KALMAN FILTER FOR INDOOR POSITIONING

• US 20140368386A1
• Filed: 06/13/2013
• Published: 12/18/2014
• Est. Priority Date: 06/13/2013
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: MANSOUR, Mohamed Farouk
• Granted Patent: US 9,459,339 B2
• Time in Patent Office: Days
• US Class Current: 342/461
• CPC Class Codes: G01S 5/0294 Trajectory determination or predictive filtering, e.g. target tracking or Kalman filtering

Abstract

A position estimation method for indoor positioning includes filtering an initial position estimate that includes a corresponding covariance that reflects the quality of the geometry of the reference points and a previous initial position estimate that includes a corresponding covariance that reflects the quality of the geometry of the reference points by a Kalman filter to generate an updated previous position estimate, analyzing the updated previous position estimate to determine if the value is outside of a range, and constraining the updated previous position estimate based on the value being outside of the range.

Claims

1. A position estimation method, comprising:

filtering, by a Kalman filter, an initial position estimate that includes a corresponding covariance and a previous initial position estimate that includes a corresponding covariance to generate an updated previous position estimate;

analyzing the updated previous position estimate to determine if the updated previous position estimate is outside of a range; and

constraining the updated previous position estimate based on the value being outside of the range.

2. The method of claim 1, further comprising filtering the initial position estimate by another Kalman filter to generate an instantaneous initial position estimate that includes a corresponding covariance.

3. The method of claim 1, wherein the initial position estimate is the instantaneous initial position estimate.

4. The method of claim 1, further comprising providing a correction measurement with a corresponding small covariance to the Kalman filter to constrain the updated previous position estimate.

5. The method of claim 1, wherein the corresponding covariance associated with the initial position estimate is computed using a Wi-Fi dilution of position (WDOP), which is the covariance associated with the initial position estimate.

6. The method of claim 5, wherein the WDOP reflects a quality of the geometry of a plurality of wireless local area network access points.

7. A positioning system, comprising:

a position estimator to determine a current position estimate and a corresponding covariance derived from a Wi-Fi dilution of position (WDOP) algorithm;

a position smoother to update a previous position estimate using the current position estimate and a delayed position estimate; and

a constraints module to limit a change in the updated previous position estimate with respect to a prior updated previous position estimate.

8. The system of claim 7, the position smoother comprising:

a delay element to delay a position estimate;

a multiplexer (MUX) to receive the current position estimate; and

a first Kalman filter coupled to an output of the delay element and an output of the MUX to update the previous position estimate.

9. The system of claim 7, further comprising:

a second Kalman filter to filter the current position estimate and an output of the second Kalman filter coupled to an input of the MUX.

10. The system of claim 8, wherein the MUX receives a control signal that directs the MUX to output either the current position estimate or the output of the second Kalman filter.

11. The system of claim 7, wherein the constraints module the position smoother, and the constraints module is to analyze the updated previous position estimate to determine if the data is outside of a range.

12. The system of claim 11, wherein the constraints module provides a correction measurement and a small covariance to the first Kalman filter based on the constraints module determining that the updated previous position estimate is outside of the range to limit the previous position estimate.

13. The system of claim 12, wherein the position estimator calculates the corresponding covariance using a Wi-Fi dilution of position (WDOP) algorithm that reflects a quality of the geometry of a plurality of wireless local area network access points based on the initial position estimate.

14. A wireless indoor positioning device, comprising:

a delay element and a multiplexer to receive a current position estimate that includes a corresponding covariance and the delay element to delay the current position estimate;

a first Kalman filter coupled to the output of the delay element and an output of a multiplexer (MUX) to generate an updated previous position estimate using the current position estimate and a delayed position estimate; and

a constraints module, coupled to the first Kalman filter, to limit the change in the updated position estimate based on a difference between a prior updated previous position estimate and the updated previous position estimate, wherein the constraint will not be applied if the difference is within a range.

15. The device of claim 14, wherein the constraints module limits the updated previous position estimate based on the difference being outside of the range.

16. The device of claim 14, further comprising a second Kalman filter to receive the current position estimate and to generate an instantaneous position estimate.

17. The device of claim 16, wherein an output of the second Kalman filter is coupled to an input of the MUX and the MUX outputs either the current position estimate or the instantaneous position estimate based on a control signal.

18. The device of claim 14, wherein the corresponding covariance is calculated by a measurement engine using a Wi-Fi dilution of precision method (WDOP) that reflects a quality of the geometry of a plurality of wireless local area network access points based on the initial position estimate.

19. The device of claim 14, wherein the constraints module provides a correction measurement with a corresponding small covariance to limit the updated previous position estimate.

20. The device of claim 14, wherein the current position estimate contains a coordinate and a velocity describing the current estimate of the device'"'"'s position and an estimate of the velocity of the device.