LOW-COMPLEXITY TIGHTLY-COUPLED INTEGRATION FILTER FOR SENSOR-ASSISTED GNSS RECEIVER

• US 20100079334A1
• Filed: 09/28/2009
• Published: 04/01/2010
• Est. Priority Date: 09/24/2008
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Waters, Eric W., Hosur, Srinath, Dutta, Goutam, Roh, June Chul
• Granted Patent: US 8,380,433 B2
• Time in Patent Office: Days
• US Class Current: 342/357.290
• CPC Class Codes: G01C 21/1654 with electromagnetic compass
• CPC Class Codes: G01C 21/28 with correlation of data from several navigational instruments
• CPC Class Codes: G01C 22/006 Pedometers
• CPC Class Codes: G01S 19/47 the supplementary measurement being an inertial measurement, e.g. tightly coupled inertial
• CPC Class Codes: G06F 17/16 Matrix or vector computation , e.g. matrix-matrix or matrix-vector multiplication, matrix factorization matrix transposition G06F7/78

Abstract

Embodiments of the invention provide a blending filter based on extended Kalman filter (EKF), which optimally integrates the IMU navigation data with all other satellite measurements tightly-coupled integration filter. This blending filter can be easily implemented with minor modification to the position engine of stand-alone GNSS receiver. Provided is a low-complexity tightly-coupled integration filter for sensor-assisted global navigation satellite system (GNSS) receiver. The inertial measurement unit (IMU) contains inertial sensors such as accelerometer, magnetometer, and/or gyroscopes Embodiments also include method for pedestrian dead reckoning (PDR) data conversion for ease of GNSS/PDR integration. The PDR position data is converted to user velocity measured at the time instances where GNSS position/velocity estimates are available.

Claims

1. An apparatus comprising:

an integration filter for a sensor-assisted global navigation satellite system (GNSS) receiver

a GNSS measurement engine for providing GNSS measurement data to said integration filter;

an inertial measurement unit (IMU); and

an inertial navigation system (INS) block for calculating navigation information using a plurality of inertial sensor outputs.

2. The apparatus of claim 1, wherein said inertial sensors in said IMU is selected from the list consisting of:

accelerometer;

magnetometer; and

gyroscope.

3. The apparatus of claim 1, wherein said integration filter performs a blending operation for combining said GNSS measurement data and said INS navigation information.

4. The apparatus of claim 1, wherein said integration filter is selected from the list consisting of:

a Kalman Filter,

a Linearized Kalman Filter (LKF),

an Extended Kalman Filter (EKF),

an Unscented Kalman Filter (UKF),

a particle filter,

a least squares filter, and

an intelligent/learning filter.

5. The apparatus of claim 1, wherein said measurement data comprises a pseudorange measurement and a delta range measurement for at least one satellite.

6. The apparatus of claim 1, wherein measurement engine providing measurement noise variance for said pseudorange measurement and said delta range measurements.

7. The apparatus of claim 1, wherein said inertial navigation system (INS) block calculates INS user velocity data using a plurality of inertial sensor outputs, and provides said INS user velocity data to said integration filter.

8. The apparatus of claim 7, wherein said INS user velocity data from said INS is synchronized to GNSS measurement data.

9. The apparatus of claim 7, wherein INS providing measurement noise variance for said INS user velocity data.

10. The apparatus of claim 1, wherein a state definition and a system equation of said integration filter is the same as those of a stand-alone GNSS position engine, but a plurality of measurements from said INS can be added in a measurement equation of said integration filter to have a blended navigation information.

11. The apparatus of claim 10, wherein integration filter processes said INS user velocity data from said INS in said measurement equation of said integration filter using a method comprising:

a plurality of INS measurements in a local navigation coordinate are included in said measurement equation in a way that said INS measurements are a function of velocity variables of an integration filter state with a plurality of measurement noises.

12. The apparatus of claim 11, wherein a vertical INS measurement from said plurality of INS measurements is set to zero.

13. The apparatus of claim 11, wherein said INS measurement includes a speed measurement using a magnitude of an INS velocity vector on horizontal navigation plane.

14. The apparatus of claim 11, wherein said INS measurement includes a heading measurement using an angle of said INS velocity vector on horizontal navigation plane.

15. The apparatus of claim 11, wherein said INS measurement includes a speed measurement using a magnitude of an INS velocity vector on horizontal navigation plane and a heading measurement using an angle of said INS velocity vector on horizontal navigation plane.

16. The apparatus of claim 3, wherein said integration filter is configured as a stand-alone GNSS position engine and does not integrate said INS measurements when GNSS measurements are reliable when GNSS signal condition is good.

17. The apparatus of claim 16, wherein said GNSS signal condition metric is the number of GNSS satellites whose signal level with respect to noise level is greater than a threshold.

