Using accelerometer data to determine movement direction

• US 9,086,281 B1
• Filed: 05/21/2013
• Issued: 07/21/2015
• Est. Priority Date: 05/21/2013
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Thiagarajan, Arvind, Wolfe, James Joseph, Balakrishnan, Anand
• Primary Examiner(s): Cheung, Calvin
• Assistant Examiner(s): Martinez Borrero, Luis A
• Application Number:
  US13/898,635
• Time in Patent Office: 791 Days
• Field of Search: 701/502, 701/434, 701/484, 701/500, 701/541, 701/4, 701/25, 701/506, 701/526, 701/468, 702/160, 702/189, 702/141, 702/6, 702/135, 702/19, 702/95, 702/9, 702/92, 702/94, 455/418, 455/3.01, 455/412.1, 455/574
• US Class Current: 1/1
• CPC Class Codes: G01C 21/10 by using measurements of speed or acceleration G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G01C 21/16 by integrating acceleration or speed, i.e. inertial navigation
• CPC Class Codes: G01C 22/006 Pedometers

Abstract

Techniques are described for determining a direction in which a computing device is being transported by a user who is walking, running, or otherwise moving relative to a direction in which the device is being held. Acceleration data may be collected by accelerometers of the device during a time window, and transformed to determine a rotated acceleration along one or more candidate directions. The rotated acceleration is transformed to a frequency domain, and a peak power is determined within the frequency transformed acceleration data for each candidate direction. The peak powers are compared across multiple candidate directions, and the candidate direction having a higher peak power may be designated as a movement direction of the device relative to the device orientation. A gross movement direction may be determined based on magnetometer data, and may be employed for navigation or other applications.

Claims

1. A computer-implemented method, comprising:

accessing acceleration data generated by one or more accelerometers that are configured as one or more hardware resources of a computing device, the acceleration data indicating acceleration of the computing device in a plane while the computing device is being transported;

determining rotated acceleration data based on a rotation of the acceleration data toward each of a plurality of candidate directions within the plane;

transforming at least a portion of the rotated acceleration data to a frequency domain to determine transformed acceleration data for each of the plurality of candidate directions;

determining a peak power of the transformed acceleration data for each of the plurality of candidate directions;

designating at least one candidate direction of the plurality of candidate directions as a translational movement direction of the computing device relative to an orientation of the computing device, the designating of the at least one candidate direction based at least partly on the at least one candidate direction exhibiting a higher peak power compared to the peak power determined for other candidate directions of the plurality of candidate directions;

accessing orientation data indicating a gross orientation of the computing device relative to a geomagnetic field; and

combining the gross orientation with the translational movement direction to determine a gross movement direction of the computing device relative to the geomagnetic field.

2. The method of claim 1, wherein the transforming of at least the portion of the rotated acceleration data to the frequency domain includes performing one or more of a Fast Fourier Transform (FFT) or a Discrete Fourier Transform (DFT).

3. The method of claim 1, further comprising:

designating two directional axes based on the acceleration data indicating a greater magnitude of the acceleration along the two directional axes than along a third directional axis, the acceleration data indicating the acceleration of the computing device along three directional axes; and

determining the plane based on the two directional axes.

4. The method of claim 1, wherein:

the transforming of at least the portion of the rotated acceleration data to the frequency domain is performed in at least one of two orthogonal directions within the plane; and

the determining of the peak power of the transformed acceleration data is for the at least one of two orthogonal directions for each of the plurality of candidate directions.

5. A computing device, comprising:

a memory storing computer-executable instructions; and

a processor in communication with the memory, the processor configured to access the memory and execute the computer-executable instructions to;

access acceleration data generated by one or more accelerometers comprising one or more hardware resources of the computing device, the acceleration data indicating acceleration of the computing device;

determine rotated acceleration data based on a rotation of the acceleration data toward each of a plurality of candidate directions;

transform at least a portion of the rotated acceleration data to a frequency domain to determine transformed acceleration data for each of the plurality of candidate directions;

determine a peak power of the transformed acceleration data for each of the plurality of candidate directions; and

designate at least one candidate direction of the plurality of candidate directions as a movement direction of a translational movement of the computing device relative to an orientation of the computing device, the designating of the at least one candidate direction based at least partly on the at least one candidate direction exhibiting a higher peak power compared to the peak power determined for other candidate directions of the plurality of candidate directions.

