Granted Patent B2

US 12,405,110 · App. 18/058,421 · Granted Sep 2, 2025

Inventors: Michael Dormody (San Jose, CA); Badrinath Nagarajan (Cupertino, CA); Guiyuan Han (San Jose, CA)

Assignee: NextNav, LLC

G01C5/06G01L27/002

Abstract

Claims (97)

1. A method comprising:

determining if a reference-level pressure value based on one or more measurements of pressure from a network of weather stations should or should not be used to calibrate a pressure sensor of a mobile device by:

determining a period of time since an initial time when the reference-level pressure value was determined or since the one or more measurements of pressure on which the reference-level pressure value is based were measured; and

determining if the period of time is less than a first threshold amount of time, wherein a first determination is made that the reference-level pressure value should be used when the period of time is less than the first threshold amount of time, and a second determination is made that the reference-level pressure value should not be used when the period of time is not less than the first threshold amount of time;

if the first determination is made that the reference-level pressure value should be used, using the reference-level pressure value to calibrate the pressure sensor of the mobile device; and

if the second determination is made that the reference-level pressure value should not be used:

(a) determining a trend in a plurality of reference-level pressure values, wherein the trend indicates a change in the plurality of reference-level pressure values in time;

(b) at a first time, determining a first estimated reference-level pressure value based on the trend; and

(c) using the first estimated reference-level pressure value to calibrate the pressure sensor of the mobile device.

2. The method of claim 1 , wherein the method further comprises:

at a second time, determining a second estimated reference-level pressure value based on the trend, and

using the second estimated reference-level pressure value to calibrate the pressure sensor of the mobile device.

3. The method of claim 1 , wherein determining the trend in the plurality of reference-level pressure values comprises:

receiving the plurality of reference-level pressure values determined using measurements of pressure from the network of weather stations at past instances of time prior to the first time; and

using extrapolation to determine the trend in the plurality of reference-level pressure values.

4. The method of claim 3 , wherein the extrapolation is linear extrapolation, and the plurality of reference-level pressure values comprise two or more reference-level pressure values determined at two or more respective times prior to the first time.

5. The method of claim 3 , wherein the extrapolation is quadratic extrapolation.

6. The method of claim 3 , wherein determining the trend in the plurality of reference-level pressure values comprises:

determining a fit metric associated with the trend in the plurality of reference-level pressure values;

determining if the fit metric meets a threshold fit condition; and

if the fit metric meets the threshold fit condition, using the trend in the plurality of reference-level pressure values as the trend, wherein determining the first estimated reference-level pressure value based on the trend and using the first estimated reference-level pressure value to calibrate the pressure sensor of the mobile device are performed only if the fit metric meets the threshold fit condition.

7. The method of claim 6 , wherein if the fit metric does not meet the threshold fit condition, the method further comprises:

receiving a new reference-level pressure value based on one or more measurements of pressure from the network of weather stations that were measured after the first time;

using the new reference-level pressure value to calibrate the pressure sensor of the mobile device.

8. The method of claim 6 , wherein if the fit metric does not meet the threshold fit condition, the method further comprises:

receiving an additional plurality of reference-level pressure values determined using measurements of pressure from the network of weather stations during past instances of time prior to a second time that is later than the first time;

using extrapolation to determine an additional trend in the additional plurality of reference-level pressure values;

determining an additional fit metric associated with the additional trend in the plurality of reference-level pressure values; and

if the additional fit metric meets the threshold fit condition, determining a second estimated reference-level pressure value based on the additional trend, and using the second estimated reference-level pressure value to calibrate the pressure sensor of the mobile device.

9. The method of claim 1 , wherein the trend is determined using decision tree regression.

10. The method of claim 1 , wherein determining the trend in the plurality of reference-level pressure values comprises:

identifying a set of input variables;

determining values of the input variables for input into a decision tree regression flow that was formulated using historical values of the input variables and an output variable; and

providing the determined values of the input variables to the decision tree regression flow to produce a value of the output variable that represents the trend in the plurality of reference-level pressure values.

11. The method of claim 1 , wherein the period of time is a difference in time between a time stamp of the reference-level pressure value and a later time when the determination is made as to whether the reference-level pressure value should or should not be used to calibrate the pressure sensor of the mobile device.

12. The method of claim 1 , wherein determining a trend in the plurality of reference-level pressure values comprises:

receiving the plurality of reference-level pressure values determined using measurements of pressure from the network of weather stations at past instances of time prior to the initial time when the reference-level pressure value was determined or since the one or more measurements of pressure on which the reference-level pressure value is based were measured;

using linear extrapolation to determine a linearly-extrapolated trend in the plurality of reference-level pressure values;

using quadratic extrapolation to determine a quadratically-extrapolated trend in the plurality of reference-level pressure values;

using the linearly-extrapolated trend to determine a linearly-extrapolated reference-level pressure for the initial time;

using the quadratically-extrapolated trend to determine a quadratically-extrapolated reference-level pressure for the initial time;

determining a first difference between the reference-level pressure value and the linearly-extrapolated reference-level pressure;

determining a second difference between the reference-level pressure value and the quadratically-extrapolated reference-level pressure;

determining a smallest difference from among the first difference and the second difference;

if the first difference is the smallest difference, determining that the trend in the plurality of reference-level pressure values is the linearly-extrapolated trend; and

if the second difference is the smallest difference, determining that the trend in the plurality of reference-level pressure values is the quadratically-extrapolated trend.

