Capacitive Touch Localization with Machine Learning Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing touch localization methods on capacitive interfaces are inaccurate due to variations in user interaction angles and noise sources, leading to errors in activating unintended functions and increased user burden.
Innovation Solution
A machine learning model, such as a convolutional neural network, is trained on capacitive sensor data to estimate touch locations by learning user touch behavior, and position-based adjustments are applied to correct systematic errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional capacitance change measurement methods are used for touch localization, then the system is simple and easy to implement, but the touch localization accuracy deteriorates due to variations in user interaction angles and noise sources
Solution Approach 1:
A machine learning model is introduced as an intermediary between the capacitive sensor data and the touch location determination. The model processes the raw capacitance changes and outputs corrected touch location coordinates, effectively mediating between the simple sensor measurement and the accurate location result while compensating for systematic errors.
Solution Approach 2:
The system transforms the raw capacitance change measurements into corrected touch location coordinates by applying a machine learning model that learns the relationship between capacitance patterns and actual touch positions. This parameter transformation compensates for systematic errors and improves localization accuracy.
2Measurement precision
If machine learning models are used to estimate touch locations, then touch localization accuracy improves by capturing nuanced touch features, but device complexity increases due to the need for training and processing
Solution Approach 1:
The machine learning model is trained in advance on a dataset of capacitive sensor readings and corresponding ground truth touch locations. This preliminary training phase allows the model to learn systematic error patterns and touch behavior characteristics before deployment, enabling accurate touch localization during actual use without requiring complex real-time processing.
Solution Approach 2:
The machine learning model creates a computational copy or representation of the relationship between capacitance patterns and touch positions by learning from training data. This learned model can then be applied repeatedly to new touch inputs, capturing nuanced touch features without requiring the complexity of re-analyzing raw sensor data from scratch each time.
3Measurement precision
If position-based adjustments are applied to correct systematic errors, then touch localization accuracy improves, but the processing time and computational load increase
Solution Approach 1:
Position-based adjustment parameters are pre-computed during the model training phase by comparing predicted touch locations with ground truth positions. This preliminary computation of correction parameters allows the system to apply simple, fast adjustments during actual touch detection without requiring complex real-time optimization or iterative processing.
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
Improves touch localization accuracy by capturing nuanced touch features and compensating for device-specific errors, resulting in precise touch location estimation.
Implementation Method 1
The input panel of a capacitive touchscreen consists of an insulator that is coated with a transparent conductor, such as indium tin oxide. When an input device (e.g., a user's finger, conductive stylus, etc.) touches or is brought near the surface of a capacitive touchscreen, the local electrostatic field is distorted. The resultant change in capacitance can be measured and used to detect the touch input and determine its location on the touchscreen.
Implementation Method 2
When an input device (e.g., a user's finger, conductive stylus, etc.) touches or is brought near the surface of a capacitive touchscreen, the local electrostatic field is distorted.
Data Source
AI summary
Implementations of the present application provide methods for estimating a location of a detected touch input on a capacitive interface. A capacitive image representing capacitive sensor data associated with the detected touch input is obtained, and a touch region of the capacitive image is identified. The touch region corresponds to an area of the capacitive interface associated with the detected touch input. A first set of features of the identified touch region is extracted using a machine learning model. The extracted first set of features is used to determine first coordinates of the location of the touch input.