Methods and systems to determine a context of a device

• US 8,180,583 B1
• Filed: 11/16/2011
• Issued: 05/15/2012
• Est. Priority Date: 11/16/2011
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Miller, James Brooks, Gossweiler, Rich
• Primary Examiner(s): Lau, Tung S
• Application Number:
  US13/297,425
• Time in Patent Office: 181 Days
• Field of Search: 702/62
• US Class Current: 702/62
• CPC Class Codes: G06F 1/3231 Monitoring the presence, absence or movement of users
• CPC Class Codes: G06F 3/011 Arrangements for interaction with the human body, e.g. for user immersion in virtual reality blind teaching G09B21/00
• CPC Class Codes: H04M 1/72454 according to context-related or environment-related conditions
• CPC Class Codes: H04M 2250/10 including a GPS signal receiver
• CPC Class Codes: H04M 2250/12 including a sensor for measuring a physical value, e.g. temperature or motion telephones including GPS signal receivers H04M2250/10
• CPC Class Codes: H04W 52/0254 detecting a user operation or a tactile contact or a motion of the device
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management
• CPC Class Codes: Y02D 30/70 in wireless communication networks

Abstract

Methods and systems for determining that a device is currently within a given context are described. A device may include multiple sensors coupled to the device. A sensor utility value combining a rate of power consumption for a sensor and an accuracy score indicating a degree to which the sensor may provide outputs indicative of the device being currently in the given context may be determined for each sensor. Based on sensor utility values, the multiple sensors coupled to the device may be ranked. A sensor may be identified based on a ranking of the sensor. A probability associated with a determination that the device is currently within the given context may be assigned to the sensor. More sensors may be identified to determine with increasing probability and increasing power requirements that the device is currently within the given context.

Claims

1. A method comprising:

determining, for each sensor of a subset of sensors coupled to a device, a respective rate of power consumption for that sensor;

determining, for each sensor of the subset of sensors, a respective accuracy score indicating a degree to which information associated with that sensor is indicative of a given context;

for each sensor of the subset of sensors, determining a sensor utility value for that sensor, each respective sensor utility value being based at least in part on the respective rate of power consumption for that sensor and the respective accuracy score for that sensor;

identifying, based on the sensor utility value for each sensor of the subset of sensors, a given sensor;

receiving information output by the given sensor;

determining, based on the information output by the given sensor, a probability that the device is currently within the given context; and

determining, by a processor, that the device is currently within the given context when the probability exceeds a predetermined threshold.

2. The method of claim 1, wherein the given sensor is a first sensor, the probability is a first probability, and the predetermined threshold is a first predetermined threshold, the method further comprising:

when the first probability does not exceed the first predetermined threshold;

(i) identifying, based on the sensor utility value for each sensor of the subset of sensors, a second sensor, (ii) receiving information output by the second sensor, (iii) determining, based on the information output by the second sensor, a second probability that the device is currently within the given context, (iv) combining the first probability and the second probability, and (v) determining that the device is currently within the given context when the combined probability exceeds a second predetermined threshold.

3. The method of claim 2, wherein the respective rate of power consumption for the first sensor is lower than the respective rate of power consumption for the second sensor.

4. The method of claim 2, wherein the second probability is higher than the first probability.

5. The method of claim 2, wherein combining the first probability and the second probability comprises determining a weighted combination of the first probability and the second probability, and wherein weights associated with the weighted combination of the first probability and the second probability vary in correlation with an input to the device.

6. The method of claim 5, wherein the weights vary in correlation with a configuration of the device, and wherein the configuration of the device includes types and number of sensors coupled to the device.

7. The method of claim 1, wherein the sensor utility value for each sensor includes a weighted combination of the respective rate of power consumption for that sensor and the respective accuracy score for that sensor, and wherein weights associated with the weighted combination of the respective rate of power consumption and the respective accuracy score vary in correlation with an input to the device.

8. The method of claim 1, further comprising:

based on each sensor utility value for each respective sensor, ranking the subset of sensors based on each respective rate of power consumption for each respective sensor.

9. The method of claim 1, wherein each respective accuracy score varies in correlation with an input to the device.

10. A non-transitory computer readable medium having stored thereon instructions executable by a computing device to cause the computing device to perform functions comprising:

determining, for each sensor of a subset of sensors coupled to a device, a respective rate of power consumption for that sensor;

determining, for each sensor of the subset of sensors, a respective accuracy score indicating a degree to which information associated with that sensor is indicative of a given context;

for each sensor of the subset of sensors, determining a sensor utility value for that sensor, each respective sensor utility value being based at least in part on the respective rate of power consumption for that sensor and the respective accuracy score for that sensor;

identifying, based on the sensor utility value for each sensor of the subset of sensors, a given sensor;

receiving information output by the given sensor;

determining, based on the information output by the given sensor, a probability that the device is currently within the given context; and

determining that the device is currently within the given context when the probability exceeds a predetermined threshold.

