Fast decimeter-level GNSS positioning

• US 20080165053A1
• Filed: 04/09/2007
• Published: 07/10/2008
• Est. Priority Date: 04/17/2006
• Status: Active Grant

• Current Assignee: Trimble Navigation Limited (Trimble Inc.)
• Original Assignee: Trimble Navigation Limited (Trimble Inc.)
• Inventors: Liu, Junjie, West, Peter Ian, Vollath, Ulrich, Klose, Soeren Ulf
• Granted Patent: US 7,498,979 B2
• Time in Patent Office: Days
• US Class Current: 342/357.03
• CPC Class Codes: G01S 19/32 Multimode operation in a single same satellite system, e.g. GPS L1/L2
• CPC Class Codes: G01S 19/44 Carrier phase ambiguity resolution; Floating ambiguity; LAMBDA [Least-squares AMBiguity Decorrelation Adjustment] method

Abstract

Methods and apparatus for processing of data from GNSS receivers are presented. A post-processing engine and a post-processed accuracy predictor are described. The post-processing engine provides high accuracy GNSS (GPS) position determination with short occupation time for GIS applications. The post-processed accuracy predictor calculates during data collection an estimate of the accuracy likely to be achieved after post-processing. This helps to optimize productivity when collecting GNSS data for which post-processed accuracy is important. The predictor examines the quality of carrier measurements and estimates how well the post-processed float solution will converge in the time since carrier lock was obtained.

Claims

1. A method of preparing GNSS reference data for processing data acquired by a GNSS receiver at times corresponding to first intervals, comprising:

a. Accessing reference data having measurements taken at times corresponding to second intervals longer than the first intervals,

b. Determining a geometric component for each measurement,

c. Producing for each measurement a geometry-free residual by subtracting the geometric component from the measurement,

d. Interpolating the geometry-free residuals to produce interpolated residuals for times corresponding to the first intervals,

e. Determining a geometric component for each time corresponding to an interpolated residual,

f. For each time corresponding to the first intervals, combining the interpolated residual for that time with the geometric component for that time to produce an interpolated reference measurement for that time.

2. The method of claim 1, wherein interpolating the geometry-free residual for given time comprises preparing a linear interpolation over a second interval spanning the given time.

3. The method of claim 2, wherein preparing a linear interpolation comprises determining phase change rate of the residuals.

4. The method of claim 2, wherein the measurements are taken from GNSS signals of at least two carrier frequencies, and wherein preparing a linear interpolation comprises determining a phase-change rate and determining an interpolated code residual using the phase-change rate.

5. The method of claim 1, wherein interpolating the geometry-free residual for a given time comprises running a linear Kalman filter over period including the given time.

6. The method of claim 1, wherein the reference data comprises carrier-phase measurements, and wherein the method comprises performing steps 1.a.-1.e. for the carrier-phase measurements.

7. The method of claim 1, wherein the reference data comprises code measurements, and wherein the method comprises performing steps 1.a.-1.e. for the code measurements.

8. The method of claim 6, wherein the reference data comprises code measurements, and wherein the method comprises performing steps 1.a.-1.e. for the code measurements.

9. The method of claim 1, further comprising processing data acquired by a GNSS receiver during an acquisition period by,

g. Accessing GNSS reference data acquired during a reference period substantially longer than the acquisition period, the GNSS reference data comprising code measurements and carrier-phase measurements,

h. Preparing a geometry-free and ionosphere-free code-minus-carrier observable from GNSS reference data,

i. Averaging the code-minus-carrier observable over the reference period to obtain respective mean carrier-ambiguity values for each of a plurality of carrier frequencies,

j. Producing a set of pre-processed reference data by combining the mean carrier-ambiguity values with interpolated reference measurements for multiple times during the acquisition period.

10. The method of claim 9, wherein the interpolated reference measurements are interpolated carrier-phase measurements.

11. The method of claim 8, wherein the acquisition period ranges from one minute to several minutes and wherein the reference period ranges from one hour to several hours.

12. The method of claim 10, further comprising preparing single-differenced minimum-error combination observations between a set of data acquired by a GNSS rover and the pre-processed reference data

13. The method of claim 12, further comprising filtering the single-difference minimum-error combination observations with a Kalman filter having states for receiver position, receiver clock and ambiguities.

14. The method of claim 12, wherein the minimum-error combination observations comprise L-1 phase and an L1-phase plus L1-code combination.

15. The method of claim 13, wherein the Kalman filter has two rover-clock states and alternates between the two rover-clock states when processing successive epochs of data.

16. The method of claim 15, wherein the Kalman filter has two rover-position states and alternates between the rover-position states when processing successive epochs of data.

17. The method of claim 16, wherein the Kalman filter has two rover-position states, further comprising determining from the rover position states whether the receiver has changed position by more than a predetermined amount from a prior epoch to a current epoch

18. The method of claim 17, further comprising estimating change of rover position between a prior epoch and a current epoch from phase measurement differences between the prior epoch and the current epoch.

19. The method of claim 18, wherein estimating change of rover position comprises applying a decorrelated delta-phase observable.

20. The method of claim 1, further comprising applying a whitening of noise technique to reduce non-Gaussian positioning errors.

21. The method of claim 9, further comprising applying a whitening of noise technique to reduce non-Gaussian positioning errors.

22. A method of differentially-processing GNSS data using GNSS reference data prepared in accordance with claim 1, further comprising:

applying a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a backward estimate of position and ambiguities for the GNSS receiver, and applying a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a forward estimate of position and ambiguities for the GNSS receiver.

23. The method of claim 22, wherein the backward estimate and the forward estimate are prepared with GNSS receiver data collected over a period substantially longer than a feature-occupation time.

24. The method of claim 22, further comprising combining the backward estimate and the forward estimate to produced a smoothed position estimate.

