Google has been embedding new health-sensing features on its devices over the last few years. The latest Pixel Watch 2 not only features numerous cardiac monitoring aspects and stress tracking but also supports fall detection. The company’s smart speaker Nest Hub was one of the first devices to feature remote sleep monitoring.
In a new patent released this week, Google’s researchers showcase how the next- generation Nest Hubs could feature radar-based Fall detection.
The rationale behind embedding fall detection on the Nest hub definitely has some merits.
A smartwatch or a ring that determines if a person has fallen is only useful if the person remembers to use it properly, such as keeping the wearable device charged and worn properly.
The researchers argue that for at least some of the population that is most at risk for falling, such as the elderly population, such proper use may be difficult to sustain.
Related Reading:
- Google’s Research on Earbuds with Health monitoring confirms New Product features
- How to disable unused built-in apps like YouTube Music or Fitbit on Google Pixel Watch
- How to track your sleep with Google Fit
Therefore, it may be desirable to monitor a person at risk for falling without requiring the person to wear any wearable device.
Smartwatches use motion sensors to look for your body’s instinctive reactions to a fall, like how you might move your arms on the way down, and the impact force of a fall. Radar- based monitoring to detect falls is quite different.
The researchers have put forth an interesting proposition in the new patent “Privacy-Preserving Radar Based Fall Monitoring”.
How will Google’s Voice Assistant devices aid in Fall detection?
According to the patent, a smart-home device may transmit radar waves. Based on reflected radar waves, raw waveform data may be created. The raw waveform data may be processed to determine that a fall by a person has occurred. Speech may then be output announcing that the fall has been detected via the speaker of the smart home device.
The home assistant device may comprise a monolithic radar integrated circuit that consists of an antenna array for receiving reflected radar.
The home assistant device will transmit radar waves within the room. Based on the reflected waves, a raw-waveform data is created and interpreted by the home assistant.
Processing the raw waveform data to determine that the fall by the person has occurred may comprise classifying the moving object based on the center-of-mass tracking using a pre-trained machine learning model.
The Patent US20240112559 was published this week. This clearly shows the company’s ambition to embed newer health-sensing features into its popular Nest Hub product line and offer a path for users who may be hesitant to wear a smartwatch device consistently.
