Tracking activity, velocity, and heading using sensors in mobile devices or other systems

• US 8,788,193 B2
• Filed: 10/16/2012
• Issued: 07/22/2014
• Est. Priority Date: 10/17/2011
• Status: Active Grant

• Current Assignee: Gen9, Inc. (Ginkgo Bioworks Holdings, Inc.)
• Original Assignee: Gen9, Inc. (Ginkgo Bioworks Holdings, Inc.)
• Inventors: Fauci, Mark A., Ma, Yinghao, Poppe, Craig, Hee, Thung-Han, Agrawal, Rahul
• Primary Examiner(s): Cheung, Mary
• Assistant Examiner(s): Sweeney, Brian P
• Application Number:
  US13/653,009
• Publication Number:
  US 20130096817A1
• Time in Patent Office: 644 Days
• Field of Search: 701/411, 701/410
• US Class Current: 701/411
• CPC Class Codes: H04W 4/027 using movement velocity, acceleration information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services

Abstract

Embodiments of the invention provide systems and methods for tracking a mobile device using sensors within the device and without using external signals to determine location, velocity, or heading. According to one embodiment, locating and tracking a mobile device can comprise reading sensor data from each of a plurality of sensors within the mobile device. The sensors can provide the sensor data without relying on a signal from a device external to the mobile device. For example, the plurality of sensors can comprise a compass, a gyroscope, and an accelerometer. Motion, velocity, and heading of the mobile device can be determined based on the sensor data.

Claims

1. A method for locating and tracking a mobile device, the method comprising:

reading sensor data from each of a plurality of sensors within the mobile device, the sensors providing the sensor data without relying on a signal from any device external to the mobile device; and

determining motion, velocity, and heading of the mobile device through a machine learning process based on the sensor data without relying on a signal from any device external to the mobile device, wherein the machine learning process comprises;

collecting sensor data and noise from each of the plurality of sensors over a time period,

fusing the collected sensor data and noise to produce estimates for each of a plurality of indications related to orientation of the mobile device,

normalizing each of the estimates,

forming a plurality of vectors representing activity of the mobile device based on the normalized estimates,

classifying the activity of the mobile device over the time period as motion or static based on the plurality of vectors, and

in response to classifying activity of the mobile device as motion, determining a velocity for the mobile device based on the sensor data and determining a heading for the mobile device based on the sensor data.

2. The method of claim 1, wherein the plurality of sensors comprises a compass, a gyroscope, and an accelerometer.

3. The method of claim 1, wherein determining the velocity for the mobile device comprises:

calculating a mean and a standard deviation for each of a plurality of sensor data readings;

generating a feature vector based on each calculated mean and standard deviation;

comparing the generated feature vector to a reference data set;

assigning each feature vector to a cluster based on said comparing; and

determining the velocity for the mobile device based on the clustered feature vector data.

4. The method of claim 1, wherein determining the heading for the mobile device comprises:

determining a magnitude of a set of angular velocity data;

determining whether the magnitude exceeds a pre-defined threshold; and

in response to determining the magnitude exceeds the pre-defined threshold, considering the mobile device to be rotating, correcting the sensor data for drift over time, and recording the corrected sensor data as a heading for the mobile device.

5. The method of claim 1, further comprising mapping a path of the mobile device, wherein mapping the path comprises:

identifying each of a plurality of axes of a plane in which the mobile device is moving;

determining a first heading vector along the identified axes based on the sensor data;

determining a second heading vector along the identified axes based on the sensor data;

filtering the determined heading vectors; and

drawing a map of the path of the mobile device on the plane based on the filtered heading vectors.

6. The method of claim 5, wherein filtering the determined heading vectors comprises:

storing the first heading vector as a third heading vector when a difference between the first heading vector and the second heading vector is determined to be large;

copying the third heading vector to a fourth heading vector when a heading of a previous point is consistent with a heading of a current point;

storing the first heading vector as a fifth heading vector when there is a large consistent difference between the first heading vector and the fourth heading vector;

discarding heading vectors that indicate quick changes; and

filling gaps created by said discarding based on previous heading vectors and next heading vectors of the discarded heading vectors.