11. The non-transitory computer readable medium of claim 10, wherein the given sensor is a first sensor, the probability is a first probability, and the predetermined threshold is a first predetermined threshold, and wherein the instructions are further executable by the computing device to cause the computing device to perform functions comprising:

when the first probability does not exceed the first predetermined threshold;

(i)

identifying, based on the sensor utility value for each sensor of the subset of sensors, a second sensor, (ii) receiving information output by the second sensor, (iii) determining, based on the information output by the second sensor, a second probability that the device is currently within the given context, (iv) combining the first probability and the second probability, and (v) determining that the device is currently within the given context when the combined probability exceeds a second predetermined threshold.

12. The non-transitory computer readable medium of claim 10, wherein the instructions are further executable by the computing device to cause the computing device to perform functions comprising determining that the device is currently within a context that describes an area where a user is not making use of the device.

13. The non-transitory computer readable medium of claim 12, wherein the instructions are further executable by the computing device to cause the computing device to perform functions comprising:

based on determining that the device is currently within a context that describes the area where the user is not making use of the device, disabling at least one sensor coupled to the device.

14. The non-transitory computer readable medium of claim 12, wherein the instructions are further executable by the computing device to cause the computing device to perform functions comprising:

based on determining that the device is currently within a context that describes the area where the user is not making use of the device, reducing functionality of the device.

15. The non-transitory computer readable medium of claim 10, wherein the sensor utility value for each sensor includes a weighted combination of the respective rate of power consumption for that sensor and the respective accuracy score for that sensor, and wherein weights associated with the weighted combination of the respective rate of power consumption and the respective accuracy score vary in correlation with an input to the device.

16. A system comprising:

a utility function module configured to receive, for each sensor of a subset of sensors coupled to a device, information associated with a respective rate of power consumption for that sensor, and, for each sensor of the subset of sensor, a respective accuracy score indicating a degree to which information associated with that sensor is indicative of a given context, wherein the utility function module is further configured to determine a sensor utility value for each sensor of the subset of sensors, each respective sensor utility value being based at least in part on the respective rate of power consumption for that sensor and the respective accuracy score for that sensor; and

a probability module in communication with the utility function module configured to identify, based on the sensor utility value for each sensor of the subset of sensors, a given sensor, receive information output by the given sensor, and determine, based on the information output by the given sensor, a probability that the device is currently within the given context, wherein the probability module is configured to determine that the device is currently within the given context when the probability exceeds a predetermined threshold.

17. The system of claim 16, wherein the given sensor is a first sensor, the probability is a first probability, and the predetermined threshold is a first predetermined threshold, and wherein the probability module is further configured to:

when the first probability does not exceed the first predetermined threshold;

(i)

identify, based on the sensor utility value for each sensor of the subset of sensors, a second sensor, (ii) receive information output by the second sensor, (iii) determine, based on the information output by the second sensor, a second probability that the device is currently within the given context, (iv) combine the first probability and the second probability, and (v) determine that the device is currently within the given context when the combined probability exceeds a second predetermined threshold.

18. The system of claim 16, further comprising a temporal statistical inference module, in communication with the utility function module and the probability module, configured to determine the probability associated with the device being in the given context given information output by the given sensor, wherein the probability comprises at least an emission probability including a likelihood of the device being currently within the given context given the information output by the given sensor.

19. The system of claim 18, wherein the temporal statistical inference module is further configured to combine information output by multiple sensors and emission probabilities associated with the multiple sensors to determine with an increased probability that the device is currently within the given context.

20. The system of claim 18, wherein the probability comprises a multiplication of a starting probability and the emission probability, wherein the starting probability comprises an average probability of the device being currently within the given context.

21. The system of claim 18, wherein the probability module determines the emission probability by training a machine learning system, and wherein the training comprises:

positioning the device in the given context;

receiving information output by each sensor of the subset of sensors over time; and

determining a statistical distribution including an average and a variance for the output of each sensor.

22. The system of claim 16, wherein the probability module is configured to disable at least one sensor coupled to the device and to reduce the functionality of the device when the device is currently within a context that describes an area where a user is not making use of the device.

23. The system of claim 16, wherein the probability module is configured to enable at least one sensor coupled to the device and to increase the functionality of the device when the device is currently within a context that describes an area where a user is making use of the device.