DRAWING AN INFERENCE USING MULTIPLE SENSORS

• US 20150317565A1
• Filed: 07/23/2014
• Published: 11/05/2015
• Est. Priority Date: 04/30/2014
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola Mobility LLC (Lenovo Group Ltd.)
• Inventors: Li, Jianguo, Davis, Paul C., Rumpler, James A., You, Di
• Granted Patent: US 9,697,465 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/043 Distributed expert systems; Blackboards
• CPC Class Codes: G06N 5/048 Fuzzy inferencing

Abstract

Disclosed are techniques (300, 500, 600) and apparatuses (102, 700) for drawing an inference using multiple sensors. These techniques and apparatuses enable a computing device to choose (302) a set of sensors that are capable of providing information for an inference, invoke (304) the chosen set of sensors to provide the information, receive (306) the information from at least a subset of the chosen sensors, and draw (308), based on the received information, the inference. In some cases, the set of sensors are chosen such that time or resources of the computing device can be conserved.

Claims

1. On a computing device, a method for drawing an inference of a usage context of the computing device, the method comprising:

choosing (302), by the computing device from a plurality of sensors capable of providing information for the inference, a set of sensors;

invoking (304), by the computing device, the chosen set of sensors to provide information for the inference;

receiving (306), by the computing device from at least a subset of the chosen set of sensors, information for the inference; and

drawing (308), by the computing device, based, at least in part, on the received information, the inference of the usage context of the computing device.

2. The method of claim 1 wherein the inference is selected from a group consisting of:

a measurement, a determination, a Boolean value, a prediction, a geographical location, and an activity determination.

3. The method of claim 1 wherein choosing the set of sensors comprises choosing a single sensor of the plurality of sensors or every sensor in the plurality of sensors.

4. The method of claim 1 wherein choosing a set of sensors is based, at least in part, on an element selected from the group consisting of:

an expected margin of error for information to be provided by a sensor, an expected degree of confidence for information to be provided by a sensor, an expected amount of resources to be consumed by a sensor, an expected overlap of information to be provided by a set of sensors, an expected amount of time used by a sensor to provide information, an expected price for using a sensor, a privacy consideration for using a sensor, an importance of the inference to be drawn, and an acceptability threshold for information for the inference.

5. The method of claim 1 wherein information received from each of the chosen sensors is associated with an error margin or a degree of confidence.

6. The method of claim 1:

wherein the set of chosen sensors comprises a plurality of sensors;

the method further comprising;

drawing, based at least in part on information received from a chosen sensor, the interference with an intermediate degree of confidence or an intermediate margin of error;

comparing the intermediate degree of confidence or the intermediate margin of error with a confidence threshold or an error threshold, respectively; and

if the intermediate degree of confidence of the inference is greater than or equal to the confidence threshold or if the intermediate margin of error of the inference is less than or equal to the error threshold, then cancelling a chosen sensor from which the computing device has not yet received information.

7. The method of claim 6 further comprising:

if the intermediate degree of confidence of the inference is less than the confidence threshold or if the intermediate margin of error is greater than the error threshold, then;

waiting to receive information from at least one other chosen sensor; and

repeating the drawing and comparing steps with the information received from the other chosen sensor until the intermediate degree of confidence is greater than or equal to the confidence threshold, the intermediate margin of error is less than or equal to the error threshold, or until information has been received from all chosen sensors.

8. The method of claim 1:

wherein the set of chosen sensors comprises a single sensor;

the method further comprising;

drawing, based at least in part on information received from the chosen sensor, the interference with an intermediate degree of confidence or an intermediate margin of error; and

comparing the intermediate degree of confidence or the intermediate margin of error with a confidence threshold or an error threshold, respectively.

9. The method of claim 8 further comprising:

if the intermediate degree of confidence of the inference is less than the confidence threshold or the intermediate margin of error is greater than the error margin, then;

choosing, from the plurality of sensors capable of providing information for the inference, another sensor;

invoking the other chosen sensor to provide information for the inference;

receiving, from the other chosen sensor, information for the inference; and

repeating the drawing and comparing steps with the information received from the other chosen sensor until the intermediate degree of confidence is great than or equal to the confidence threshold, the intermediate margin of error is less than or equal to the error threshold, or until information has been received from each sensor of the plurality of sensors.

10. An apparatus (102, 700) comprising:

a plurality of sensors (212, 214), each sensor of the plurality of sensors capable of providing information useful to make an inference of a usage context of a computing device; and

an inference controller (210) implemented at least partially via hardware to;

select (302), from the plurality of sensors, a set of sensors;

cause (304) the selected set of sensors to provide information for the inference;

receive (306), from at least a subset of the selected set of sensors, information for the inference; and

draw (308), based at least in part on the information received from the subset of sensors, the inference of the usage context of the computing device.

11. The apparatus of claim 10 further comprising a data interface to a sensor remote from the apparatus wherein the inference controller selects the set of sensors from the plurality of sensors and the sensor remote from the apparatus.

12. The apparatus of claim 10 wherein the inference is selected from a group consisting of:

a measurement, a determination, a Boolean value, a predication, a geographical location, and an activity determination.

13. The apparatus of claim 10 wherein selecting the set of sensors comprises selecting a single sensor of the plurality of sensors or every sensor in the plurality of sensors.

14. The apparatus of claim 10:

wherein the selected set of sensors comprises a plurality of sensors;

the inference controller further implemented to;

draw, based at least in part on information received from a selected sensor, the interference with an intermediate degree of confidence;

compare the intermediate degree of confidence of the inference with a confidence threshold; and

if the intermediate degree of confidence of the inference is greater than or equal to the confidence threshold, then cancelling a selected sensor from which the inference controller has not yet received information.

15. The apparatus of claim 10 wherein the selected set of sensors comprises at least one sensor of a group consisting of:

an accelerometer, an acoustic sensor, an infrared sensor, a magnetometer, a global positioning system, a short-range radio, and a wireless local-area-network radio.

16. A method for drawing an inference, the method comprising:

receiving (604), from one or more sensors of a computing device, information useful to draw the inference of a usage context, the information having an associated statistical value;

comparing (606), by the computing device, the statistical value associated with the information with a predefined threshold; and

if the statistical value meets or exceeds the predefined threshold, then drawing (608), by the computing device, the inference using the information received from the one or more sensors;

or

if the statistical value does not meet the predefined threshold, then waiting (610) for additional information from another sensor of the computing device.

17. The method of claim 16 further comprising determining, for the information, the associated statistical value.

18. The method of claim 16 wherein the statistical value is a degree of confidence associated with the information or a margin of error associated with the information.

19. The method of claim 16 wherein the information is received from a single sensor of the computing device, the statistical value does not meet the predefined threshold, and the method further comprises invoking the other sensor of the computing device from which to receive the additional information.

20. The method of claim 16 wherein the statistical value meets or exceeds the predefined threshold and the method further comprises cancelling other sensors of the computing device effective to conserve resources of the computing device.