Wearable Gesture Interface Using Spaced MEMS Microphones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hands-free remote control schemes for electronic devices are unreliable due to their reliance on complex machine learning algorithms for detecting and identifying patterns in highly variable user-specific data, leading to inconsistent control and operation without encumbering the user's hands.
Innovation Solution
A wearable human-electronics interface device featuring a band with spaced-apart MEMS microphones and an inertial sensor that detects vibrations and motions to classify tapping gestures, allowing for reliable and consistent control of electronic devices without requiring direct hand interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If complex machine learning algorithms are used to detect and identify patterns in highly variable user-specific data, then the capability to recognize gestures is improved, but the reliability of hands-free remote control deteriorates due to inconsistent control and operation
Solution Approach 1:
The patent divides the gesture detection task into multiple independent sensor channels (multiple microphones and inertial sensors) that each capture specific aspects of the gesture. By segmenting the detection across multiple sensors with known spatial relationships, the system avoids relying on a single complex algorithm while maintaining high detection capability and consistency.
Solution Approach 2:
The patent replaces complex software-based machine learning pattern recognition with a physics-based acoustic model. The system uses the known geometry of the sensor array and acoustic wave propagation principles to directly calculate gesture parameters from sensor signals, substituting mechanical/physical laws for complex computational algorithms and improving reliability.
2Measurement precision
If multiple spaced-apart MEMS microphones and inertial sensors are used to detect vibrations and motions, then the precision of gesture classification is improved, but the device complexity increases
Solution Approach 1:
The patent makes each sensor serve multiple functions: the microphones detect both acoustic vibrations from finger taps and motion-induced vibrations, while the inertial sensors detect both linear acceleration and rotational motion. This multi-functionality allows the system to achieve high measurement precision with a relatively small number of sensors, reducing overall device complexity.
Solution Approach 2:
The patent adds the temporal dimension to gesture detection by analyzing the sequence and timing of signals from multiple sensors. By examining how vibrations propagate through the device over time across multiple sensor channels, the system achieves high classification precision without needing an excessive number of spatial sensors.
3Ease of operation
If hands-free remote control schemes are implemented without encumbering the user's hands, then the ease of operation is improved, but the reliability deteriorates due to unreliable detection and identification of gestures
Solution Approach 1:
The patent enables the device to automatically detect, classify, and interpret gestures without requiring user training or calibration. The system uses its pre-configured sensor array and acoustic model to self-determine gesture parameters directly from raw sensor signals, providing reliable hands-free control without encumbering the user's hands while maintaining high reliability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables reliable and consistent control of electronic devices by distinguishing between different tapping gestures using MEMS microphones and inertial sensors, reducing the need for complex pattern recognition and improving user interaction without hand encumbrance.
Implementation Method 1
a set of sensors (e.g., microelectromechanical systems ("MEMS") microphones) carried by the band and physically spaced apart from one another... detecting, by at least one sensor in the set of sensors, a vibration at the appendage of the user
Implementation Method 2
the at least one inertial sensor, the at least one inertial sensor detects at least one of: a muscle vibration at the appendage of the user... a tendon vibration at the appendage of the user... a bioacoustic vibration at the appendage of the user... and/or a mechanical vibration at the appendage of the user
Data Source
AI summary
Systems, articles, and methods for wearable human-electronics interfaces are described. A wearable human-electronics interface device includes a band that in use is worn on an appendage (e.g., a wrist, arm, finger, or thumb) of a user. The band carries multiple sensors that are responsive to vibrations. The sensors are physically spaced apart from one another on or within the band. The band also carries an on-board processor. The sensors detect vibrations at the appendage of the user when the user performs different finger tapping gestures (i.e., tapping gestures involving different individual fingers or different combinations of fingers) and provide corresponding vibration signals to the processor. The processor classifies the finger tapping gesture(s) based on the vibration signals and an on-board transmitter sends a corresponding signal to control, operate, or interact with a receiving electronic device. The sensors include inertial sensors, digital MEMS microphones, or a combination thereof.