Inertial/magnetic measurement device

• US 7,587,277 B1
• Filed: 11/21/2005
• Issued: 09/08/2009
• Est. Priority Date: 11/21/2005
• Status: Active Grant

• Current Assignee: Miltec, LLC
• Original Assignee: Miltec Corporation (General Atomics, Inc.)
• Inventors: Wells, Randy
• Primary Examiner(s): Beaulieu; Yonel
• Application Number:
  US11/285,644
• Time in Patent Office: 1,387 Days
• Field of Search: 701/200, 701/220, 731/78.R, 342/357.14
• US Class Current: 701/220
• CPC Class Codes: G01C 21/1654 with electromagnetic compass

Abstract

Disclosed is a system and method for onboard optimal estimation of heading, pitch, and roll through real-time measurement of magnetic field, acceleration and angular motion in three dimensions. Magnetometer information is used to create an initial reference from which movement is measured. Thus, the process does not have to start when the body is in a known position. Further, the device does not have to continually rely on accelerometer data to get roll and pitch. To do this magnetic field data is used to complement gyro information. The magnetic data is used to estimate pitch, roll, and heading.

Claims

1. One or more computer-readable media adapted for use with a computing device which is adapted to execute at least one of direction and motion measurement, said media having computer-usable instructions embodied thereon for performing a method comprising:

computing body-sensed navigational frame magnetic field vector using sensed magnetic data;

calculating attitude residuals between a group of body-sensed navigational frame magnetic field components and an initial magnetic reference frame;

outputting said residuals to a Kalman filter process;

measuring the observability for an attitude axis for each of roll, pitch, and yaw;

producing a measurement-noise covariance matrix;

using said measurement-noise covariance matrix in said Kalman filter process;

detecting for motion;

generating said initial magnetic reference frame if no motion is detected;

taking running averages for magnetic field until a predetermined amount of time has elapsed; and

outputting said running averages to serve as said initial magnetic field reference frame in said calculating step.

2. One or more computer-readable media adapted for use with a comprising device which is adapted to execute at least one of direction and motion measurement, said media having computer-usable instructions embodied thereon for performing a method comprising:

computing body-sensed navigational frame magnetic field vector using sensed magnetic data;

calculating attitude residuals between a group of body-sensed navigational frame magnetic field components and an initial magnetic reference frame;

outputting said residuals to a Kalman filter process;

measuring the observability for an attitude axis for each of roll, pitch, and yaw;

producing a measurement-noise covariance matrix;

using said measurement-noise covariance matrix in said Kalman filter process; and

wherein said producing a measurement-noise covariance matrix step comprises;

using simulation generated empirical data to relate observability with angle measurement uncertainty.

3. The media of claim 2 wherein said using step in said method occurs immediately after said producing step.

4. One or more computer-readable media adapted for use with a computing device which is adapted to execute at least one of direction and motion measurement, said media having computer-usable instructions embodied thereon for performing a method comprising:

computing body-sensed navigational frame magnetic field vector using sensed magnetic data;

calculating attitude residuals between a group of body-sensed navigational frame magnetic field components and an initial magnetic reference frame;

outputting said residuals to a Kalman filter process;

measuring the observability for an attitude axis for each of roll, pitch, and yaw;

producing a measurement-noise covariance matrix;

using said measurement-noise covariance matrix in said Kalman filter process; and

said method including an intermittent gyro calibration module which assists in the calibration of gyro data, said intermittent gyro calibration module comprising;

sensing for motion;

low-pass filtering the gyro data;

taking running averages for bias estimations if no motion is detected; and

outputting said running averages for bias estimations to a gyro compensation process.

5. The media of claim 4, said intermittent gyro calibration process comprising:

monitoring an onboard compensated translational acceleration and a compensated onboard angular rate in detecting motion.

6. The media of claim 4, wherein said intermittent gyro calibration process comprises:

attenuating said gyro data by suppressing high frequency noise using a low pass filter.

7. The media of claim 4, wherein said intermittent gyro calibration process comprises:

providing a mechanism which is adapted to enable a user to manually terminate the taking of running averages and to cause these averages to be output to said gyro compensation process.

8. One or more computer-readable media adapted for use with an inertial measurement device, said media having computer-usable instructions embodied thereon for performing a method for of measuring at least one of direction and motion which makes use of magnetometer data, said method incorporating an anomaly filter process, said process comprising:

identifying whether an anomaly exists in said magnetometer data;

notifying other processes of the existence of an anomaly if one exists; and

submitting said magnetometer data to said other processes as reliable if no anomaly is identified

wherein said identifying process comprises;

computing a first magnitude of a first magnetic field vector which is derived from sensed onboard compensated magnetic field data;

computing a second magnitude of a second magnetic field vector which is derived from initial reference magnetic field data;

comparing said first magnitude to said second magnitude; and

executing said notifying step if said first and second magnitudes are significantly different.

9. The media of claim 8, wherein said identifying process comprises:

computing a third magnitude of a magnetometer-sensed angular rate vector;

computing a fourth magnitude of a gyro-sensed angular rate vector;

comparing said third magnitude to said fourth magnitude; and

executing said notifying step if said third and fourth magnitudes are significantly different.

10. One or more computer-readable media adapted for use with an inertial measurement device, said media having computer-usable instructions embodied thereon for performing a method for of measuring at least one of direction and motion which makes use of magnetometer data, said method incorporating an anomaly filter process, said process comprising:

identifying whether an anomaly exists in said magnetometer data;

notifying other processes of the existence of an anomaly if one exists; and

submitting said magnetometer data to said other processes as reliable if no anomaly is identified

wherein said identifying process comprises;

computing a first magnitude of a magnetometer-sensed angular rate vector;

computing a second magnitude of a gyro-sensed angular rate vector;

comparing said first magnitude to said second magnitude; and

executing said notifying step if said first and second magnitudes are significantly different.

