Systems and methods for speed estimation based on stride data using predictive modeling

• US 9,158,740 B2
• Filed: 11/08/2012
• Issued: 10/13/2015
• Est. Priority Date: 11/08/2012
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Inventors: Gyorfi, Julius S.
• Primary Examiner(s): Craig, Dwin M
• Application Number:
  US13/672,356
• Publication Number:
  US 20140129189A1
• Time in Patent Office: 1,069 Days
• Field of Search: 342/461, 703/2, 703/14
• US Class Current: 1/1
• CPC Class Codes: G01C 21/165 combined with non-inertial navigation instruments
• CPC Class Codes: G01C 22/006 Pedometers
• CPC Class Codes: G06F 17/10 Complex mathematical operations function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544

Abstract

Systems and methods for speed estimation based on stride data using predictive modeling relate to portable wireless devices (202) that contain GPS or other location resources to generate position and speed data. The device (202) also incorporates an accelerometer (212) or other motion sensor. When GPS signals are interrupted, an estimation module (216) and associated logic generates a speed estimate based solely on stride frequency data from the accelerometer (212). Unlike conventional estimators, the estimation module (216) operates on a non-deterministic basis, generating an estimated speed using a joint probability distribution derived from previous speed and stride data captured while positioning signals are available. The estimation module (216) can continue to operate when GPS signals are available, and either filter the GPS speed measures for spurious signals, or further train the estimation module (216) with new position data. The estimation module (216) can also incorporate altitude measurements when estimating speed.

Claims

1. A method of processing data, comprising:

receiving a stride frequency measurement at a portable device;

accessing a predictive model based on a nondeterministic association between a previous set of stride frequency data and a previous set of calculated speed data at corresponding time intervals; and

generating an estimated speed value based on the stride frequency measurement and the predictive model.

2. The method of claim 1, wherein the nondeterministic association comprises a joint probability distribution between the previous set of stride frequency data and the previous set of calculated speed data.

3. The method of claim 1, wherein the predictive model comprises a Bayesian model.

4. The method of claim 1, further comprising:

capturing a current set of position data at a current set of time data in the portable device; and

generating a current calculated speed value based on the current set of position data and the current set of time data.

5. The method of claim 4, wherein capturing a current set of position data at a current set of time data comprises at least one of:

receiving a set of global positioning system (GPS) data, a set of Doppler data, or a set of cellular triangulation positioning data.

6. The method of claim 4, wherein capturing a current set of position data at a current set of time data comprises:

receiving a set of Doppler data.

7. The method of claim 6, further comprising:

filtering the current calculated speed value based on the estimated speed value.

8. The method of claim 4, further comprising:

filtering the current calculated speed value based on the estimated speed value.

9. The method of claim 4, further comprising:

recording the stride frequency measurement;

updating the predictive model based on the current calculated speed value and the stride frequency measurement.

10. The method of claim 1, wherein generating an estimated speed value comprises:

automatically generating the estimated speed value based on the stride frequency measurement when a current set of position data is unavailable.

11. The method of claim 1, wherein accessing a predictive model comprises:

accessing a predictive model associated exclusively with an individual user of the portable device.

12. The method of claim 1, wherein the predictive model is also based on a previous set of slope data.

13. The method of claim 12, wherein the estimated speed value is generated based on the stride frequency measurement, a current slope value, and the predictive model.

14. A device, comprising:

an accelerometer module; and

a processor, communicating with the accelerometer module, the processor being configured to

receive a stride frequency measurement from the accelerometer module,

access a predictive model based on a nondeterministic association between a previous set of stride frequency data and a previous set of calculated speed data at corresponding time intervals, and

generate an estimated speed value based on the stride frequency measurement and the predictive model.

15. The device of claim 14, further comprising:

a position module; and

wherein the processor is further configured to

receive a current set of position data from the position module, and

generate a current calculated speed value based on the current set of position data.

16. The device of claim 15, wherein the position module comprises:

a global positioning system (GPS) module or a cellular triangulation positioning module.

17. The device of claim 15, wherein the processor is further configured to update the predictive model based on the current calculated speed value and the stride frequency measurement.

18. The device of claim 14, further comprising:

an altimeter module coupled to the processor,

wherein the processor is configured to generate a slope measurement and the predictive model is also based on a previous set of calculated slope data.

19. The device of claim 18, wherein the estimated speed value is generated based on the stride frequency measurement, the slope measurement, and the predictive model.

20. The device of claim 14, further comprising:

a position module; and

wherein the processor is further configured to generate a current calculated speed value based on received Doppler data.

21. The device of claim 14, further comprising:

a display, coupled to the processor, configured to annunciate the estimated speed value.