7. The method of claim 5, wherein drawing the map of the path of the mobile device comprises:

finding a current line of best fit for information received;

comparing the current line of best fit to previously saved lines;

in response to determining the current line of best fit is close to parallel to a previously saved line, adjusting the current line of best fit to be parallel to the previously saved line;

in response to determining the current line of best fit is close in position to a previously saved line, adjusting the position of the current line of best fit to match the position of the previously saved line;

in response to determining the current line of best fit includes a corner and in response to determining the corner of the current line of best fit matches a corner of a previously saved line, adjusting points of the current line of best fit to match points of the previously saved line.

8. The method of claim 5, further comprising:

detecting aberrations in a magnetic field around the mobile device;

augmenting the map of the path based the detected aberrations, wherein augmenting the map comprises adding one or more magnetic landmarks to the map based on the detected aberrations; and

correcting a position estimate for the mobile device based on the augmented map and a currently detected aberration.

9. A system comprising:

a processor; and

a memory coupled with and readable by the processor and having stored therein a sequence of instructions which, when executed by the processor, causes the processor to locate and track a mobile device by;

reading sensor data from each of a plurality of sensors within the mobile device, the sensors providing the sensor data without relying on a signal from any device external to the mobile device; and

determining motion, velocity, and heading of the mobile device through a machine learning process based on the sensor data without relying on a signal from any device external to the mobile device, wherein the machine learning process comprises;

collecting sensor data and noise from each of the plurality of sensors over a time period,

fusing the collected sensor data and noise to produce estimates for each of a plurality of indications related to orientation of the mobile device,

normalizing each of the estimates,

forming a plurality of vectors representing activity of the mobile device based on the normalized estimates,

classifying the activity of the mobile device over the time period as motion or static based on the plurality of vectors, and

in response to classifying activity of the mobile device as motion, determining a velocity for the mobile device based on the sensor data and determining a heading for the mobile device based on the sensor data.

10. The system of claim 9, wherein the plurality of sensors comprises a compass, a gyroscope, and an accelerometer.

11. The system of claim 9, wherein determining the velocity for the mobile device comprises:

calculating a mean and a standard deviation for each of a plurality of sensor data readings;

generating a feature vector based on each calculated mean and standard deviation;

comparing the generated feature vector to a reference data set;

assigning each feature vector to a cluster based on said comparing; and

determining the velocity for the mobile device based on the clustered feature vector data.

12. The system of claim 9, wherein determining the heading for the mobile device comprises:

determining a magnitude of a set of angular velocity data;

determining whether the magnitude exceeds a pre-defined threshold; and

in response to determining the magnitude exceeds the pre-defined threshold, considering the mobile device to be rotating, correcting the sensor data for drift over time, and recording the corrected sensor data as a heading for the mobile device.

13. The system of claim 9, further comprising mapping a path of the mobile device, wherein mapping the path comprises:

identifying each of a plurality of axes of a plane in which the mobile device is moving;

determining a first heading vector along the identified axes based on the sensor data;

determining a second heading vector along the identified axes based on the sensor data;

filtering the determined heading vectors; and

drawing a map of the path of the mobile device on the plane based on the filtered heading vectors.

14. The system of claim 13, wherein filtering the determined heading vectors comprises:

storing the first heading vector as a third heading vector when a difference between the first heading vector and the second heading vector is determined to be large;

copying the third heading vector to a fourth heading vector when a heading of a previous point is consistent with a heading of a current point;

storing the first heading vector as a fifth heading vector when there is a large consistent difference between the first heading vector and the fourth heading vector;

discarding heading vectors that indicate quick changes; and

filling gaps created by said discarding based on previous heading vectors and next heading vectors of the discarded heading vectors.

15. The system of claim 13, wherein drawing the map of the path of the mobile device comprises:

finding a current line of best fit for information received;

comparing the current line of best fit to previously saved lines;

in response to determining the current line of best fit is close to parallel to a previously saved line, adjusting the current line of best fit to be parallel to the previously saved line;

in response to determining the current line of best fit is close in position to a previously saved line, adjusting the position of the current line of best fit to match the position of the previously saved line;

in response to determining the current line of best fit includes a corner and in response to determining the corner of the current line of best fit matches a corner of a previously saved line, adjusting points of the current line of best fit to match points of the previously saved line.

