Providing energy-efficient features using human presence detection

Abstract

Disclosed herein are system, method, and computer program product embodiments for the detection of human presence in an energy efficient manner using a plurality of sensors such as those of a battery-powered device such as a television remote, and a device with a processor, such as a television. Data gathered from an initial television WiFi radio scan, or an initial low-powered detection scan from the television remote, may be analyzed by the processor to determine a potential presence of one or more humans are present proximate to the device. If there is such a potential presence, the device remote can enter a full-powered detection mode to accurately determine the presence or absence of one or more humans, and take further actions.

What is claimed is: 
     
         1 . A computer implemented method, comprising:
 executing, by at least one processor, a collection routine at regular intervals to gather raw data from a plurality of sensors;   receiving, by the at least one processor, results of the collection routine in a form of the raw data from the sensors;   storing, by the at least one processor, the received results in a central data repository;   setting, by the at least one processor, a device remote in a low-power mode;   analyzing, by the at least one processor, the stored sensor data to determine when a television is turned on or off, and when it is turned on,
 using, by the at least one processor, a WiFi radio of the television to initially scan for a potential presence of one or more humans within a predetermined geographical range; and 
 when one or more humans are determined by the at least one processor to be potentially present, executing, by the at least one processor, at least one further action; 
   
       when the television is turned off, executing, by the at least one processor, at least one further action. 
     
     
         2 . The method of  claim 1 , wherein the executing of at least one further action when the television is turned on comprises:
 fully powering up the device remote by the at least one processor so that the device remote transitions from a low-power mode to a full power mode;   performing, by the at least one processor, a full-presence detection using a WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range;   when one or more humans are determined to be present through said analyzing, determining, by the at least one processor, a respective geographic position of the one or more humans.   
     
     
         3 . The method of  claim 1 , wherein the sensors form a high-resolution detection zone in front of the at least one processor, in 2 or 3 dimensions in a shape configurable by placement of the sensors. 
     
     
         4 . The method of  claim 1 , wherein the executing of at least one further action when the television is turned off further comprises:
 periodically sending, by the at least one processor, a low power WiFi beacon from the television to the device remote;   entering, by the at least one processor, a low power detection mode on the device remote;   using, by the at least one processor, a WiFi radio of the device remote to initially scan at a low transmission power for the potential presence of one or more humans within the predetermined geographical range;   when one or more humans are determined by the at least one processor to be potentially present,
 fully powering up the device remote by the at least one processor so that it transitions from a low-power mode to a full power mode; 
 performing, by the at least one processor, a full-presence detection using a WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range;
 when one or more humans are determined to be present through said analyzing, determining, by the at least one processor, a respective geographic position of the one or more humans. 
 
   
     
     
         5 . The method of  claim 2 , wherein the full-presence detection using the WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range further includes:
 transmitting signals to the television and an internet router using the WiFi radio at full power, by the at least one processor;   gathering, by the at least one processor, raw data from the sensors of the device remote;   storing, by the at least one processor, the received raw data from the sensors of the device remote in the central data repository; and   analyzing, by the at least one processor, the stored remote sensor data to determine if one or more humans are present proximate to the at least one processor within the predetermined geographical range.   
     
     
         6 . The method of  claim 5 , wherein the analyzing of the stored remote sensor data to determine if one or more humans are present proximate to the at least one processor within the predetermined geographical range further comprises:
 feeding, by the at least one processor, the raw data as input to a neural network machine learning classifier, the neural network machine learning classifier having an input layer that receives the raw data as a plurality of inputs, and an output layer; and   comparing, by the at least one processor, values of nodes in the output layer to determine a presence or absence of one or more humans proximate to the at least one processor in the predetermined geographical range.   
     
     
         7 . The method of  claim 6 , wherein the analyzing of the stored remote sensor data to determine if one or more humans are proximate to the at least one processor within the predetermined geographical range further comprises:
 comparing, by the at least one processor, the determined presence or absence of humans by neural network machine learning to a known presence or absence of a user through user feedback; and   conducting, on the basis of the comparison through the user feedback, backpropagation of the neural network machine learning classifier by the at least one processor to adjust weights of nodes of the input layer.   
     
     
         8 . A system, comprising:
 a memory; and   at least one processor coupled to the memory and configured to:   execute a collection routine at regular intervals to gather raw data from a plurality of sensors;   receive results of the collection routine in a form of the raw data from the sensors;   store the received results in a central data repository in the memory;   set a device remote in a low-power mode;   analyze the stored sensor data to determine when a television is turned on or off, and when it is turned on,
 use a WiFi radio of the television to initially scan for the potential presence of one or more humans within a predetermined geographical range; and 
 when one or more humans are determined to be potentially present, execute at least one further action: 
   
       when the television is turned off, execute at least one further action. 
     
     
         9 . The system of  claim 8 , wherein to execute the at least one further action when the television is turned on, the at least one processor is further configured to:
 fully power up the device remote such that the device remote transitions from a low-power mode to a full power mode;   perform a full-presence detection using a WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range;   when one or more humans are determined to be present through said analyzing, determine a respective geographic position of the one or more humans.   
     