18. The apparatus of claim 16, wherein said GNSS signal condition metric is GNSS position uncertainty metric.

19. The apparatus of claim 16, wherein said GNSS signal condition metric is GNSS velocity uncertainty metric.

20. The apparatus of claim 3, wherein integration filter uses time-varying measurement noise variances for said GNSS measurements and said INS measurement in calculating blended navigation information.

21. The apparatus of claim 20, wherein said measurement noise variances per a satellite vehicle (SV) for GNSS measurement are determined based on a GNSS signal quality, wherein said measurement noise is time-varying and location-dependent.

22. The apparatus of claim 21, determination of said measurement noise variance for INS measurement is based on a subset from the group consisting of:

accuracy of sensors,

mounting condition of IMU, and

dynamics of the receiver.

23. The apparatus of claim 3, wherein said measurement noise variance for INS measurement is set to a constant (not time-varying) and said INS measurement is used in calculating blended navigation information.

24. The apparatus of claim 1, wherein said INS is a motion detector operable to provide motion information of user and wherein low-cost sensors can be used even if sensor is uncalibrated.

25. The apparatus of claim 24, wherein said when said motion detector determines when said user is stationary or has negligible motions zeros are used in the INS user velocity data with low measurement variances.

26. The apparatus of claim 24, wherein said when motion detector determines when said user is in motion, said integration filter is configured as stand-alone GNSS position engine and does not integrate said INS measurements.

27. A method of step detection comprising:

obtaining accelerometer measurement a(k), where a(k) is multi-dimensional acceleration vector at a k-th sample;

filtering the magnitude of said accelerometer measurement |a(k)| using a low-pass filter;

providing a threshold for a down-crossing (dc);

providing a threshold for an up-crossing (uc); and

trigging a step detection if said magnitude of said accelerometer measurement |a(k)| is greater than or equal to said threshold for an up-crossing (uc).

28. The method of claim 27, wherein trigging a step detection if said magnitude of said accelerometer measurement |a(k)| is less than or equal to said threshold for an down-crossing (dc).

29. The method of claim 27, further comprising transitioning a static state if there is no up-crossing or down-crossing event for a predetermined time period.

30. The method of claim 27, further comprising transitioning to a down-crossing (dc) state if said magnitude of said accelerometer measurement |a(k)| is equal to or below said threshold for a down-crossing (dc).

31. The method of claim 27, further comprising:

providing a time window from time instance for the latest down-crossing; and

triggering a step detection when a transition to an up-crossing (uc) state from down-crossing (dc) state within said time window.

32. A method for pedestrian dead reckoning (PDR) data conversion for ease of global navigation satellite system/pedestrian dead reckoning (GNSS/PDR) integration by converting PDR position data to user velocity measured at a plurality of time instances where GNSS position/velocity estimates are available, said method comprising:

obtaining a heading and a step length for a previous step;

calculating a partial step using said heading and said step interval for said previous step;

obtaining a pedestrian dead reckoning (PDR) position at a current GNSS clock, comprising;

adding said partial step to said PDR position, if a step detection state is not in Static;

keeping said position for said previous step, if said step detection state is in Static; and

taking a difference between them to obtain said user velocity.

33. The method of claim 32, further comprising updating said position at the previous GNSS clock by interpolating with a plurality measured step instances.

34. A method of blending velocity data from an inertial navigation system (INS) in a measurement equation of GNSS/IMU integration filter, said method comprising:

creating a coordinate transformation matrix with a plurality of measurement noises;

including a plurality of INS measurements in a local navigation coordinate in said measurement equation in a way that said INS measurements are a function of velocity variables of an integration filter state with a plurality of measurement noises; and

outputting a blended position fix.

35. The method of claim 34, wherein a vertical INS measurement in said plurality of INS measurements is set to zero.

36. The method of claim 34, wherein said INS user velocity data includes a speed measurement using a magnitude of an INS velocity vector on horizontal navigation plane.

37. The method of claim 34, wherein said INS user velocity data includes a heading measurement using an angle of said INS velocity vector on horizontal navigation plane.

38. The method of claim 34, wherein said INS user velocity data includes a speed measurement using a magnitude of an INS velocity vector on horizontal navigation plane and a heading measurement using an angle of said INS velocity vector on horizontal navigation plane.

39. The method of claim 34, wherein integration filter is configured as a stand-alone GNSS position engine and does not integrate said INS measurements when GNSS measurements are reliable when GNSS signal condition is good.

40. The method of claim 34, further comprising determining a measurement noise variance per a satellite vehicle (SV) for GNSS measurement based on a signal quality, wherein said measurement noise is time-varying and location-dependent.

41. The method of claim 39, wherein determination of said measurement noise variance for INS user velocity is selected from the group consisting of:

accuracy of sensors, mounting condition of IMU, and dynamics of the receiver.