Shoe Sensor Walk Discrimination Using Attitude Angle
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing step counting devices using acceleration sensors often erroneously detect movements other than walking, such as sitting or fidgeting, as walking due to reliance on acceleration data alone.
Innovation Solution
A walk discrimination system that utilizes both acceleration and angular velocity data from sensors in shoes to calculate attitude angles and determine the presence of walking steps by identifying specific patterns in these data, including negative and positive peaks, to accurately differentiate between walking and non-walking movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If step counting is performed based solely on acceleration data, then the device complexity is reduced and ease of operation is improved, but measurement precision deteriorates due to erroneous detection of non-walking movements
Solution Approach 1:
The patent combines acceleration sensor data and angular velocity sensor data into a unified walk detection system. By merging multiple sensor inputs and analyzing their combined patterns (including attitude angle calculations), the system achieves higher measurement precision in distinguishing walking from non-walking movements, resolving the contradiction between accuracy and simplicity.
2Measurement precision
If multiple sensor data types are used to improve walk detection accuracy, then measurement precision is improved, but device complexity increases due to additional processing requirements
Solution Approach 1:
The patent segments the walk detection process into distinct analytical stages: attitude angle calculation from sensor data, threshold comparison, peak detection, and sequential pattern recognition. This segmentation of the processing workflow makes the complex multi-sensor analysis more manageable and systematic, reducing the perceived complexity while maintaining high measurement precision.
Solution Approach 2:
The patent performs preliminary calculations of attitude angles from acceleration and angular velocity data before conducting the actual walk detection. By pre-processing the sensor data into meaningful derived parameters (attitude angles), the system simplifies subsequent detection logic and reduces real-time processing complexity while improving detection accuracy.
3Reliability
If acceleration threshold detection is used alone, then the response speed is improved, but reliability deteriorates due to false positive detections
Solution Approach 1:
The patent calculates attitude angles as a preliminary step before performing threshold comparisons and peak detections. This pre-processing enables the system to establish more reliable detection criteria based on attitude changes rather than raw acceleration alone, improving reliability without significantly increasing detection time.
Solution Approach 2:
The patent continuously monitors attitude angles and their changes throughout the detection process, maintaining continuous analysis of the sensor data stream. This continuous monitoring ensures that walk detection relies on sustained pattern recognition rather than isolated threshold crossings, improving reliability while keeping detection time efficient through streamlined processing.
Data Source
AI summary
A walk discrimination device includes: a data reception unit that receives sensor data including an acceleration and an angular velocity acquired by a sensor installed in a shoe; an acceleration determination unit that determines whether the acceleration included in the sensor data is positive or negative; an attitude angle calculation unit that calculates an attitude angle by use of the acceleration and the angular velocity included in the sensor data; an attitude angle determination unit that determines whether the attitude angle calculated by the attitude angle calculation unit has exceeded a threshold, and determines whether a peak of the attitude angle has been detected; a walk discrimination unit that discriminates a walk based on a determination order by the acceleration determination unit and the attitude angle determination unit; and an output unit that outputs a discrimination result by the walk discrimination unit.