Inertial Sensor Performance Indicator Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for analyzing human or animal movements during exercise provide limited real-time feedback and are hindered by the need for multiple sensors, which can interfere with the athlete's performance and are difficult to use in dynamic environments like cross-country skiing or running.
Innovation Solution
A method and system using a single inertial sensor to collect time series data, partition it into movement periods, transform the data into a defined representation, and classify it using a Markov chain of multivariate Gaussian distributions to determine performance indicators in real-time, minimizing disruption and providing accurate feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors are placed on the athlete's body to capture movement data, then measurement precision is improved, but device complexity and interference with the athlete's movements increase
Solution Approach 1:
The patent combines multiple sensor functions into a single inertial measurement unit (IMU) that integrates accelerometer, gyroscope, and magnetometer components. This consolidation captures comprehensive movement data (acceleration, orientation, position) from one device rather than requiring multiple separate sensors, thereby maintaining measurement precision while reducing device complexity and minimizing interference with the athlete's movements
Solution Approach 2:
The single inertial sensor system is designed to perform multiple measurement functions simultaneously - tracking position, velocity, acceleration, and orientation of the athlete's movement. This multi-functional approach replaces what would traditionally require several specialized sensors, achieving both comprehensive data capture and reduced system complexity
2Loss of information
If multiple sensors are used to capture comprehensive movement data, then information completeness is improved, but ease of operation deteriorates due to difficulty in analyzing large amounts of data
Solution Approach 1:
The system processes sensor data in real-time and provides immediate feedback to the athlete through visual or audio signals indicating performance quality and technique correctness. This continuous feedback loop transforms complex analytical data into actionable, easy-to-understand guidance, maintaining information completeness while dramatically improving ease of operation during exercise
Solution Approach 2:
The system includes automated algorithms that independently analyze the captured movement data, classify movement patterns, and generate performance assessments without requiring manual intervention from trainers or analysts. This self-processing capability handles the complexity of data analysis automatically, preserving complete movement information while eliminating the operational burden of manual data interpretation
3Measurement precision
If video analysis tools are used to capture and analyze movements, then measurement precision is improved, but productivity deteriorates due to post hoc analysis only and difficulty in capturing movements over large areas
Solution Approach 1:
The inertial sensor system operates continuously throughout the exercise, capturing movement data in real-time as the athlete performs. This continuous data collection and processing enables immediate performance feedback during the activity itself, rather than requiring post-exercise analysis, thereby maintaining high measurement precision while dramatically improving productivity through instantaneous insights
Solution Approach 2:
The patent replaces the mechanical/video-based analysis system with an electronic inertial sensing system. Unlike video analysis that requires cameras, lighting, and complex image processing, the inertial sensors electronically capture and process movement data directly, enabling both high precision measurement and rapid real-time feedback without the logistical constraints of video equipment
Data Source
AI summary
A method for determining a value of a performance indicator for a periodic movement performed by a human or an animal. The method includes receiving (210) time series data from an inertial sensor used for sensing the periodic movement, partitioning (220) the received time series data into periods of the movement, and transforming (230) the time series data of each period into a data representation of a defined size. The method further includes classifying (240) each period into a movement class, and determining (250) a value of a performance indicator for at least one period of the movement, based on the data representation and on the movement class of the at least one period.