Touchscreen Controller Predictive Reporting via Linear Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional touchscreen report methods are limited by the scan time, which cannot be reduced without compromising data precision, resulting in a slow touch report rate that hinders user interaction.
Innovation Solution
Implementing a machine learning model on the touchscreen controller to predict touch coordinates based on historical data using linear regression, allowing for anticipatory reporting of touches without increasing scan time, thereby enhancing the touch report rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the scan time is reduced to increase touch report rate, then the touch report rate improves, but the data precision deteriorates
Solution Approach 1:
The system performs preliminary actions by collecting historical touch data and training machine learning models in advance. The ML model predicts future touch coordinates based on patterns learned from historical data, allowing the system to report predicted touches between actual scan cycles without compromising the actual scan time and data precision.
Solution Approach 2:
The system creates copies of touch information by generating predicted touch coordinates based on historical patterns. These predicted touch copies are reported alongside or between actual touch data, effectively increasing the touch report rate without requiring more frequent actual scans that would compromise precision.
2Measurement precision
If the scan time is increased to improve data precision, then the data precision improves, but the touch report rate deteriorates
Solution Approach 1:
The machine learning model acts as an intermediary between the slow actual scan process and the need for frequent touch reporting. It processes historical scan data to generate predicted touch coordinates, filling the gaps between actual scans and enabling high-rate reporting without increasing actual scan frequency or compromising precision.
Solution Approach 2:
The system performs preliminary data collection and model training during normal operation, building up historical touch data that the ML model can later use to generate predictions. This preliminary preparation enables the system to maintain low scan times while still providing high-precision predicted touch reports.
3Productivity
If machine learning model is implemented to predict touch coordinates, then the touch report rate improves, but the device complexity increases
Solution Approach 1:
The touchscreen controller performs self-service by implementing the machine learning model internally to generate predicted touch coordinates. This self-contained approach eliminates the need for external processing systems, reducing overall system complexity while still achieving improved touch report rates through intelligent prediction.
Data Source
AI summary
A method for reporting touch on a touchscreen includes detecting first touch data from the touchscreen corresponding to a first touch on the touchscreen; determining coordinates of the first touch from the first touch data; reporting the coordinates of the first touch at a first time; determining predicted coordinates of a second touch based on a linear regression of historical touch data; and reporting the predicted coordinates of the second touch at a second time, where the second time occurs after the first time.