Accelerometer Wrist Stride Analysis via FFT Attenuation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing devices struggle to accurately measure the biomechanical efficiency of a runner's stride, as they are either uncomfortable, not portable, or provide inaccurate data when placed away from the runner's center of gravity, such as on the wrist, shoe, or ankle.
Innovation Solution
A method using a triaxial accelerometer device on the wrist, upper arm, or shoe that measures acceleration data, applies a Fast Fourier Transform (FFT) to extract the cadence frequency, and employs a machine learning system, like a neural network, to determine biomechanical parameters by attenuating extraneous motion frequencies, effectively converting wrist or shoe data into equivalent center of mass data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If an accelerometer is placed on the wrist or shoe, then the device is portable and comfortable, but the measurement precision of biomechanical parameters deteriorates
Solution Approach 1:
The patent uses machine learning algorithms as an intermediary to process the acceleration data collected from remote locations (wrist, shoe). The algorithm learns the relationship between remote sensor data and center of mass motion patterns, enabling accurate biomechanical parameter extraction without requiring direct placement at the center of gravity. This mediator bridges the gap between convenient remote sensing and accurate biomechanical measurement.
Solution Approach 2:
The patent transforms the measurement approach by changing from direct physical measurement at the center of mass to indirect measurement through machine learning parameter transformation. The system collects acceleration data from accessible locations and uses learned parameters to infer biomechanical characteristics that would otherwise require direct center of mass measurement, thereby maintaining portability while achieving measurement accuracy.
2Measurement precision
If an accelerometer is placed close to the center of gravity (chest, torso, back), then the measurement precision of biomechanical parameters improves, but the ease of operation deteriorates due to discomfort
Solution Approach 1:
The machine learning algorithm serves as an intermediary that eliminates the need for direct placement at the center of gravity. By learning the kinematic relationships between remote body parts and the center of mass, the system can accurately infer biomechanical parameters from wrist or shoe sensors, removing the discomfort associated with chest or back mounting while preserving measurement accuracy.
Solution Approach 2:
The system creates a virtual model of center of mass motion by processing data from remote sensors through machine learning. Instead of physically placing the sensor at the center of gravity, the algorithm copies and reconstructs the biomechanical information that would be obtained from direct measurement, enabling accurate analysis without the discomfort of improper device placement.
3Device complexity
If traditional methods are used to analyze stride efficiency, then the device complexity is low, but the measurement precision deteriorates due to inability to accurately capture center of gravity motion
Solution Approach 1:
The patent replaces complex mechanical measurement systems (direct center of mass sensors, motion capture equipment) with a simpler accelerometer-based system combined with machine learning. The algorithm substitutes for the need for complex mechanical setups by computationally inferring center of mass motion from accessible sensor locations, thereby reducing device complexity while improving or maintaining measurement precision.
Solution Approach 2:
The system changes the measurement parameters from direct physical quantities (center of mass position, velocity) to derived parameters obtained through machine learning processing of acceleration data. This parameter transformation enables accurate stride efficiency analysis using simple accelerometers, avoiding the need for complex measurement equipment while maintaining high precision.
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 accurate, real-time analysis of biomechanical parameters like cadence, stride efficiency, and musculoskeletal performance without the need for uncomfortable chest or back-mounted devices, providing portable and user-friendly data for runners.
Implementation Method 1
measuring an initial sequence of acceleration data in at least the vertical direction with said accelerometer device
Implementation Method 2
performing a FFT of said sample window, so as to compute a FFT signal
Data Source
AI summary
A method for determining biomechanical parameters of the stride of a runner, using an accelerometer device on the wrist, on the upper arm, on the head or on the shoe, comprising the steps of measuring an initial sequence of acceleration data in at least the vertical direction using said accelerometer device; identifying in said initial sequence of acceleration data at least one frequency component caused by relative motion of the accelerometer device relative to the runner center of masse; attenuating said frequency component, so as to determine a modified sequence of acceleration data; determining said biomechanical parameters from said modified sequence of acceleration data.