Sensor Fusion for Activity Identification

• US 20120232432A1
• Filed: 05/21/2012
• Published: 09/13/2012
• Est. Priority Date: 08/29/2008
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Arthur Kinsolving, Brian Y. Lee, David Vogel, Mark A. Christensen, Philippe Kahn
• Inventors: Kahn, Philippe, Kinsolving, Arthur, Christensen, Mark Andrew, Lee, Brian Y, Vogel, David
• Granted Patent: US 8,568,310 B2
• Time in Patent Office: Days
• US Class Current: 600/595
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 2562/08 Sensors provided with means for identification, e.g. barcodes or memory chips
• CPC Class Codes: A61B 5/0022 Monitoring a patient using a global network, e.g. telephone networks, internet
• CPC Class Codes: A61B 5/1038 Measuring plantar pressure during gait
• CPC Class Codes: A61B 5/1112 Global tracking of patients, e.g. by using GPS
• CPC Class Codes: A61B 5/1113 Local tracking of patients, e.g. in a hospital or private home
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/4866 Evaluating metabolism A61B5/083 takes precedence
• CPC Class Codes: A61B 5/6898 Portable consumer electronic devices, e.g. music players, telephones, tablet computers
• CPC Class Codes: A61B 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence
• CPC Class Codes: G01S 19/49 whereby the further system is an inertial position system, e.g. loosely-coupled
• CPC Class Codes: G16H 40/67 for remote operation
• CPC Class Codes: Y10S 482/901 having computer circuitry

Abstract

A mobile device monitors accelerations using one or more inertial sensors. A user activity is identified based on the accelerations. A first estimation is made of a user activity statistic associated with the user activity based on the accelerations. Location information is obtained by one or more location based sensors. A second estimation is made of the user activity statistic based on the location information. The user activity statistic is calculated based on the first estimation and the second estimation.

Claims

1. A method of monitoring human activity using multiple sensors comprising:

monitoring accelerations using an inertial sensor;

identifying a user activity based on the accelerations;

making a first estimation of a user activity statistic associated with the user activity based on the accelerations, the user activity statistic being one of;

a distance traveled or a speed of travel;

making a second estimation of the user activity statistic based on location information obtained by a location based sensor; and

calculating the user activity statistic based on the first estimation and the second estimation, wherein the calculated user activity statistic is a result of a weighted combination of the first estimation and the second estimation.

2. The method of claim 1, further comprising:

transmitting the accelerations and the location information to an external computing device; and

receiving an externally calculated user activity statistic.

3. The method of claim 2, wherein the accelerations and the location information are transmitted to the external computing device after the user activity ends.

4. The method of claim 3, wherein the externally calculated user activity statistic includes a reconstruction of an event that started when the user activity was initially detected and ended when the user activity ceased to be detected using the accelerations and the location information.

5. The method of claim 1, wherein the user activity statistic is one of a distance traveled, a route traveled, a speed of travel, a current position or a periodic human motion count.

6. The method of claim 1, wherein calculating the user activity statistic comprises:

determining a first accuracy of the first estimation and a second accuracy of the second estimation;

weighting the first estimation and the second estimation based on the first accuracy and the second accuracy; and

combining the weighted first estimation and the weighted second estimation.

7. The method of claim 1, further comprising:

verifying the user activity based on the location information.

8. The method of claim 1, further comprising:

verifying current location information based on the accelerations and past location information.

9. The method of claim 1, further comprising:

if the location information indicates that a user is at a particular location, automatically performing a preconfigured action.

10. The method of claim 1, wherein the user activity statistic is a distance traveled, the method further comprising:

calibrating a stride length correlation model based on the accelerations and the second estimation, wherein the stride length correlation model is used to make the first estimation.

11. The method of claim 10, wherein the stride length correlation model is calibrated if the first estimation and the second estimation do not match, and if an accuracy of the second estimation is high.

12. A non-transitory computer readable medium including instructions that, when executed by a processing system, cause the processing system to perform a method comprising:

monitoring accelerations using an inertial sensor;

identifying a user activity based on the accelerations;

making a first estimation of a user activity statistic associated with the user activity based on the accelerations;

making a second estimation of the user activity statistic based on location information obtained by a location based sensor; and

calculating the user activity statistic based on the first estimation and the second estimation, wherein the calculated user activity statistic is a result of a weighted combination of the first estimation and the second estimation.

13. The computer readable medium of claim 12, the method further comprising:

transmitting the accelerations and the location information to an external computing device; and

receiving an externally calculated user activity statistic.

14. The computer readable medium of claim 13, wherein the accelerations and the location information are transmitted to the external computing device after the user activity ends.

15. The computer readable medium of claim 14, wherein the externally calculated user activity statistic includes a reconstruction of an event that started when the user activity was initially detected and ended when the user activity ceased to be detected using the accelerations and the location information.

16. The computer readable medium of claim 12, wherein the user activity statistic is one of a distance traveled, a route traveled, a speed of travel, a current position or a periodic human motion count.

17. The computer readable medium of claim 12, wherein calculating the user activity statistic comprises:

determining an first accuracy of the first estimation and a second accuracy of the second estimation;

weighting the first estimation and the second estimation based on the first accuracy and the second accuracy; and

combining the weighted first estimation and the weighted second estimation.

18. The computer readable medium of claim 12, the method further comprising:

verifying the user activity based on the location information.

19. The computer readable medium of claim 12, the method further comprising:

verifying current location information based on the accelerations and past location information.

20. The computer readable medium of claim 12, the method further comprising:

if the location information indicates that a user is at a particular location, automatically performing a preconfigured action.

21. The computer readable medium of claim 12, wherein the user activity statistic is a distance traveled, the method further comprising:

calibrating a stride length correlation model based on the accelerations and the second estimation, wherein the stride length correlation model is used to make the first estimation.

22. The computer readable medium of claim 21, wherein the stride length correlation model is calibrated if the first estimation and the second estimation do not match, and if an accuracy of the second estimation is high.