Apple has filed a patent application that describes an advanced machine learning system for home applications using microlocations’ tagged data to determine a user’s position within their home. With modern mobile devices having hundreds of applications, it can be challenging for a user to find and use an application amongst all the available ones. Apple’s invention seeks to improve this by determining a user’s position in their home and identifying an application based on that.
The mobile application can control other devices like smart locks on doors, windows, window blinds, and kitchen appliances using sensor measurements. A sensor position is used as a proxy for physical position, and a cluster corresponds to a group of sensor positions at which measurements have been made. A location is referred to as a microlocation because it refers to a specific area in the user’s home.
The application can automatically generate tagged samples without the user requesting them by measuring signal values at that location and labeling that tagged sample with the location’s name. After the machine learning model has been trained, the application can provide a recommendation on the user interface or automatically open the front door when it predicts that the user is in the driveway.
Additionally, a wireless streaming application can use a semi-supervised machine learning model to predict target devices to project video or audio. After the machine learning model has been trained, the wireless streaming application can provide the living room TV as a recommendation when the model predicts that the user is in the living room.
This patent application by Apple could be a game-changer in determining a user’s position within their home, making it easy to identify an application that is likely to be used in that particular room. It could provide a user with their Apple TV app to turn on the TV when they enter the family room or provide a garage door opener app as they enter the garage.
