Method and apparatus for classifying multiple device states

• US 9,195,309 B2
• Filed: 09/23/2011
• Issued: 11/24/2015
• Est. Priority Date: 05/27/2011
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Grokop, Leonard H., Sarah, Anthony
• Primary Examiner(s): Eisen, Alexander
• Assistant Examiner(s): Yang, Nan-Ying
• Application Number:
  US13/244,170
• Publication Number:
  US 20120299828A1
• Time in Patent Office: 1,523 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/017 Gesture based interaction, e.g. based on a set of recognized hand gestures interaction based on gestures traced on a digitiser G06F3/04883
• CPC Class Codes: G06F 3/0346 with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors

Abstract

Techniques are described herein for classifying multiple device states using separate Bayesian classifiers. An example of a method described herein includes accessing sensor information of a device, wherein at least some of the sensor information is used in a first feature set and at least some of the sensor information is used in a second feature set; processing the first feature set using a first classification algorithm configured to determine a first proposed state of a first state type and a first proposed state of a second state type; processing the second feature set using a second classification algorithm configured to determine a second proposed state of the first state type and a second proposed state of the second state type; and determining a proposed state of the device as the first proposed state of the first state type and the second proposed state of the second state type.

Claims

1. A method of classifying a plurality of state types for a device, the method comprising:

accessing sensor information from a plurality of sensors of the device, wherein at least some of the sensor information is used in a first set of sensor information and at least some of the sensor information is used in a second set of sensor information;

processing the first set of sensor information using a first classification algorithm, wherein the first classification algorithm is configured to determine a first proposed state of a first state type and a first proposed state of a second state type;

processing the second set of sensor information using a second classification algorithm, wherein the second classification algorithm is configured to determine a second proposed state of the first state type and a second proposed state of the second state type, wherein the first proposed state of the first state type is different than the second proposed state of the first state type; and

after completion of the processing of the first set of sensor information and the processing of the second set of sensor information, determining a proposed state of the device as the first proposed state of the first state type and the second proposed state of the second state type based on the results of the processing of the first set of sensor information and the processing of the second set of sensor information.

2. The method of claim 1 wherein the first state type is a motion state and the second state type is a position state.

3. The method of claim 2 wherein:

the sensor information used in the first set of sensor information includes accelerometer output, and

the first set of sensor information includes a variance of accelerometer output over a time interval.

4. The method of claim 2 wherein:

the sensor information used in the second set of sensor information includes accelerometer output for a plurality of accelerometer axes of a multi-axis accelerometer having at least a first axis and a second axis, and

the second set of sensor information includes a mean ratio of acceleration between two accelerometer axes of the multi-axis accelerometer.

5. The method of claim 1 wherein the first classification algorithm comprises a maximize joint likelihood algorithm using the first set of sensor information and a first probability distribution function.

6. The method of claim 1 wherein the first classification algorithm comprises a maximize marginal posterior algorithm using the first set of sensor information and a first probability distribution function.

7. The method of claim 1 wherein the second classification algorithm comprises a maximize joint likelihood algorithm using the second set of sensor information and a second probability distribution function.

8. The method of claim 1 wherein the second classification algorithm comprises a maximize marginal posterior algorithm using the second set of sensor information and a second probability distribution function.

9. The method of claim 1 wherein the first classification algorithm and the second classification algorithm each comprise a Bayesian classifier.

10. A mobile device comprising:

one or more sensors configured to generate sensor output;

a feature set generator module communicatively coupled to the one or more sensors and configured to associate respective sensor information indicative of the sensor output to at least one of a first set of sensor information and a second set of sensor information; and

a classification module communicatively coupled to the feature set generator module and configured to perform operations comprising;

processing the first set of sensor information using a first classification algorithm, wherein the first classification algorithm is configured to determine a first proposed state of a first state type and a first proposed state of a second state type,

processing the second set of sensor information using a second classification algorithm, wherein the second classification algorithm is configured to determine a second proposed state of the first state type and a second proposed state of the second state type, wherein the first proposed state of the first state type is different than the second proposed state of the first state type, and

determining a proposed state of the device as the first proposed state of the first state type and the second proposed state of the second state type.

11. The device of claim 10 wherein the first state type is a motion state and the second state type is a position state.

12. The device of claim 11 wherein:

the one or more sensors include one or more accelerometers,

the sensor information used in the first set of sensor information includes accelerometer output, and

the first set of sensor information includes a variance of accelerometer output over a time interval.

