Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11409371-B2
Publication Date
2022-08-09
Expiration Date
Abstract
Systems and methods for gesture-based control are described. In some embodiments, a system may include a device configured to be worn at a person's wrist. The wearable device may include a biopotential sensor, a wrist location sensor, and a wireless transmitter. The system may have a first state and a second state. In the first state, the system may be configured to classify first data based on an output from the wrist location sensor to detect a wake word gesture. In the second state, the system may be configured to classify second data to detect a second gesture, the second data being based on both wrist location data and biopotential data. The system may be configured to transition from the first state to the second state based, at least in part, on a detection of the wake word gesture.
Core Innovation
A wrist-worn human-machine interface system uses a wearable device configured to be worn at a wrist of a person, including a biopotential sensor configured to obtain biopotential data indicating a state of the person's hand, a wrist location sensor configured to obtain wrist location data, and a wireless transmitter configured to communicate with a responsive device. The wearable device operates with a semi-responsive first state and a responsive second state.
In the semi-responsive first state, first data based on an output from the wrist location sensor is classified to detect a wake word gesture, while the system does not classify biopotential data obtained from the biopotential sensor. The system transitions from the semi-responsive first state to the responsive second state based, at least in part, on detection of the wake word gesture, where the detection is based on the wrist location data obtained by the wrist location sensor.
In the responsive second state, the system classifies second data to detect a second gesture, where the second data is based on both the wrist location data obtained from the wrist location sensor and the biopotential data obtained from the biopotential sensor. The architecture described includes sensor-to-responsive-device processing with sampling, conditioning, filtering, feature extraction, and ML-based gesture detection, and includes concepts such as motion silencing to ignore biopotential sensing during arm motion and wake/sleep word behavior tied to state switching.
Claims Coverage
The independent claim specifies a two-state wearable gesture interface with wake-word detection using wrist location data only, followed by second-gesture classification using both wrist location data and biopotential data, transitioning between states based on the wake word gesture detection.
Two-state wrist interface with wake-word gating
A wearable device at a wrist includes a biopotential sensor, a wrist location sensor, and a wireless transmitter; the wearable device comprises a semi-responsive first state and a responsive second state where, in the semi-responsive first state, first data based on wrist location sensor output is classified to detect a wake word gesture and the system does not classify biopotential data, and in the responsive second state the system classifies second data based on both wrist location data and biopotential data, transitioning from the semi-responsive first state to the responsive second state based, at least in part, on detection of the wake word gesture based on the wrist location data.
Across the claim set reflected in the provided material, coverage centers on gating gesture classification with a semi-responsive first state that detects a wake word gesture using wrist location data only, then switching to a responsive second state that classifies subsequent gestures using both wrist location data and biopotential data.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?