Magnetic Force Sensor Layout Using ML-Calibrated Field Components
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic sensor systems face challenges in accurately measuring force components and tilting angles of joysticks or thumbsticks while being less sensitive to external disturbance fields, temperature variations, and demagnetization, often requiring complex analytical formulas and conflicting with other performance requirements.
Innovation Solution
A magnetic sensor system with a semiconductor substrate and a movable axially magnetized two-pole magnet, using a combination of 2D and 3D magnetic sensors, and a processing circuit that determines physical quantities through a predefined algorithm with machine-learned constants, reducing sensitivity to external disturbances and temperature variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If magnetic sensor systems use complex analytical formulas to accurately measure force components and tilting angles, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex analytical calculation systems with a machine learning-based system. Instead of using explicit analytical formulas to compute force components and tilting angles from magnetic field measurements, the system employs a trained machine learning model that directly maps magnetic field sensor outputs to physical quantities. This substitution of the computational mechanism reduces device complexity while maintaining measurement precision.
Solution Approach 2:
The patent changes the computational parameters from requiring explicit analytical formulas to using machine learning model parameters. The system transforms the measurement approach by using a trained model with learned parameters that directly relate magnetic field characteristics to physical quantities, eliminating the need for complex real-time analytical calculations and reducing computational burden.
2Device complexity
If magnetic sensor systems use simple or cheap magnetic structures, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent replaces the need for complex magnetic structures with a machine learning-based computational system. Instead of relying on sophisticated magnetic field generation structures to achieve high measurement precision, the system uses a trained machine learning model that can accurately interpret magnetic field measurements from simpler, cheaper magnetic structures, thereby maintaining precision while reducing structural complexity.
3Adaptability or versatility
If magnetic sensor systems use resiliently mounted magnets to measure force components, then measurement capability is improved, but sensitivity to external disturbance fields increases
Solution Approach 1:
The patent employs a machine learning model that learns from training data to distinguish between magnetic field changes caused by applied forces and those caused by external disturbance fields. The model incorporates feedback mechanisms where it processes multiple magnetic field measurements and uses learned patterns to filter out disturbances, maintaining measurement capability while reducing sensitivity to harmful external fields.
Solution Approach 2:
The patent changes the system's response parameters by using a machine learning model that adapts to different measurement conditions. The model learns optimal parameter relationships during training that account for typical disturbance patterns, allowing the system to maintain versatility in measuring force components while becoming less sensitive to external disturbances through learned parameter adjustments.
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
The system achieves accurate measurement of force components and tilting angles with reduced sensitivity to external disturbances and temperature variations, allowing for robust and efficient operation without the need for explicit analytical formulas.
Implementation Method 1
a permanent magnet which is movable relative to the integrated circuit, and configured for generating a magnetic field
Implementation Method 2
a plurality of magnetic sensors configured for measuring at least two first magnetic field components (Bx1, Bx2) oriented in a first direction (X), and for measuring at least two second magnetic field components (Bz1, Bz2) oriented in a second direction (Z)
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
Magnetic sensor system comprising: an integrated circuit comprising a semiconductor substrate, the semiconductor substrate comprises a plurality of magnetic sensors configured for measuring at least two first magnetic field components (Bx1, Bx2) oriented in a first direction (X), and for measuring at least two second magnetic field components (Bz1, Bz2) oriented in a second direction (Y; Z); a permanent magnet movable relative to the integrated circuit and configured for generating a magnetic field; a processing circuit configured for determining at least two physical quantities (Fx, Fy, Fz) related to a position of the magnet, using a predefined algorithm based on the measured first and second magnetic field components (Bx1, Bx2; Bz1, Bz2) or values derived therefrom, as inputs, and that uses a plurality of at least eight constants which are determined using machine learning. A force sensor system. A joystick or thumbstick system. A method.