Machine learning motion sensing with auxiliary sensors

1. A system comprising one or more computers and one or more storage devices on which are stored instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

receiving, from a passive infrared (PIR) sensor at the property, PIR data that represents motion within an area of the property, wherein the motion triggered a detection event when the motion satisfies a first set of motion detection criteria;

receiving, from an auxiliary sensor at the property, auxiliary sensor data that depicts one or more objects within the area of the property;

classifying, using the auxiliary sensor data, a first object of the one or more objects as an object of non-interest;

evaluating the first set of motion detection criteria, including determining, using the PIR data and the auxiliary sensor data, that the motion that triggered the detection event was motion of the object of non-interest; and

in response to determining that the motion that triggered the detection event was motion of the object of non-interest, adjusting at least one parameter of the first set of motion detection criteria to generate a second set of motion detection criteria.

2. The system of claim 1 , wherein evaluating the first set of motion detection criteria comprises:

classifying, based on the auxiliary sensor data, an object of interest in the area of the property; and

determining, based on the PIR data, that the triggered detection event does not correspond to the object of interest.

3. The system of claim 2 , wherein classifying the object of interest in the area of the property comprises: classifying, using a machine learning algorithm trained to identify objects of interest, an object represented by the auxiliary sensor data as an object of interest.

4. The system of claim 1 , wherein classifying the first object of the one or more objects as an object of non-interest comprises:

determining, using stored settings, that the first object is not classified as an object of interest.

5. The system of claim 1 , wherein classifying the first object of the one or more objects as an object of non-interest comprises:

classifying, using a machine learning algorithm trained to identify objects of non-interest, the first object as an object of non-interest.

6. The system of claim 1 , wherein evaluating the first set of motion detection criteria comprises:

determining, based on the auxiliary sensor data, an expected time of motion detection of an object of interest;

determining, based on the PIR data, a time of the triggered detection event; and

determining that the time of the triggered detection event was later than the expected time of motion detection of the object of interest.

7. The system of claim 1 , wherein determining the second set of motion detection criteria comprises processing the PIR data and the auxiliary sensor data using a machine learning algorithm trained to generate motion detection criteria.

8. The system of claim 1 , wherein the monitoring server is configured to:

obtain environmental data indicating a particular environmental condition at the property; and

determine the second set of motion detection criteria based on the environmental data, the second set of motion detection criteria being designated for use when an environmental condition at the property matches the particular environmental condition, wherein the particular environmental condition comprises one or more of a lighting level, a temperature, a time of day, a day of year, a season, or a weather condition.

9. The system of claim 1 , wherein determining the second set of motion detection criteria comprises setting a motion detection sensitivity in one or more segments of a field of view of the PIR sensor.

10. The system of claim 1 , wherein determining the second set of motion detection criteria comprises:

analyzing the auxiliary sensor data to generate a model of a scene within a field of view of the PIR sensor; the model comprising two or more spatial segments;

classifying an object within the scene as a background object;

identifying an associated spatial segment where the background object is located in the scene; and

reducing a motion detection sensitivity of the associated spatial segment.

11. The system of claim 1 , wherein the auxiliary sensor comprises one or more of an infrared camera or a visible light camera, and the auxiliary sensor data comprises one or more images of the area of the property.

12. The system of claim 1 , wherein the auxiliary sensor comprises one or more of a light sensor, a structured light sensor, a time of flight sensor, a RADAR sensor, a Doppler RADAR sensor, a LIDAR sensor, or a microphone.

13. The system of claim 1 , wherein the PIR sensor has a first field of view and the auxiliary sensor has a second field of view, wherein the first field of view overlaps with the second field of view, and

wherein the monitoring server is configured to:

map the auxiliary sensor data from an area of the second field of view to a corresponding area of the first field of view; and

identify, based on the auxiliary sensor data mapped from the area of the second field of view to the corresponding area of the first field of view, an object that caused the triggered detection event.

14. The system of claim 1 , wherein the first set of motion detection criteria and the second set of motion detection criteria each comprise one or more of a threshold PIR differential voltage or a threshold distance from the PIR sensor.

15. The system of claim 1 comprising a plurality of auxiliary sensors, wherein obtaining the auxiliary sensor data from the auxiliary sensor comprises:

receiving data indicating an environmental condition at the property; and

selecting, for activation, one or more of the plurality of auxiliary sensors based on the environmental condition at the property; and

activating the selected one or more auxiliary sensors to cause the selected one or more auxiliary sensors to generate the auxiliary sensor data.

16. The system of claim 1 , wherein the monitoring server is configured to provide the second set of motion detection criteria to the PIR sensor.

17. The system of claim 1 , wherein the PIR sensor is configured to:

obtain the second set of motion detection criteria from the monitoring server; and

detect additional events based on determining that motion within the area of the property satisfies the second set of motion detection criteria.

18. A method, comprising:

receiving, by one or more computers and from a passive infrared (PIR) sensor at a property, PIR data that represents motion within an area of the property, wherein the motion triggered a detection event when the motion satisfies a first set of motion detection criteria;

receiving, by the one or more computers and from an auxiliary sensor at the property, auxiliary sensor data that depicts one or more objects within the area of the property;

classifying, using the auxiliary sensor data, a first object of the one or more objects as an object of non-interest;

evaluating, by the one or more computers, the first set of motion detection criteria, including determining, using the PIR data and the auxiliary sensor data, that the motion that triggered the detection event was motion of the object of non-interest; and

in response to determining that the motion that triggered the detection event was motion of the object of non-interest, adjusting at least one parameter of the first set of motion detection criteria to generate, by the one or more computers, a second set of motion detection criteria.

19. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

receiving, by the one or more computers and from a passive infrared (PIR) sensor at a property, PIR data that represents motion within an area of the property, wherein the motion triggered a detection event when the motion satisfies a first set of motion detection criteria;

receiving, by the one or more computers and from an auxiliary sensor at the property, auxiliary sensor data that depicts one or more objects within the area of the property;

classifying, using the auxiliary sensor data, a first object of the one or more objects as an object of non-interest;

evaluating, by the one or more computers, the first set of motion detection criteria, including determining, using the PIR data and the auxiliary sensor data, that the motion that triggered the detection event was motion of the object of non-interest; and

in response to determining that the motion that triggered the detection event was motion of the object of non-interest, adjusting at least one parameter of the first set of motion detection criteria to generate, by the one or more computers, a second set of motion detection criteria.

20. The system of claim 1 , wherein classifying the first object of the one or more objects as an object of non-interest comprises:

determining, using preferences input by a user, that the first object is not classified as an object of interest.