Gyroscope Swim Activity Detection for Accurate Rest Interval Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current swim tracking devices require manual input for non-swim interval detection, leading to inaccuracies and false positives due to accelerometer-based detection, especially during short rest periods and arm movements, affecting swim metrics accuracy.
Innovation Solution
A gyroscope-based controller processes raw signals through a Band Pass Filter and energy envelope estimator to differentiate swim and non-swim activities using sliding window averaging and threshold comparisons, eliminating the need for manual input and reducing false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If accelerometer-based detection is used for non-swim detection, then automated detection is achieved, but false positives occur due to normal hand movements during rest time
Solution Approach 1:
The patent replaces accelerometer-based mechanical detection with gyroscope-based detection. The gyroscope measures angular velocity and orientation changes, which are fundamentally different from the linear acceleration measurements of an accelerometer. This substitution allows the system to detect the characteristic rotational patterns of swimming strokes while ignoring the linear hand movements that occur during rest, thereby maintaining automation while eliminating false positives.
Solution Approach 2:
The patent changes the detection parameter from linear acceleration (accelerometer) to angular velocity and orientation (gyroscope). By monitoring rotational movement parameters instead of linear acceleration, the system can distinguish between swimming activities (which produce characteristic rotational patterns) and rest activities (which produce minimal rotational movement), thus improving detection reliability while maintaining automation.
2Reliability
If manual button pressing is required for non-swim detection, then detection accuracy is maintained, but user convenience deteriorates
Solution Approach 1:
The patent enables the system to automatically detect and record non-swim intervals without requiring user intervention. The gyroscope continuously monitors movement patterns, and the processing unit automatically identifies when swimming stops and when it resumes, eliminating the need for manual button pressing while maintaining accurate detection through objective motion analysis.
Solution Approach 2:
The patent replaces the manual mechanical action of button pressing with automated sensor-based detection. The gyroscope and processing unit work together to automatically identify non-swim intervals based on movement patterns, substituting the need for manual user input with an automated sensing and analysis system that provides both accuracy and convenience.
3Difficulty of detecting and measuring
If accelerometer detects hand movements during rest, then stroke detection sensitivity increases, but swim time accuracy decreases due to false stroke detection
Solution Approach 1:
The patent replaces accelerometer-based stroke detection with gyroscope-based detection. The gyroscope's ability to measure rotational movement allows it to detect the characteristic twisting and turning motions of swimming strokes while filtering out the linear hand movements that occur during rest. This substitution maintains high stroke detection sensitivity while eliminating false positives that contaminate swim time measurements.
Solution Approach 2:
The patent changes the detection parameter from linear acceleration to angular velocity and orientation. Swimming strokes produce distinctive rotational patterns that are easily captured by the gyroscope, while rest-period hand movements produce minimal rotational activity. This parameter change enables the system to maintain sensitive stroke detection while achieving precise swim time measurement by excluding false detections.
Data Source
AI summary
A controller is configured to receive raw signals from at least one gyroscope for at least two axes. The controller processes the raw signals through a BPF and output a filtered signal. The controller processes the filtered signal through an energy envelope estimator and determines an energy envelope signal for at least one axis. The controller then determines a nonswim activity in a segment based on the energy envelope signal through a detector. The energy envelope estimator is configured to generate the energy envelope signal using sliding window average of a preset window size over the filtered signal. The detector is configured to compare values of the energy envelope signal for at least one axis against respective threshold value, and classify the segment of the raw signals as non-swim activity upon satisfactory comparison.