13. The method of claim 12 , wherein determining a trend in the plurality of reference-level pressure values comprises:

determining values of input variables for input into a decision tree regression flow that was formulated using historical values of the input variables and an output variable;

producing a value of the output variable that represents a regression-based trend in the plurality of reference-level pressure values for the initial time by providing the determined values of the input variables to the decision tree regression flow;

using the regression-based trend in pressure to determine a regression-based estimated reference-level pressure for the initial time;

determining a third difference between the reference-level pressure value and the regression-based estimated reference-level pressure;

determining the smallest difference from among the first difference, the second difference, and the third difference; and

if the third difference is the smallest difference, determining that the trend in pressure is the regression-based trend in pressure.

14. The method of claim 1 , wherein determining a trend in the plurality of reference-level pressure values comprises:

determining a first linearly-extrapolated trend in pressure using a first set of reference-level pressure values of the plurality of reference-level pressure values;

determining a first quadratically-extrapolated trend in pressure using the first set of reference-level pressure values;

determining a first linearly-extrapolated reference-level pressure for a first historical time using the first linearly-extrapolated trend in pressure;

determining a first quadratically-extrapolated reference-level pressure for the first historical time using the first quadratically-extrapolated trend in pressure;

determining a first historical reference-level pressure value for the first historical time that is based on a first set of one or more measurements of pressure from the network of weather stations;

determining a difference between the first historical reference-level pressure value and the first linearly-extrapolated reference-level pressure;

determining a difference between the first historical reference-level pressure value and the first quadratically-extrapolated reference-level pressure;

determining a second linearly-extrapolated trend in pressure using a second set of reference-level pressure values of the plurality of reference-level pressure values;

determining a second quadratically-extrapolated trend in pressure using the second set of reference-level pressure values;

determining a second linearly-extrapolated reference-level pressure for a second historical time using the second linearly-extrapolated trend in pressure;

determining a second quadratically-extrapolated reference-level pressure for the second historical time using the second quadratically-extrapolated trend in pressure;

determining a second historical reference-level pressure value for the second historical time that is based on a second set of one or more measurements of pressure from the network of weather stations;

determining a difference between the second historical reference-level pressure value and the second linearly-extrapolated reference-level pressure;

determining a difference between the second historical reference-level pressure value and the second quadratically-extrapolated reference-level pressure;

determining a first difference by combining at least (i) the difference between the first historical reference-level pressure value and the first linearly-extrapolated reference-level pressure and (ii) the difference between the second historical reference-level pressure value and the second linearly-extrapolated reference-level pressure;

determining a second difference by combining at least (i) the difference between the first historical reference-level pressure value and the first quadratically-extrapolated reference-level pressure and (ii) the difference between the second historical reference-level pressure value and the second quadratically-extrapolated reference-level pressure;

determining a smallest difference from among the first difference and the second difference;

if the first difference is the smallest difference:

(a) using linear extrapolation to determine a linearly extrapolated trend in a plurality of reference-level pressure values that were determined using measurements of pressure from a network of weather stations at past instances of time prior to the first time; and

(b) using the linearly extrapolated trend in the plurality of reference-level pressure values as the trend in the plurality of reference-level pressure values; and

if the second difference is the smallest difference:

(c) using quadratic extrapolation to determine a quadratically extrapolated trend in the plurality of reference-level pressure values, and

(d) using the quadratically extrapolated trend in the plurality of reference-level pressure values as the trend in the plurality of reference-level pressure values.

15. The method of claim 14 , wherein determining a trend in the plurality of reference-level pressure values comprises:

determining a first set of values of input variables for input into a decision tree regression flow that was formulated using historical values of the input variables and an output variable, and producing a first value of the output variable that represents a first regression-based trend in pressure for the first historical time by providing the first set of values of the input variables to the decision tree regression flow;

determining a first regression-based reference-level pressure for the first historical time using the first regression-based trend in pressure;

determining a difference between the first historical reference-level pressure value and the first regression-based reference-level pressure;

determining a second set of values of the input variables for input into the decision tree regression flow, and producing a second value of the output variable that represents a second regression-based trend in pressure for the second historical time by providing the second set of values of the input variables to the decision tree regression flow;

determining a second regression-based reference-level pressure for the second historical time using the second regression-based trend in pressure;

determining a difference between the second historical reference-level pressure value and the second regression-based reference-level pressure;

determining a third difference by combining at least (i) the difference between the first historical reference-level pressure value and the first regression-based reference-level pressure and (ii) the difference between the second historical reference-level pressure value and the second regression-based reference-level pressure;

determining the smallest difference from among the first difference, the second difference, and the third difference; and

if the third difference is the smallest difference, using the decision tree regression flow to determine the trend in the plurality of reference-level pressure values.

16. The method of claim 1 , wherein the method further comprises:

determining if the period of time is less than a second threshold amount of time;

if the period of time is less than the second threshold amount of time, making the first determination that the reference-level pressure value should be used;

if the period of time is not less than the second threshold amount of time, determining if a pressure variation of reference-level pressure values of a time period does not exceed a pressure variation threshold;

if the pressure variation does not exceed the pressure variation threshold, making the first determination that the reference-level pressure value should be used;

if the pressure variation exceeds the pressure variation threshold, making the second determination that the reference-level pressure value should not be used, and determining if different pressure variations during the time period meet a pressure variation condition;

if the different pressure variations do not meet the pressure variation condition, using linear extrapolation or decision tree regression to determine the trend in the plurality of reference-level pressure values; and

if the different pressure variations meet the pressure variation condition, using linear extrapolation, quadratic extrapolation or decision tree regression to determine the trend in the plurality of reference-level pressure values.

17. One or more non-transitory machine-readable media embodying program instructions that, when executed by one or more machines, cause the one or more machines to implement the method of claim 1 .

18. A system for pressure-based calibration of a pressure sensor, the system comprising: a memory comprising computer-executable instructions; and a machine configured to execute the computer-executable instructions and cause a processing system to perform the method of claim 1 .

Assignments (2)

Continuity (3)

References Cited (32)