Stride Length Estimation via Dynamic Acceleration Correlation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing stride length estimation methods, such as those using fixed tables, fail to accurately account for variations in user physical attributes and gait, leading to inaccurate stride length calculations.

Innovation Solution

A method that detects vertical movement acceleration and adjusts the correlation between acceleration and stride length based on user speed, using a correlation modeling equation to estimate stride length more accurately, and includes extracting low-frequency components from acceleration data to improve estimation accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a fixed table is used to estimate stride length, then the device complexity is reduced, but the measurement precision deteriorates because stride length variations due to user physical attributes and gait cannot be accounted for

Engineering Contradiction:

Improvestride length estimation accuracyVSAvoidestimation method complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the correlation parameter between acceleration and stride length dynamically based on user speed. Instead of using a fixed correlation, the system adjusts the correlation value according to the detected user speed, allowing accurate stride length estimation across different walking conditions while maintaining a relatively simple device structure

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system transitions from a static fixed-table approach to a dynamic estimation method where the correlation between acceleration and stride length is adjusted in real-time based on user speed. This dynamic adaptation enables the system to account for variations in gait and physical attributes without requiring complex individual user profiling

Inventive Principle:

Principle #15Dynamics

2Measurement precision

If high-frequency noise is included in acceleration data, then the data utilization is maximized, but the measurement precision deteriorates due to noise interference

Engineering Contradiction:

Improvestride length estimation accuracyVSAvoidacceleration data quality

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts the low-frequency component from the acceleration signal while removing high-frequency noise. By applying frequency filtering to separate the useful low-frequency motion information from the harmful high-frequency noise, the system achieves accurate stride length estimation without losing essential motion data

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS8718970B2Method of estimating stride length, method of calculating movement trajectory, and stride length estimating device

Publication Date: 2014.05.06 SEIKO EPSON CORP

AI summary

A method of estimating a stride length includes: detecting acceleration of a user's vertical movement; and estimating a stride length using the detected acceleration by changing the degree of correlation between the acceleration and the stride length, in which the stride length is increased as the detected acceleration increases, based on user's speed.