11. One or more computer-readable media adapted for use with an inertial measurement device, said media having computer-usable instructions embodied thereon for performing a method for of measuring at least one of direction and motion which makes use of magnetometer data, said method incorporating an anomaly filter process, said process comprising:

identifying whether an anomaly exists in said magnetometer data;

notifying other processes of the existence of an anomaly if one exists; and

submitting said magnetometer data to said other processes as reliable if no anomaly is identified

wherein said process comprises;

estimating said anomaly;

filtering out said anomaly if said notifying step has occurred; and

outputting filtered values to at least one other process for the purpose of estimating at least one of position and motion.

12. The media of claim 11, wherein said process comprises:

taking the difference between an sensed onboard compensated magnetic field vector and an initial reference field vector to create a resultant vector to accomplish said estimating step; and

transmitting said resultant vector to a magnetometer compensation module for the purpose of removing the anomaly.

13. One or more computer-readable media having computer-usable instructions embodied thereon for performing when executed by a computing device a method comprising:

receiving magnetic data including a plurality of simultaneously body-sensed measurements from a plurality of differently-oriented magnetic-sensing devices mounted on a body;

using said magnetic data in calculating a magnetic-reading-based three-dimensional orientation for said body in relation to a known magnetic field.

14. The media of claim 13 wherein said calculating step comprises:

determining a cross product of (i) a first vector comprising a magnetic reference frame for said known magnetic field; and

(ii) a second vector derived from said body-sensed measurements;

determining a dot product of a norm of the first vector with the norm of the second vector; and

dividing said cross product by said dot product to determine the attitude innovations for use in a Kalman filter in determining an attitude axis for each of roll, pitch, and yaw enabling said calculation of said magnetic-reading-based three-dimensional orientation.

15. The media of claim 14, said method comprising:

producing a measurement-noise covariance matrix using simulation generated empirical data to relate observability with angle measurement uncertainty.

16. The media of claim 13, said method comprising:

using a magnetic field of the earth to serve as said known magnetic field.

17. The media of claim 16 wherein said calculating step comprises:

determining a cross product of (i) a first vector comprising a magnetic reference frame for said known magnetic field; and

(ii) a second vector derived from said body-sensed measurements;

determining a dot product of a norm of the first vector with the norm of the second vector; and

dividing said cross product by said dot product to determine the attitude innovations for use in a Kalman filter in determining an attitude axis for each of roll, pitch, and yaw enabling said calculation of said magnetic-reading-based three-dimensional orientation.

18. The media of claim 17, said method comprising:

producing a measurement-noise covariance matrix using simulation generated empirical data to relate observability with angle measurement uncertainty.

19. A process for measuring at least one of direction and motion, said process comprising:

computing body-sensed navigational frame magnetic field vector using sensed magnetic data from a plurality of magnatometers;

calculating attitude residuals between a group of body-sensed navigational frame magnetic field components and an initial magnetic reference frame using one or more computing devices;

outputting said residuals to a Kalman filter process module on said one or more computing devices;

measuring the observability for an attitude axis for each of roll, pitch, and yaw;

producing a measurement-noise covariance matrix using said one or more computing devices;

using said measurement-noise covariance matrix in said Kalman filter process module;

detecting for motion;

generating said initial magnetic reference frame if no motion is detected using said one or more computing devices;

taking running averages for magnetic field until a predetermined amount of time has elapsed using said one or more computing devices; and

outputting said running averages to serve as said initial magnetic field reference frame in said calculating step one said one or more computing devices.

20. An inertial measurement unit comprising:

a computer processing device, said computer processing device adapted to receive magnetic data including a plurality of simultaneously body-sensed measurements from a plurality of differently-oriented magnetic-sensing devices mounted on a body; and

said computing device adapted to calculate a magnetic-reading-based three-dimensional orientation for said body in relation to a known magnetic field using said magnetic data.

21. The inertial measurement unit of claim 20 wherein said computer processing device includes a module for determining a cross product of (i) a first vector comprising a magnetic reference frame for said known magnetic field;

and (ii) a second vector derived from said body-sensed measurements;

determining a dot product of a norm of the first vector with the norm of the second vector; and

dividing said cross product by said dot product to determine the attitude innovations for use in a Kalman filter in determining an attitude axis for each of roll, pitch, and yaw enabling said calculation of said magnetic-reading-based three-dimensional orientation.

22. The inertial measurement unit of claim 21 wherein said module includes a process for producing a measurement-noise covariance matrix using simulation generated empirical data to relate observability with angle measurement uncertainty.

23. The inertial measurement unit of claim 20 wherein said computer processing device includes a process adapted to use a magnetic field of the earth to serve as said known magnetic field.

24. The inertial measurement unit of claim 23 including a calculation module which:

determines a cross product of (i) a first vector comprising a magnetic reference frame for said known magnetic field; and

(ii) a second vector derived from said body-sensed measurements;

determines a dot product of a norm of the first vector with the norm of the second vector; and

divides said cross product by said dot product to determine the attitude innovations for use in a Kalman filter in determining an attitude axis for each of roll, pitch, and yaw enabling said calculation of said magnetic-reading-based three-dimensional orientation.

25. The inertial measurement device of claim 24 wherein said calculation module comprises a process for producing a measurement-noise covariance matrix using simulation generated empirical data to relate observability with angle measurement uncertainty.