13. The device of claim 11 wherein:

the one or more sensors include one or more accelerometers,

the sensor information used in the second set of sensor information includes accelerometer output for a plurality of accelerometer axes of a multi-axis accelerometer having at least a first axis and a second axis, and

the second set of sensor information includes a mean ratio of acceleration between two accelerometer axes of the multi-axis accelerometer.

14. The device of claim 10 wherein the first classification algorithm comprises a maximize joint likelihood algorithm using the first set of sensor information and a first probability distribution function.

15. The device of claim 10 wherein the first classification algorithm comprises a maximize marginal posterior algorithm using the first set of sensor information and a first probability distribution function.

16. The device of claim 10 wherein the second classification algorithm comprises a maximize joint likelihood algorithm using the second set of sensor information and a second probability distribution function.

17. The device of claim 10 wherein the second classification algorithm comprises a maximize marginal posterior algorithm using the second set of sensor information and a second probability distribution function.

18. The device of claim 10 wherein the first classification algorithm and the second classification algorithm each comprise a Bayesian classifier.

19. A mobile device comprising:

means for accessing sensor information of the device;

means for associating at least some of the sensor information with a first set of sensor information;

means for associating at least some of the sensor information with a second set of sensor information;

means for processing the first set of sensor information using a first classification algorithm, wherein the first classification algorithm is configured to determine a first proposed state of a first state type and a first proposed state of a second state type;

means for processing the second set of sensor information using a second classification algorithm, wherein the second classification algorithm is configured to determine a second proposed state of the first state type and a second proposed state of the second state type, wherein the first proposed state of the first state type is different than the second proposed state of the first state type; and

means for determining a proposed state of the device as the first proposed state of the first state type and the second proposed state of the second state type.

20. The device of claim 19 wherein the first state type is a motion state and the second state type is a position state.

21. The device of claim 20 wherein:

the sensor information associated with the first set of sensor information includes accelerometer output, and

the first set of sensor information includes a variance of accelerometer output over a time interval.

22. The device of claim 20 wherein:

the sensor information associated with the second set of sensor information includes accelerometer output for a plurality of accelerometer axes of a multi-axis accelerometer having at least a first axis and a second axis, and

the second set of sensor information includes a mean ratio of acceleration between two accelerometer axes of the multi-axis accelerometer.

23. The device of claim 19 wherein:

the first classification algorithm comprises at least one of a maximize joint likelihood algorithm or a maximize marginal posterior algorithm, and

the first classification algorithm uses the first set of sensor information and a first probability distribution function.

24. The device of claim 19 wherein:

the second classification algorithm comprises at least one of a maximize joint likelihood algorithm or a maximize marginal posterior algorithm, and

the second classification algorithm uses the second set of sensor information and a second probability distribution function.

25. A computer program product comprising a non-transitory processor-readable medium and processor-readable instructions residing thereon configured to cause a processor to:

access sensor information of a device, wherein at least some of the sensor information is used in a first feature set and at least some of the sensor information is used in a second feature set;

process the first feature set using a first classification algorithm configured to determine a first proposed state of a first state type and a first proposed state of a second state type;

process the second feature set using a second classification algorithm configured to determine a second proposed state of the first state type and a second proposed state of the second state type, wherein the first proposed state of the first state type is different than the second proposed state of the first state type; and

determine a proposed state of the device as the first proposed state of the first state type and the second proposed state of the second state type.

26. The computer program product of claim 25 wherein the first state type is a motion state and the second state type is a position state.

27. The computer program product of claim 26 wherein:

the sensor information used in the first feature set includes accelerometer output, and

the first feature set includes a variance of accelerometer output over a time interval.

28. The computer program product of claim 26 wherein:

the sensor information used in the second feature set includes accelerometer output for a plurality of accelerometer axes of a multi-axis accelerometer having at least a first axis and a second axis, and

the second feature set includes a mean ratio of acceleration between two accelerometer axes of the multi-axis accelerometer.

29. The computer program product of claim 25 wherein:

the first classification algorithm comprises at least one of a maximize joint likelihood algorithm or a maximize marginal posterior algorithm, and

the first classification algorithm uses the first feature set and a first probability distribution function.

30. The computer program product of claim 25 wherein:

the second classification algorithm comprises at least one of a maximize joint likelihood algorithm or a maximize marginal posterior algorithm, and

the second classification algorithm uses the second set of sensor information and a second probability distribution function.