Methods and systems for implementing an iterated extended Kalman filter within a navigation system

• US 7,873,472 B2
• Filed: 10/07/2009
• Issued: 01/18/2011
• Est. Priority Date: 03/08/2006
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Leftin, Lawrence J., Zeng, Yueping, Briske, Richard W., Anderson, John F., Kirkland, Jeffrey
• Primary Examiner(s): Beaulieu; Yonel
• Application Number:
  US12/574,957
• Publication Number:
  US 20100036613A1
• Time in Patent Office: 468 Days
• Field of Search: 701/200, 701213-215, 701/220, 342/357.14
• US Class Current: 701/220
• CPC Class Codes: G01C 21/165 combined with non-inertial navigation instruments
• CPC Class Codes: G01C 21/188 for accumulated errors, e.g. by coupling inertial systems with absolute positioning systems

Abstract

Methods and systems for implementing an iterated extended Kalman filter within a navigation system are provided. In one embodiment, a Kalman filter programmed with time updating error states and measurement updating error states relating to received data sets is provided. The said Kalman filter configured to: iterate both the time updating error states and the measurement updating error states based on a first criteria; and iterate only the measurement updating error states based on a second criteria.

Claims

1. A Kalman filter programmed with time updating error states and measurement updating error states relating to received data sets, said Kalman filter configured to:

iterate both the time updating error states and the measurement updating error states when a first criteria is satisfied; and

iterate only the measurement updating error states when the first criteria is not satisfied.

2. The Kalman filter according to claim 1, wherein the first criteria is satisfied by data indicative of a stationary position for a vehicle and wherein the first criteria is not satisfied by data indicative of a changing position for the vehicle.

3. The Kalman filter according to claim 1, wherein the first criteria is satisfied by data indicative of a gyroscope compass alignment mode.

4. The Kalman filter according to claim 1, wherein a virtual time step is used to iterate at a higher frequency when the first criteria is satisfied than when the first criteria is not satisfied.

5. The Kalman filter according to claim 4, wherein using the virtual time step reduces a number of measurements required during the gyroscope compass alignment mode.

6. The Kalman filter according to claim 4, wherein using the virtual time step shortens an alignment time required during the gyroscope compass alignment mode.

7. The Kalman filter according to claim 1, wherein iteration of both time updating error states and measurement updating error states when the first criteria is satisfied is by a global iterated extended Kalman filter;

and

wherein iteration of only the measurement updating error states when the first criteria is not satisfied is by a local iterated extended Kalman filter.

8. A method comprising:

iterating both time updating error states in a Kalman filter and measurement updating error states in a Kalman filter when a first criteria is satisfied, the time updating error states and the measurement updating error states relating to received data sets; and

iterating only the measurement updating error states when the first criteria is not satisfied.

9. The method of claim 8, wherein the first criteria is satisfied by data indicative of a stationary position for a vehicle and wherein the first criteria is not satisfied by data indicative of a changing position for the vehicle.

10. The method of claim 8, wherein the first criteria is satisfied by data indicative of a gyroscope compass alignment mode.

11. The method of claim 8, further comprising:

using a virtual time step to iterate at a higher frequency when the first criteria is satisfied than when the first criteria is not satisfied.

12. The method of claim 11, wherein using the virtual time step reduces a number of measurements required during the gyroscope compass alignment mode.

13. The method of claim 11, wherein using the virtual time step shortens an alignment time required during the gyroscope compass alignment mode.

14. The method of claim 8, wherein iterating both time updating error states and measurement updating error states when the first criteria is satisfied is by a global iterated extended Kalman filter;

and

wherein iterating only the measurement updating error states when the first criteria is not satisfied is by a local iterated extended Kalman filter.