25. A method of differentially-processing GNSS data using pre-processed reference data prepared in accordance with claim 9, further comprising:

applying a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a backward estimate of position and ambiguities for the GNSS receiver, and applying a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a forward estimate of position and ambiguities for the GNSS receiver.

26. The method of 25, wherein the backward estimate and the forward estimate are prepared with GNSS receiver data collected over a period substantially longer than a feature-occupation time.

27. The method of 25, further comprising combining the backward estimate and the forward estimate to produced a smoothed position estimate.

28. Apparatus for preparing GNSS reference data for processing data acquired by a GNSS receiver at times corresponding to first intervals, comprising:

An element to access reference data having measurements taken at times corresponding to second intervals longer than the first intervals,

An element to determine a geometric component for each measurement,

An element to produce for each measurement a geometry-free residual by subtracting the geometric component from the measurement,

An element to interpolate the geometry-free residuals to produce interpolated residuals for times corresponding to the first intervals,

An element to determine a geometric component for each time corresponding to an interpolated residual,

An element to combine, for each time corresponding to the first intervals, the interpolated residual for that time with the geometric component for that time to produce an interpolated reference measurement for that time.

29. The apparatus of claim 28, wherein the element to interpolate the geometry-free residual for given time comprises an element to prepare a linear interpolation over a second interval spanning the given time.

30. The apparatus of claim 29, wherein the element to prepare a linear interpolation comprises an element to determine phase change rate of the residuals.

31. The apparatus of claim 29, wherein the measurements are taken from GNSS signals of at least two carrier frequencies, and wherein the element to prepare a linear interpolation comprises an element to determine a phase-change rate and an element to determine an interpolated code residual using the phase-change rate.

32. The apparatus of claim 28, wherein the element to interpolate the geometry-free residual for a given time comprises a linear Kalman filter running over period including the given time.

33. The apparatus of claim 28, wherein the reference data comprises carrier-phase measurements, and wherein the interpolated reference measurements comprise interpolated carrier-phase measurements.

34. The apparatus of claim 28, wherein the reference data comprises code measurements, and wherein the interpolated reference measurements comprise code measurements.

35. The apparatus of claim 33, wherein the reference data comprises code measurements, and wherein the interpolated reference measurements comprise code measurements.

36. The apparatus of claim 28, further adapted to process data acquired by a GNSS receiver during an acquisition period and comprising:

An element to access GNSS reference data acquired during a reference period substantially longer than the acquisition period, the GNSS reference data comprising code measurements and carrier-phase measurements,

An element to prepare a geometry-free and ionosphere-free code-minus-carrier observable from GNSS reference data,

An element to average the code-minus-carrier observable over the reference period to obtain respective mean carrier-ambiguity values for each of a plurality of carrier frequencies, and

An element to produce a set of pre-processed reference data by combining the mean carrier-ambiguity values with interpolated reference measurements for multiple times during the acquisition period.

37. The apparatus of claim 36, wherein the interpolated reference measurements are interpolated carrier-phase measurements.

38. The apparatus of claim 37, wherein the acquisition period ranges from one minute to several minutes and wherein the reference period ranges from one hour to several hours.

39. The apparatus of claim 36, further comprising an element to prepare single-differenced minimum-error combination observations between a set of data acquired by a GNSS rover and the pre-processed reference data

40. The apparatus of claim method of claim 39, further comprising a Kalman filter having states for receiver position, receiver clock and ambiguities to filter the single-difference minimum-error combination observations.

41. The apparatus of claim 39, wherein the minimum-error combination observations comprise L-1 phase and an L1-phase plus L1-code combination.

42. The apparatus of claim 40, wherein the Kalman filter has two rover-clock states and is adapted to alternate between the two rover-clock states when processing successive epochs of data.

43. The apparatus of claim 41, wherein the Kalman filter has two rover-position states and alternates between the rover-position states when processing successive epochs of data.

44. The apparatus of claim 42, wherein the Kalman filter has two rover-position states and is adapted to determine from the rover position states whether the receiver has changed position by more than a predetermined amount from a prior epoch to a current epoch

45. The apparatus of claim 44, further comprising an element to estimate change of rover position between a prior epoch and a current epoch from phase measurement differences between the prior epoch and the current epoch.

46. The apparatus of claim 45, wherein estimating change of rover position comprises applying a decorrelated delta-phase observable.

47. The apparatus of claim 28, further comprising an element to apply a whitening of noise technique to reduce non-Gaussian positioning errors.

48. The apparatus of claim 36, further comprising an element to apply a whitening of noise technique to reduce non-Gaussian positioning errors.

49. The apparatus of claim 28, further adapted to differentially-process GNSS data using the interpolated reference measurements and comprising:

a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a backward estimate of position and ambiguities for the GNSS receiver, and a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a forward estimate of position and ambiguities for the GNSS receiver.

50. The apparatus of claim 49, wherein the backward estimate and the forward estimate are prepared with GNSS receiver data collected over a period substantially longer than a feature-occupation time.

51. The apparatus of claim 49, further comprising an element to combine the backward estimate and the forward estimate to produced a smoothed position estimate.

52. The apparatus of claim 36, further adapted to differentially-process GNSS data using the pre-processed reference data and comprising:

a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a backward estimate of position and ambiguities for the GNSS receiver, and a Kalman filter having position states for current epoch and for previous epoch and an ambiguity state for each satellite to prepare a forward estimate of position and ambiguities for the GNSS receiver.

53. The apparatus of claim 52, wherein the backward estimate and the forward estimate are prepared with GNSS receiver data collected over a period substantially longer than a feature-occupation time.

54. The apparatus of claim 53, further comprising an element to combine the backward estimate and the forward estimate to produced a smoothed position estimate.