3-Axis Accelerometer Gravity Signal Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing physical activity monitoring systems fail to accurately monitor posture and intensity of physical activity due to issues like gravity signal suppression, alignment errors, and inability to distinguish between potential and kinetic energy expenditure, leading to measurement inaccuracies and increased complexity with additional sensors.
Innovation Solution
A 3-axis sensor system that estimates both gravity and acceleration vectors, providing separate data streams to accurately determine activity type and intensity, with features like feedback loops for error correction and integration into portable devices like smartphones, which can automatically calibrate and distinguish between different energy expenditure types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If high-pass filtering is used to suppress gravity signal, then gravity suppression is improved, but measurement precision deteriorates due to posture changes creating AC components that feed through the filter
Solution Approach 1:
The system dynamically adapts the filtering approach based on detected posture stability. When posture changes are detected, the system switches from high-pass filtering to alternative methods such as vector magnitude analysis or machine learning-based gravity separation, preventing filter feed-through while maintaining acceleration measurement accuracy during dynamic activities
Solution Approach 2:
The system changes the processing parameters by using tri-axial accelerometer data to compute vector magnitude and direction, then applies context-aware filtering that adjusts filter characteristics based on activity type and posture stability detection, rather than using fixed high-pass filtering parameters
2Object-affected harmful factors
If low-pass filtering is used to suppress acceleration signal, then acceleration interference is reduced, but measurement precision deteriorates due to feed-through of signal components in either direction
Solution Approach 1:
The system segments the acceleration signal processing into distinct components: gravity vector estimation, acceleration vector extraction, and activity classification. Each component is processed separately with appropriate filtering applied only where needed, avoiding the need for broad low-pass filtering that would distort the acceleration signal
3Measurement precision
If additional sensors are added to monitor posture and acceleration separately, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system makes the tri-axial accelerometer universal by using it to perform multiple functions: gravity detection, acceleration measurement, posture estimation, and activity classification. Through sophisticated signal processing algorithms, a single sensor achieves capabilities that would traditionally require multiple specialized sensors
Solution Approach 2:
The system replaces mechanical/postural sensors with computational methods that use accelerometer signal characteristics to infer posture and movement type. Machine learning algorithms analyze acceleration patterns to determine body orientation and activity type without requiring direct mechanical sensing of posture
4Measurement precision
If periodic calibration is performed to adjust sensor accuracy, then measurement precision is improved, but productivity deteriorates due to interruption of data capture
Solution Approach 1:
The system performs self-calibration by detecting periods when the user is stationary or performing known reference movements (such as standing still or holding the device in specific orientations). During these naturally occurring moments, the system automatically adjusts calibration parameters without requiring user intervention or interrupting the monitoring workflow
Solution Approach 2:
The system performs preliminary calibration during initial device setup and uses factory-calibrated reference data to establish baseline accuracy. Ongoing calibration is performed in the background during routine activities, preparing correction factors before they are needed for accurate measurement
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
This solution improves the accuracy of posture and acceleration estimation, reduces sensor errors, and allows for more precise monitoring of physical activity patterns, movement efficiency, and diagnostic purposes without the need for additional costly sensors.
Implementation Method 1
a 3-axis sensor operative to sense apparent acceleration along each of a first axis, a second axis, and a third axis
Data Source
AI summary
An improved apparatus and methods of posture and physical activity monitoring. The apparatus is physically mountable to or associated with an object or person, includes a multi-axis accelerometer, and derives measurements of posture and of acceleration. Methods are disclosed which provide improved estimations of posture, acceleration, energy expenditure, movement characteristics and physical activity, detect the influence of externally-caused motion, and permit automatic calibration of the apparatus in the field.