Wearable Attitude Matrix Calculation Using Gyroscope Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing attitude matrix calculation methods for wearable devices require complex convergence and matrix inversion, leading to high power consumption and memory usage, especially when calculating motion distance based on tri-axis gyroscope, accelerometer, and magnetometer data.

Innovation Solution

A method that calculates the attitude matrix using only real-time gyroscope data, eliminating the need for accelerometer and magnetometer measurements, and employs a constant transformation matrix between the sensor and foot coordinate systems to reduce computational complexity and power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complex convergence algorithms (unscented Kalman filter or extended Kalman filter) are used to calculate attitude matrix from tri-axis gyroscope, accelerometer, and magnetometer data, then measurement precision and reliability are improved, but device complexity, power consumption, and memory usage increase

Engineering Contradiction:
Improveattitude matrix calculation precisionVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts and removes the accelerometer and magnetometer from the attitude matrix calculation process, using only gyroscope data. This eliminates the need for complex multi-sensor convergence algorithms while maintaining acceptable measurement precision for motion distance calculation.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses a simpler, computationally cheaper calculation method that sacrifices some theoretical precision for significant reductions in power consumption and device complexity. The simplified algorithm is sufficient for the practical application of motion tracking.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If complex convergence algorithms are used to calculate attitude matrix, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improveattitude matrix calculation precisionVSAvoidalgorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent removes the complex convergence algorithms and multi-sensor processing requirements, retaining only the essential gyroscope-based calculation. This dramatically simplifies the device complexity while maintaining the core functionality of attitude matrix calculation.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If complex matrix inversion and covariance matrix calculation are performed during convergence, then measurement precision is improved, but loss of substance (memory usage) increases

Engineering Contradiction:
Improveattitude matrix calculation precisionVSAvoidmemory usage
Core Design Contradiction:
Measurement precisionVSLoss of substance

Solution Approach 1:

The patent extracts and eliminates the matrix inversion and covariance matrix calculation steps from the algorithm. By using only gyroscope data with a simplified calculation approach, these computationally intensive operations are removed, significantly reducing memory requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3671119B1Attitude matrix calculating method and device
Publication Date: 2024.04.10 HUAWEI TECH CO LTD
  • EP3671119B1 patent drawingFigure 1~2a
  • EP3671119B1 patent drawingFigure 2b~2c
  • EP3671119B1 patent drawingFigure 2d~2e

AI summary

An attitude matrix calculation method and a device are disclosed and relate to the field of terminal technologies, to reduce a calculation amount, reduce power consumption of a device, and save memory of the device. The method is as follows: calculating, based on a first coordinate transformation matrix and a measured value of a gyroscope that is corresponding to the ith sampling moment, a second coordinate transformation matrix corresponding to the ith sampling moment, where the first coordinate transformation matrix is a constant matrix, the first coordinate transformation matrix is a coordinate transformation matrix between a sensor coordinate system and a foot coordinate system, the second coordinate transformation matrix is a coordinate transformation matrix between the foot coordinate system and a ground coordinate system, and i is an integer greater than 0; and calculating, based on the first coordinate transformation matrix and the second coordinate transformation matrix corresponding to the ith sampling moment, an attitude matrix corresponding to the ith sampling moment, where the attitude matrix is a coordinate transformation matrix between the sensor coordinate system and the ground coordinate system.