Detecting of patterns of activity based on identified presence detection
A unified presence detection and prediction platform that is privacy aware is described. The platform is receives signals from plural sensor devices that are disposed within a premises. The platform produces profiles of entities based on detected characteristics developed from relatively inexpensive and privacy-aware sensors, i.e., non-video and non-audio sensor devices. The platform using these profiles and sensor signals from relatively inexpensive and privacy-aware sensors determines specific identification and produces historical patterns. Also described are techniques that allow users (persons), when authorized, to control remote devices/systems generally without direct interaction with such systems merely by the systems detecting and in instances predicting the specific presence of an identified individual in a location within the premises.
1. A system comprising at least one computing device including at least one processor and at least one memory, the at least one computing device configured to:
receive signals from a plurality of sensor devices;
retrieve, based on the signals, a presence profile comprising one or more parameters describing physical characteristics of a user and one or more movement flow records associated with the user, wherein each of the one or more movement flow records is associated with a corresponding user action and comprises a series of signals from the plurality of sensor devices;
analyze the signals to determine a historical pattern associated with a routine behavior of the user, wherein the historical pattern is a series of repeated actions comprising a user input causing one or more building devices to perform a particular operation at a plurality of first points in time;
store a graph representation of the historical pattern in the presence profile, wherein the graph representation comprises a set of nodes and a set of edges, each node representing a sensor device and including a time at which the sensor device was encountered by the user, and each edge representing a unidirectional temporal indicator of an order in which two sensor devices connected by that edge were encountered by the user;
predict a next action of the user to be the particular operation at a second point in time based on the graph representation of the historical pattern; and
generate one or more signals to operate the one or more building devices to reproduce the particular operation at the second point in time.
2. The system of claim 1 , wherein the at least one computing device is configured to analyze the signals to determine the historical pattern comprising a hierarchy of historical pattern rules, wherein the hierarchy of historical pattern rules comprises a category rule that invokes quantitative rules, wherein the category rule describes a specific sensor of the plurality of sensor devices present in the one or more movement flow records and the quantitative rules describe a specific signal value of the specific sensor.
3. The system of claim 1 , the historical pattern further comprising a profile ID corresponding to the presence profile and a Rule ID.
4. The system of claim 1 , wherein the at least one computing device is configured to analyze the signals by applying a pattern recognition algorithm to determine the historical pattern.
5. The system of claim 1 , wherein the at least one computing device is further configured to:
receive, from the user, conditional preferences that describe settings for the one or more building devices; and
send one or more second signals to configure the one or more building devices according to the conditional preferences.
6. The system of claim 5 , wherein a first one of the one or more building devices is a thermostat, wherein the at least one computing device is further configured to set a temperature set point for the thermostat.
7. The system of claim 1 , wherein the at least one computing device is further configured to:
receive a request to operate in an activity replication learning mode for one or more specific periods of time.
8. The system of claim 1 , wherein the historical pattern further comprises one or more upper level rules associated with one or more lower level rules, wherein the one or more upper level rules describe a time and a location of the user and the one or more lower level rules describe a variation in the series of repeated actions based on a series of movement flow records, the one or more lower level rules comprising a listing of actions performed by the user, with each action in the listing of actions associated with one or more periods of time.
9. A method, comprising:
receiving signals from a plurality of sensor devices;
retrieving, by a computing circuit, from a data store based on the signals, a presence profile comprising one or more parameters describing physical characteristics of a user and one or more movement flow records associated with the user, wherein each of the one or more movement flow records is associated with a corresponding user action and comprises a series of signals from the plurality of sensor devices;
analyzing the signals to determine a historical pattern associated with a routine behavior of the user, wherein the historical pattern is a series of repeated actions comprising a user input causing one or more building devices to perform a particular operation at a plurality of first points in time;
storing a graph representation of the historical pattern in the presence profile, wherein the graph representation comprises a set of nodes and a set of edges, each node representing a sensor device and including a time at which the sensor device was encountered by the user, and each edge representing a unidirectional temporal indicator of an order in which two sensor devices connected by that edge were encountered by the user;
predicting a next action of the user to be the particular operation at a second point in time based on the graph representation of the historical pattern; and
generating one or more signals to operate the one or more building devices to reproduce the particular operation at the second point in time.
10. The method of claim 9 , wherein the analyzing to determine the historical pattern comprises a hierarchy of historical pattern rules, wherein the hierarchy of historical pattern rules comprises a category rule that invokes quantitative rules, wherein the category rule describes a specific sensor of the plurality of sensor devices present in the movement flow records and the quantitative rules describe a specific signal value of the specific sensor.
11. The method of claim 9 , the historical pattern further comprising a profile ID corresponding to the presence profile and a Rule ID.
12. The method of claim 9 , wherein the analyzing comprises applying a pattern recognition algorithm to analyze the signals to determine the historical pattern.
13. The method of claim 9 , further comprising:
receiving, from the user, conditional preferences that describe settings for the one or more building devices; and
sending one or more second signals to configure the one or more building devices according to the conditional preferences.
14. The method of claim 13 , wherein a first one of the one or more building devices is a thermostat, the method further comprising setting a temperature set point for the thermostat.
15. The method of claim 9 , further comprising:
receiving a request to operate in an activity replication learning mode for one or more specific periods of time.
16. The method of claim 9 , wherein the historical pattern further comprises one or more upper level rules associated with one or more lower level rules, wherein the one or more upper level rules describe a time and a location of the user and the one or more lower level rules describe a variation in the series of repeated actions based on a series of movement flow records, wherein the one or more lower level rules comprise a listing of actions performed by the user, with each action in the listing of actions associated with one or more periods of time.
17. A computer program product tangibly stored on a computer readable hardware storage device of a computing device comprising a processor, the computer program product comprising instructions that when executed by the processor cause the processor to:
receive signals from a plurality of sensor devices;
retrieve from a data store, based on the signals, a presence profile comprising one or more parameters describing physical characteristics of a user and one or more movement flow records associated with the user, wherein each of the one or more movement flow records is associated with a corresponding user action and comprises a series of signals from the plurality of sensor devices;
analyze the signals to determine a historical pattern associated with a routine behavior of the user, wherein the historical pattern is a series of repeated actions comprising a user input causing one or more building devices to perform a particular operation at a plurality of first points in time;
store a graph representation of the historical pattern in the presence profile, wherein the graph representation comprises a set of nodes and a set of edges, each node representing a sensor device and including a time at which the sensor device was encountered by the user, and each edge representing a unidirectional temporal indicator of an order in which two sensor devices connected by that edge were encountered by the user;
predict a next action of the user to be the particular operation at a second point in time based on the graph representation of the historical pattern; and
generate one or more signals to operate the one or more building devices to reproduce the particular operation at the second point in time.
18. The computer program product of claim 17 , the historical pattern further comprising a profile ID corresponding to the presence profile and a Rule ID.
19. The computer program product of claim 17 , wherein the instructions when executed by the processor further cause the processor to:
receive, from the user, conditional preferences that describe settings for the one or more building devices; and
send one or more second signals to configure the one or more building devices according to the conditional preferences.
20. The computer program product of claim 17 , wherein a first one of the one or more building devices is a thermostat, wherein the instructions when executed by the processor further cause the processor to set a temperature set point for the thermostat.
21. The system of claim 1 , wherein the at least one computing device is further configured to:
receive a request to operate in a replication mode;
retrieve, in response to the request, current timing information and one or more corresponding actions stored in the historical pattern; and
send one or more control messages or signals to the one or more building devices to replicate the one or more corresponding actions.