Home automation with features to improve sleep

1 . A system comprising:

a bed having a mattress and a foundation configured to support the mattress;

a sensor configured to sense sleep readings of a user sleeping on the bed;

an automation controller configured to transmit the sleep readings sensed by the sensor to a cloud service, wherein the cloud service comprises a machine learning model, wherein the machine learning model comprises at least two sub-models, and wherein the machine learning model comprises one or more classifier machine learning models trained using a supervised training approach, wherein the supervised training approach comprises:

obtaining a set of training data comprising a plurality of training examples, each training example comprising (i) training sleep readings for a user of a bed and (ii) a target output for the training sleep readings;

for each training example of the plurality of training examples:

generating, based on the training sleep readings, one or more features comprising trend data;

processing the training sleep readings and trend data using the machine learning model to generate a training sleep inference for the user;

generating, using the machine learning model and the training sleep inference, training adjustments to the sleep environment of the user; and

training the machine learning model to optimize an objective function based on a relationship between the training adjustments to the sleep environment of the user and the target output for the training sleep readings;

hardware hosting the cloud service, the cloud service being configured to:

process the sleep readings and trend data using a first sub-model of the machine learning model to generate a sleep inference for the user;

generate, using a second sub-model of the machine learning model and the sleep inference, adjustments to a sleep environment of the user; and

provide information to the user via a graphic user interface (GUI) based on the adjustments.

2 . The system of claim 1 , wherein the classifier machine learning models have been trained using an unsupervised training approach.

3 . The system of claim 2 , wherein the unsupervised training approach is a clustering technique.

4 . The system of claim 3 , wherein the clustering technique comprises (i) a K-means algorithm, (ii) mixture models, (iii) hierarchical clustering, (iv) self organizing maps, (v) hidden Markov models, or any combination thereof.

5 . The system of claim 4 , wherein the K-means algorithm comprises minimizing a distortion score.

6 . The system of claim 1 , wherein the machine learning model comprises (i) one or more support vector machine models (ii) one or more artificial neural networks (iii) one or more decision tree models, (iv) one or more regression models, or any combination thereof.

7 . The system of claim 1 , wherein obtaining the set of training data comprises annotating a large set of data.

8 . The system of claim 1 , wherein the set of training data comprises pre-classified data.

9 . The system of claim 1 , wherein the machine learning model comprises one or more regressor machine learning models.

10 . The system of claim 1 , wherein the machine learning model comprises:

one or more classifier machine learning models; and

one or more regressor machine learning models.

11 . The system of claim 1 , wherein the hardware hosting the cloud service is further configured to:

receive one or more user inputs from the GUI; and

provide interactive personalized insights to the GUI.

12 . The system of claim 1 , wherein the hardware hosting the cloud service is further configured to:

generate, using the machine learning model, instructions to modify the sleep environment of the user based on the adjustments to sleep environment of the user.

13 . The system of claim 1 , wherein the sensor configured to sense sleep readings of a user sleeping on the bed is integral to the bed, and wherein the system further comprises:

an interface device configured to display the GUI;

a datastore storing historic user sleep data; and

a peripheral sensor that is not integral to the bed and that is configured to sense an element of the sleep environment of the user as peripheral data; and

wherein the hardware hosting the cloud service is further configured to incorporate data input, the historical user sleep data, and the peripheral data in the generation of instructions to modify the sleep environment of the user.

14 . The system of claim 13 , wherein the automation controller is one of the group consisting of a bed-controller, a pump-controller, and a voice-activated home-automation controller.

15 . The system of claim 12 , wherein the instructions to modify the sleep environment of the user include machine-readable instructions for one of the group consisting of the bed and a peripheral device.

16 . The system of claim 12 , wherein the instructions to modify the sleep environment of the user include human-readable suggestions to the user.

17 . The system of claim 1 , wherein generating the sleep inference comprises (i) identifying sleep factors that negatively impact sleep quality, (ii) detecting if a user has difficulty falling asleep or difficulty staying asleep, (iii) detecting biometrics and sleep health factor changes, (iv) detecting out-of-the-norm changes in sleep or health parameters, (v) detecting sleep and health disorders, or any combination thereof.