     
         10 . The system of  claim 8 , wherein the sensors form a high-resolution detection zone in front of the at least one processor, in 2 or 3 dimensions in a shape configurable by placement of the sensors. 
     
     
         11 . The system of  claim 8 , wherein the executing of at least one further action when the television is turned off further comprises:
 periodically send a low power WiFi beacon from the television to the device remote;   enter a low power detection mode on the device remote;   use a WiFi radio of the device remote to initially scan at a low transmission power for the potential presence of one or more humans within the predetermined geographical range;   when one or more humans are determined by the at least one processor to be potentially present,
 fully power up the device remote so that it transitions from a low-power mode to a full power mode; 
 perform a full-presence detection using a WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range;
 when one or more humans are determined to be present through said analyzing, determine a respective geographic position of the one or more humans; 
 
   
     
     
         12 . The system of  claim 9 , wherein the full-presence detection using the WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range further comprises:
 transmit signals to the television and an internet router using the WiFi radio at full power;   gather raw data from the sensors of the device remote;   store the received raw data from the sensors of the device remote in the central data repository; and   analyze the stored remote sensor data to determine if one or more humans are present proximate to the at least one processor within the predetermined geographical range.   
     
     
         13 . The system of  claim 12 , wherein the analyzing of the stored remote sensor data to determine if one or more humans are proximate to the at least one processor within the predetermined geographical range further comprises:
 feed the raw data as input to a neural network machine learning classifier, the neural network machine learning classifier having an input layer that receives the raw data as a plurality of inputs, and an output layer; and   compare values of nodes in the output layer to determine a presence or absence of one or more humans proximate to the at least one processor in the predetermined geographical range.   
     
     
         14 . The system of  claim 13 , wherein the analyzing of the stored remote sensor data to determine if one or more humans are proximate to the at least one processor within the predetermined geographical range further comprises:
 compare the determined presence or absence of humans by neural network machine learning to a known presence or absence of a user through user feedback; and   conduct, on the basis of the comparison through user feedback, backpropagation of the neural network machine learning classifier to adjusts weights of nodes of the input layer.   
     
     
         15 . A non-transitory computer-readable device having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:
 executing a collection routine at regular intervals to gather raw data from a plurality of sensors;   receiving results of the collection routine in a form of the raw data from the sensors;   storing the received results in a central data repository;   setting a device remote in a low-power mode;   analyzing the stored sensor data to determine when a television is turned on or off, and when it is turned on,
 using a WiFi radio of the television to initially scan for the potential presence of one or more humans within a predetermined geographical range; and 
 when one or more humans are determined to be potentially present, executing at least one further action; 
   
       when the television is turned off, executing at least one further action. 
     
     
         16 . The device of  claim 15 , wherein the operation of executing at least one further action when the television is turned on comprises:
 fully powering up the device remote so that the device remote transitions from a low-power mode to a full power mode;   performing a full-presence detection using a WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range;   when one or more humans are determined to be present through said analyzing, determining a respective geographic position of the one or more humans.   
     
     
         17 . The device of  claim 15 , wherein the operation of executing of at least one further action when the television is turned off further comprises:
 periodically sending a low power WiFi beacon from the television to the device remote;   entering a low power detection mode on the device remote;   using a WiFi radio of the device remote to initially scan at a low transmission power for the potential presence of one or more humans within the predetermined geographical range;   when one or more humans are determined to be potentially present,
 fully powering up the device remote so that it transitions from a low-power mode to a full power mode; 
 performing a full-presence detection using a WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range;
 when one or more humans are determined to be present through said analyzing, determining a respective geographic position of the one or more humans. 
 
   
     
     
         18 . The device of  claim 16 , wherein the operation of full-presence detection using the WiFi radio of the device remote to scan at a full transmission power for the presence of one or more humans within the predetermined geographical range further includes:
 transmitting signals to the television and an internet router using the WiFi radio at full power;   gathering raw data from the sensors of the device remote;   storing the received raw data from the sensors of the device remote in the central data repository; and   analyzing the stored remote sensor data to determine if one or more humans are present proximate to the computing device within the predetermined geographical range.   
     
     
         19 . The device of  claim 18 , wherein the operation of analyzing of the stored remote sensor data to determine if one or more humans are present proximate to the computing device within the predetermined geographical range further comprises:
 feeding the raw data as input to a neural network machine learning classifier, the neural network machine learning classifier having an input layer that receives the raw data as a plurality of inputs, and an output layer; and   comparing values of nodes in the output layer to determine a presence or absence of one or more humans proximate to the at least one computing device in the predetermined geographical range.   
     
     
         20 . The device of  claim 19 , wherein the operation of analyzing of the stored remote sensor data to determine if one or more humans are proximate to the at least one computing device within the predetermined geographical range further comprises:
 comparing the determined presence or absence of humans by neural network machine learning to a known presence or absence of a user through user feedback; and   conducting, on the basis of the comparison through user feedback, backpropagation of the neural network machine learning classifier to adjusts weights of nodes of the input layer.