SemiWiki 800x100 DAC FSWG
WP_Term Object
(
    [term_id] => 468
    [name] => Mobile
    [slug] => mobile
    [term_group] => 0
    [term_taxonomy_id] => 468
    [taxonomy] => category
    [description] => 
    [parent] => 0
    [count] => 307
    [filter] => raw
    [cat_ID] => 468
    [category_count] => 307
    [category_description] => 
    [cat_name] => Mobile
    [category_nicename] => mobile
    [category_parent] => 0
)

Apple Smart Home Artificial Intelligence Insights from Patents

Apple Smart Home Artificial Intelligence Insights from Patents

 US 20160132030 illustrate a smart home system for automating operation of the smart home devices (e.g., thermostats, lighting devices, household appliances, etc.) based on aggregation of individual user routines.

User mobile devices and smart home devices can incorporate pattern detection logic to identify patterns in the user’s behavior (e.g., going to particular places at particular times or invoking particular operation functions of a smart home device at particular times). A coordinator (e.g., user smartphone) can receive information about detected patterns and analyze the information to detect an aggregate pattern.

Based on the detected aggregate pattern, the coordinator can identify the operational behavior to automate (e.g., turn off the lights when the last user goes to bed) and implement the automated behavior by establishing the automation rule that reflects the detected aggregate pattern.

The automation rule can specify an action to be taken by the smart home device. For example, the automation rule can specify that a porch light is to be turned on if an outside ambient light sensor detects a light level below a threshold or at a specific time each night.

As another example, the automation rule can specify that a heating (or cooling) system is to be turned on to adjust the temperature of the house to a target temperature. The coordinator can analyze the pattern data of the user routines to detect aggregate patterns across the users. Any pattern of behavior of an individual can be inferred by the machine learning algorithm based on inputs indicative of the individual’s location and/or activity at various times.

US20140136451 illustrates an artificial intelligence (AI) system to determine preferential smart home device action associated to a specific user. The AI system associates observed user behavior with output of the machine learning process that is derived from attributes observed at the specific smart home device, aggregated attributes from a number of other smart home devices and prior knowledge. The AI system determines a preferential smart home device action based on results of the associating.

US9303997 illustrates an AI system to predict the future user behavior using the machine learning process based on the user-specific data.

Reference: IoT Smart Home Patents Data 2Q 2016


Comments

There are no comments yet.

You must register or log in to view/post comments.