Input Device Load Calculation for Multi-Point Touch Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional input devices face challenges in accurately calculating loads applied to a touch panel surface due to noise interference, non-existence of matrix inverses, and high computational complexity, especially when multiple points are collinear or noise is present.
Innovation Solution
An input device with a load calculation method that includes a calculation loop to adjust initial pressed point loads based on comparison values and load distribution coefficients, reducing reliance on matrix inverses and improving accuracy with reduced computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If matrix inverse calculation is used to calculate pressed point loads, then calculation accuracy is improved, but computational complexity increases and noise sensitivity worsens
Solution Approach 1:
The patent segments the load calculation process into two distinct phases: a preprocessing phase where matrix inverse is calculated once to obtain load distribution coefficients, and a real-time phase where these coefficients are simply multiplied by load sensor values. This segmentation separates the computationally intensive operation from the time-critical operation, resolving the contradiction between calculation accuracy and computational complexity.
Solution Approach 2:
The patent performs the matrix inverse calculation in advance during system initialization or calibration, storing the resulting load distribution coefficients for later use. This preliminary action eliminates the need for repeated matrix inversion during actual load measurements, significantly reducing real-time computational complexity while maintaining calculation accuracy.
2Measurement precision
If matrix inverse calculation is used to calculate pressed point loads, then calculation accuracy is improved, but noise interference worsens
Solution Approach 1:
By calculating the matrix inverse in advance during a calibration phase with controlled conditions, the patent creates a stable set of load distribution coefficients that are less sensitive to operational noise. This preliminary calculation under ideal conditions filters out the impact of noise that would otherwise affect real-time calculations.
Solution Approach 2:
The patent creates a simplified model (load distribution coefficients) that copies the essential relationship between load sensor readings and pressed point loads without requiring the full complexity of real-time matrix inversion. This copied model maintains accuracy while being robust against noise in operational conditions.
3Quantity of substance
If matrix inverse calculation is used for multiple pressed points, then load calculation is performed, but calculation time increases extremely
Solution Approach 1:
The patent performs the computationally intensive matrix inverse calculation once during initialization, regardless of how many pressed points will be detected later. The resulting load distribution coefficients are then reused for any number of pressed points, making the per-point calculation time constant and independent of the total number of pressed points.
Solution Approach 2:
The patent changes the calculation parameters from requiring full matrix inversion for each pressed point to using pre-computed load distribution coefficients. This parameter change transforms the computational complexity from O(n³) per pressed point to O(1) per pressed point, where n is the number of load sensors.
4Reliability
If pressed point is intentionally shifted to ensure matrix inverse existence, then calculation can be performed, but load calculation accuracy worsens
Solution Approach 1:
The patent performs the matrix inverse calculation in advance during a calibration phase where the actual pressed point positions are known and can be used to construct a well-conditioned matrix. This preliminary action ensures that the matrix inverse exists and is accurate for the actual application, avoiding the need for artificial point shifting during operational calculations.
Data Source
AI summary
There is provision of an input device for calculating loads applied to multiple pressed points respectively, by executing one or more calculation loops in a load calculation unit. The calculation loop includes calculating computation values representing loads detected at load sensors based on initial pressed point loads, calculating a comparison value based on a comparison of the computation value with a detection value detected by the load sensor, calculating an adjustment value for adjusting the initial pressed point load based on the comparison values and pressed positions, and calculating the pressed point loads by adjusting the initial pressed point load based on the adjustment value. The load calculation unit uses the pressed point load calculated in the calculation loop for each of the pressed positions as an initial pressed point load in the next calculation loop.