The Impact of Voice Control and AI on the Smart Home

by Tom Kerber | Jan. 18, 2018

The smart home is a comprised of a collection of products and services that serve the needs of consumers using either a centralized or a distributed control architecture. In a distributed control architecture individual products perform specific jobs for consumers, but those products also benefit from collaboration with other products. A loose collaboration between autonomous products is sufficient for many jobs; however, some tasks may be better served by if they were coordinated as a system. Today, smart speaker adoption is outpacing smart product adoption and smart product adoption is outpacing smart home system adoption, which means that most homes will have a distributed intelligence platform with voice control acting as the primary user interface. These systems will use a distributed control strategy with a voice UI as the only centralized function.

If a smart home is to understand how it can better serve its occupants, it must understand who it is serving, and what it is that they are doing or are about to do so that it can serve their needs for that task. This type of cognitive skill is perhaps the most difficult to execute as it requires understanding intent. For example, when someone sits down on the couch in the evening, how should the home respond? It depends. Do they intend to read a book or take a nap, watch a movie on the television or catch up on social media while watching TV? Understanding that someone is in the home and that they are sitting down in the living room is insufficient to understand their intention. Humans must communicate their intention to other humans, and the same is true for the smart home. Voice control combined with scenes are needed to allow the human to communicate their intentions and have the home response with coordinated action.  Until mind control becomes available, humans will need to provide high level commands if they want the home to take action to serve their unique needs.

There is another category of home control that can be automated on behalf of the homeowner. Just like there are voluntary and involuntary muscles in the human body, some actions require deliberate thought, while others happen in the background without any thought. The same is true for the smart home. Many of the individual products in the home are likely more capable of optimizing their operation than providing complicated user controls and features for end users to figure out. Dishwashers, washing machines, and clothes dryers offer different controls for users, but in many cases sensors, data, and high level commands can be used to optimize controls on behalf of users. Comfort controls are a good example of an system that should be able to run automatically with little to no human interface. Optimization strategies derived from thermodynamic models of the home are much more capable than a human in providing comfort at the lowest cost. Energy management within the home is another task that should require limited user interaction.

Further Reading:

Next: Video/Audio Analytics Add Needed Context to the Smart Home
Previous: CONNECTIONS Summit: Speaker Insights from Bernd Grohmann, Executive VP, EQ-3 AG


    Be the first to leave a comment.

Post a Comment

Have a comment? Login or create an account to start a discussion.