Wearable Computing Device Gesture Recognition Using Accelerometer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wearable computing devices lack an efficient method to recognize intuitive gestures, such as looking up, which is essential for activating devices in a power-saving manner without consuming excessive power or triggering false positives in accelerated reference frames.
Innovation Solution
A wearable computing device that utilizes an accelerometer to determine the level and tilt of the wearer's head, generating a gesture-recognition trigger when the head is tilted up from a level position, employing a low pass filter and finite state machine to accurately detect the 'look up' gesture using single or dual accelerometers, and optionally with gyroscopes or magnetometers to verify the gesture in various reference frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous monitoring of head position is performed to accurately recognize gestures, then gesture recognition accuracy is improved, but power consumption increases
Solution Approach 1:
The system uses periodic sampling of accelerometer data at specific intervals rather than continuous monitoring. The finite state machine transitions between sleep and active states, enabling periodic detection of head level and tilt events while conserving power during idle periods.
Solution Approach 2:
The system performs preliminary detection of head level position before detecting the tilt-up gesture. This two-stage approach (first detect level, then detect tilt from level) prevents false positives and improves accuracy while allowing the system to remain in low-power mode until events are detected.
2Use of energy by moving object
If simple accelerometer-based detection is used to conserve power, then power consumption is reduced, but false positives occur in accelerated reference frames
Solution Approach 1:
The system uses feedback from the finite state machine to verify gesture sequences. It requires detection of head level first, then subsequent tilt-up from that level, creating a feedback loop that confirms the gesture is intentional rather than caused by acceleration events.
Solution Approach 2:
The finite state machine acts as an intermediary between the raw accelerometer data and gesture recognition. It processes the acceleration events through defined states (sleep, level detected, tilt detected) to filter out false positives from accelerated reference frames.
3Reliability
If multiple sensors (gyroscopes, magnetometers) are added to verify gestures in various reference frames, then detection reliability is improved, but device complexity increases
Solution Approach 1:
The accelerometer serves multiple functions: detecting head level position, detecting tilt-up gestures, and providing reference frame verification through the finite state machine logic. This multi-functionality reduces the need for additional specialized sensors.
Solution Approach 2:
The system changes the parameters being monitored by the accelerometer based on the current state. In sleep mode, it monitors for level events; after level detection, it monitors for tilt events. This dynamic parameter adjustment allows reliable gesture detection using a single sensor.
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 power-efficient recognition of the 'look up' gesture, reducing false positives in accelerated frames and ensuring accurate activation of wearable devices while conserving battery life.
Implementation Method 1
determining whether a head of the wearer is level relative to gravity based on the level-indication data
Data Source
AI summary
Methods, apparatus, and computer-readable media are described herein related to recognizing a look up gesture. Level-indication data from at least an accelerometer associated with a wearable computing device (WCD) can be received. The WCD can be worn by a wearer. The WCD can determine whether a head of the wearer is level based on the level-indication data. In response to determining that the head of the wearer is level, the WCD can receive lookup-indication data from at least the accelerometer. The WCD can determine whether the head of the wearer is tilted up based on the lookup-indication data. In response to determining that the head of the wearer is tilted up, the WCD can generate a gesture-recognition trigger, where the gesture-recognition trigger indicates that the head of the wearer has moved up from level.