Wearable Magnetometer Motion Estimation via Magnetic Field Distortion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image data analysis techniques struggle to predict object motion in real-time without using image data, especially when the object is outside the camera's field of view.
Innovation Solution
The use of a wearable device equipped with magnetometers and other sensors to detect changes in local magnetic fields, allowing for the estimation of user actions and motion characteristics, even when ferromagnetic objects are outside the camera's view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image data analysis techniques are used to predict object motion, then motion prediction capability is provided, but real-time prediction is not achieved and the system cannot detect objects outside camera field of view
Solution Approach 1:
The patent replaces image data analysis (optical/mechanical system) with magnetometer-based detection (magnetic field system). Magnetometers can detect ferromagnetic objects and their motion in real-time without requiring camera field of view, enabling simultaneous real-time prediction and extended detection range.
Solution Approach 2:
The patent introduces magnetic field distortion as an intermediary between the ferromagnetic object and the detection system. The magnetometer detects changes in magnetic field caused by ferromagnetic objects, translating physical motion into detectable signal changes for real-time analysis.
2Loss of time
If magnetometers are used to detect ferromagnetic objects, then real-time detection and extended field of view are achieved, but device complexity increases
Solution Approach 1:
The patent makes the magnetometer system multi-functional by combining it with machine learning models that can classify various motions (e.g., bicep curls, shoulder presses, rows). The same magnetometer hardware serves multiple detection purposes, reducing the need for separate specialized sensors for each motion type.
Solution Approach 2:
The patent creates a virtual model of motion patterns through machine learning training. Instead of using complex hardware for each motion type, the system copies motion characteristics into a computational model that can recognize patterns from magnetometer data, simplifying the physical device while maintaining detection capability.
3Use of energy by moving object
If wearable devices with magnetometers are used for motion estimation, then power consumption is reduced for long-term use, but measurement precision may be affected
Solution Approach 1:
The patent implements periodic sampling of magnetometer data rather than continuous processing. The system samples magnetic field changes at intervals during exercise routines, reducing computational load and power consumption while maintaining sufficient measurement precision through strategic sampling points in the motion cycle.
Solution Approach 2:
The system uses the ferromagnetic objects themselves as part of the sensing mechanism. The objects being exercised with (weights, dumbbells) inherently distort the magnetic field, providing the measurement signal without requiring separate active sensors on the objects. This self-service approach reduces system complexity and power requirements.
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 approach enables accurate and real-time estimation of user motion and activity repetitions with minimal power consumption, making it suitable for long-term wear and use in various environments.
Implementation Method 1
detect change in a local magnetic field... detect the distortion of a local magnetic field as a ferromagnetic object moves in the vicinity
Data Source
AI summary
Predicting and counting repetitions of a physical activity includes capturing first sensor data, by a first magnetometer on a wearable device, a change in a magnetic field indicative of a ferromagnetic object moving in relation to the wearable device. One or more characteristics of a user motion are determined based on the first sensor data. A count of repetitions of the user motion are determined based on the one or more characteristics of the user motion, and a notification of the count of repetitions is generated.