16. The system of claim 13, further comprising:

detecting aberrations in a magnetic field around the mobile device;

augmenting the map of the path based the detected aberrations, wherein augmenting the map comprises adding one or more magnetic landmarks to the map based on the detected aberrations; and

correcting a position estimate for the mobile device based on the augmented map and a currently detected aberration.

17. A computer-readable memory having stored thereon a sequence of instructions which, when executed by a processor, causes the processor to locate and track a mobile device by:

reading sensor data from each of a plurality of sensors within the mobile device, the sensors providing the sensor data without relying on a signal from any device external to the mobile device; and

determining motion, velocity, and heading of the mobile device through a machine learning process based on the sensor data without relying on a signal from any device external to the mobile device, wherein the machine learning process comprises;

collecting sensor data and noise from each of the plurality of sensors over a time period,

fusing the collected sensor data and noise to produce estimates for each of a plurality of indications related to orientation of the mobile device,

normalizing each of the estimates,

forming a plurality of vectors representing activity of the mobile device based on the normalized estimates,

classifying the activity of the mobile device over the time period as motion or static based on the plurality of vectors, and

in response to classifying activity of the mobile device as motion, determining a velocity for the mobile device based on the sensor data and determining a heading for the mobile device based on the sensor data.

18. The computer-readable memory of claim 17, wherein the plurality of sensors comprises a compass, a gyroscope, and an accelerometer.

19. The computer-readable memory of claim 17, wherein

determining the velocity for the mobile device comprises:

calculating a mean and a standard deviation for each of a plurality of sensor data readings;

generating a feature vector based on each calculated mean and standard deviation;

comparing the generated feature vector to a reference data set;

assigning each feature vector to a cluster based on said comparing; and

determining the velocity for the mobile device based on the clustered feature vector data.

20. The computer-readable memory of claim 17, wherein

determining the heading for the mobile device comprises:

determining a magnitude of a set of angular velocity data;

determining whether the magnitude exceeds a pre-defined threshold; and

in response to determining the magnitude exceeds the pre-defined threshold, considering the mobile device to be rotating, correcting the sensor data for drift over time, and recording the corrected sensor data as a heading for the mobile device.

21. The computer-readable memory of claim 17, further comprising mapping a path of the mobile device, wherein mapping the path comprises:

identifying each of a plurality of axes of a plane in which the mobile device is moving;

determining a first heading vector along the identified axes based on the sensor data;

determining a second heading vector along the identified axes based on the sensor data;

filtering the determined heading vectors; and

drawing a map of the path of the mobile device on the plane based on the filtered heading vectors.

22. The computer-readable memory of claim 21, wherein filtering the determined heading vectors comprises:

storing the first heading vector as a third heading vector when a difference between the first heading vector and the second heading vector is determined to be large;

copying the third heading vector to a fourth heading vector when a heading of a previous point is consistent with a heading of a current point;

storing the first heading vector as a fifth heading vector when there is a large consistent difference between the first heading vector and the fourth heading vector;

discarding heading vectors that indicate quick changes; and

filling gaps created by said discarding based on previous heading vectors and next heading vectors of the discarded heading vectors.

23. The computer-readable memory of claim 21, wherein drawing the map of the path of the mobile device comprises:

finding a current line of best fit for information received;

comparing the current line of best fit to previously saved lines;

in response to determining the current line of best fit is close to parallel to a previously saved line, adjusting the current line of best fit to be parallel to the previously saved line;

in response to determining the current line of best fit is close in position to a previously saved line, adjusting the position of the current line of best fit to match the position of the previously saved line;

in response to determining the current line of best fit includes a corner and in response to determining the corner of the current line of best fit matches a corner of a previously saved line, adjusting points of the current line of best fit to match points of the previously saved line.

24. The computer-readable memory of claim 21, further comprising:

detecting aberrations in a magnetic field around the mobile device;

augmenting the map of the path based the detected aberrations, wherein augmenting the map comprises adding one or more magnetic landmarks to the map based on the detected aberrations; and

correcting a position estimate for the mobile device based on the augmented map and a currently detected aberration.