Automatically learning and controlling connected devices

Abstract

A first input is received from a plurality of sensors. A first state including a first location based on the first input is determined. The first state is associated with a first probability. A second input is received from the plurality of sensors. A second state including a second location is determined based on the second input associated with a second probability. It is determined that the second state corresponds to an actual state based on a transition model and the second probability. The transition model associates the first state with the second state and indicates a likelihood of a transition from the first state to the second state. A rule to change a state of at least one network connected device is triggered based on the second state.

What is claimed is: 
     
         1 . A computer system, comprising:
 a processor;   a storage device coupled to the processor and storing instructions, which when executed by the processor cause the processor to perform a method for controlling network connected devices, the method comprising:
 receiving, from a plurality of sensors, sensor data; 
 determining, based on the sensor data, a probability of a subject being in a current state; 
 determining the current state to be an actual state of the subject based on the probability, a previous state of the subject, and a transition model; and 
 controlling at least one network connected device based on the current state. 
   
     
     
         2 . The computer system of  claim 1 , wherein the plurality of sensors includes one or more of:
 a camera;   a microphone;   a motion sensor; and   a temperature sensor.   
     
     
         3 . The computer system of  claim 1 , wherein the subject comprises one or more of:
 a person; and   a pet.   
     
     
         4 . The computer system of  claim 3 , wherein a state of the subject comprises one or more of:
 an identifier of the subject;   a type of the subject;   a location; and   an activity.   
     
     
         5 . The computer system of  claim 1 , wherein controlling the at least one network connected device further comprises determining that a triggering condition for a rule is met based on the current state and modifying a behavior of the at least one network connected device based on the rule, and wherein the triggering condition comprises one or more different states and/or a range of states. 
     
     
         6 . The computer system of  claim 5 , wherein the rule is automatically generated based on observations detected using the plurality of the sensors. 
     
     
         7 . The computer system of  claim 5 , wherein the method further comprises:
 in response to determining that triggering conditions for a plurality of rules are met, selecting a particular rule for triggering based on a comparison of priority values of the plurality of rules.   
     
     
         8 . The computer system of  claim 7 , wherein method further comprises:
 updating the particular rule's priority value based on a feedback received after the particular rule is triggered.   
     
     
         9 . The computer system of  claim 1 , wherein the transition model indicates a likelihood that the subject transitions from the previous state to the current state, and wherein the transition model was generated using deep learning to analyze a history of state transitions. 
     
     
         10 . A computer-executed method for controlling network connected devices, the method comprising:
 receiving, by a computer from a plurality of sensors, sensor data;   determining, based on the sensor data, a probability of a subject being in a current state;   determining the current state to be an actual state of the subject based on the probability, a previous state of the subject, and a transition model; and   controlling at least one network connected device based on the current state.   
     
     
         11 . The method of  claim 10 , wherein the plurality of sensors includes one or more of:
 a camera;   a microphone;   a motion sensor; and   a temperature sensor.   
     
     
         12 . The method of  claim 10 , wherein the subject comprises one or more of: a person and a pet, and wherein a state of the subject comprises one or more of: an identifier of the subject, a type of the subject, a location, and an activity. 
     
     
         13 . The method of  claim 10 , wherein controlling the at least one network connected device further comprises determining that a triggering condition for a rule is met based on the current state and modifying a behavior of the at least one network connected device based on the rule, and wherein the rule is automatically generated based on observations detected using the plurality of the sensors. 
     
     
         14 . The method of  claim 13 , further comprising:
 in response to determining that triggering conditions for a plurality of rules are met, selecting a particular rule for triggering based on a comparison of priority values of the plurality of rules.   
     
     
         15 . The method of  claim 10 , wherein the transition model indicates a likelihood that the subject transitions from the previous state to the current state, and wherein the transition model was generated using deep learning to analyze a history of state transitions. 
     
     
         16 . A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform method for controlling network connected devices, the method comprising:
 receiving, from a plurality of sensors, sensor data;   determining, based on the sensor data, a probability of a subject being in a current state;   determining the current state to be an actual state of the subject based on the probability, a previous state of the subject, and a transition model; and   controlling at least one network connected device based on the current state.   
     
     
         17 . The computer-readable storage medium of  claim 16 , wherein the plurality of sensors includes one or more of:
 a camera;   a microphone;   a motion sensor; and   a temperature sensor.   
     
     
         18 . The computer-readable storage medium of  claim 16 , wherein the subject comprises one or more of: a person and a pet, and wherein a state of the subject comprises one or more of: an identifier of the subject, a type of the subject, a location, and an activity. 
     
     
         19 . The computer-readable storage medium of  claim 16 , wherein controlling the at least one network connected device further comprises determining that a triggering condition for a rule is met based on the current state and modifying a behavior of the at least one network connected device based on the rule, and wherein the rule is automatically generated based on observations detected using the plurality of the sensors. 
     
     
         20 . The computer-readable storage medium of  claim 16 , wherein the transition model indicates a likelihood that the subject transitions from the previous state to the current state, and wherein the transition model was generated using deep learning to analyze a history of state transitions.