6. The computing device of claim 5, wherein:

the acceleration data indicates the acceleration of the computing device within a plane; and

the plurality of candidate directions are separated by an angular distance;

at least two of the plurality of candidate directions are non-perpendicular to each other; and

an arc including the plurality of candidate directions is at least 180 degrees within the plane.

7. The computing device of claim 6, wherein the angular distance is 10 degrees.

8. The computing device of claim 5, wherein:

the acceleration data indicates the acceleration of the computing device within a plane; and

the plurality of candidate directions are determined based on a search pattern to designate the at least one candidate direction within the plane.

9. The computing device of claim 5, wherein the transforming of at least the portion of the rotated acceleration data to the frequency domain includes performing one or more of a Fast Fourier Transform (FFT) or a Discrete Fourier Transform (DFT).

10. The computing device of claim 5, wherein:

the processor is further configured to;

designate two directional axes based on the acceleration data indicating a greater magnitude of the acceleration along the two directional axes than along a third directional axis, the acceleration data indicating the acceleration of the computing device along three directional axes; and

determine a plane based on the two directional axes; and

an arc including the plurality of candidate directions is at least 180 degrees within the plane.

11. The computing device of claim 5, wherein:

the at least one candidate direction includes two candidate directions separated by 180 degrees;

the processor is further configured to access orientation data generated by one or more of a gyroscopic sensor or a magnetometer, the orientation data indicating the orientation of the computing device; and

the designating of the movement direction includes selecting one of the two candidate directions based on the orientation data.

12. The computing device of claim 5, the processor further configured to:

access orientation data indicating a gross orientation of the computing device relative to a geomagnetic field; and

combine the movement direction with the gross orientation of the computing device to determine a gross movement direction of the computing device relative to the geomagnetic field.

13. The computing device of claim 12, wherein the orientation data is generated using a magnetometer of the computing device.

14. The computing device of claim 5, wherein:

the acceleration data is generated by the one or more accelerometers within a time window; and

the movement direction describes a movement of the computing device during the time window.

15. One or more non-transitory computer-readable media storing instructions which, when executed by a processor, instruct the processor to perform actions comprising:

accessing acceleration data indicating acceleration of a computing device, the acceleration data determined at least partly by at least one accelerometer that is a hardware resource;

determining rotated acceleration data, based on a rotation of the acceleration data toward each of a plurality of candidate directions;

transforming at least a portion of the rotated acceleration data to a frequency domain to determine transformed acceleration data for each of the plurality of candidate directions;

determining a peak power of the transformed acceleration data for each of the plurality of candidate directions; and

designating at least one candidate direction of the plurality of candidate directions as a movement direction of a translational movement of the computing device relative to an orientation of the computing device, the designating of the at least one candidate direction based at least partly on the at least one candidate direction exhibiting a higher peak power compared to the peak power determined for other candidate directions of the plurality of candidate directions.

16. The one or more non-transitory computer-readable media of claim 15, wherein the at least one accelerometer is included in the computing device.

17. The one or more non-transitory computer-readable media of claim 15, the actions further comprising:

adjusting the acceleration data to remove a portion of the acceleration caused by a gravitational field.

18. The one or more non-transitory computer-readable media of claim 15, the actions further comprising:

applying a smoothing function to the rotated acceleration data prior to the transforming.

19. The one or more non-transitory computer-readable media of claim 15, the actions further comprising:

accessing orientation data generated by a magnetometer of the computing device, the orientation data indicating a gross orientation of the computing device relative to a geomagnetic field; and

combining the movement direction with the gross orientation of the computing device to determine a gross movement direction of the computing device relative to the geomagnetic field.

20. The one or more non-transitory computer-readable media of claim 15, wherein:

the acceleration data indicates the acceleration of the computing device within a plane;

the plurality of candidate directions are separated by a predetermined angular distance; and

an arc including the plurality of candidate directions is at least 180 degrees within the plane.