequency motion of the marine vessel on position measurement data provided by the respective sensor before the position measurement data is provided to the means for deriving the first estimates; and
means for combining the measurement data supplied by each sensor and the overall estimates for each sensor.

Assignees

• Ge Energy Power Conversion Technology Ltd

Inventors

• Stephens Richard Ian

Classifications

• B63H25/04

  automatic, e.g. reacting to compass · CPC title

• G01C21/28

  with correlation of data from several navigational instruments · CPC title

• G01C21/16

  by integrating acceleration or speed, i.e. inertial navigation · CPC title

• G01C21/20 Primary

  Instruments for performing navigational calculations (G01C21/24, G01C21/26 take precedence) · CPC title

• G01B21/16 Primary

  for measuring distance of clearance between spaced objects · CPC title