Finger-Worn Device for External Computer Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing finger-worn devices for controlling external computers face challenges such as high bandwidth requirements, energy consumption, and precision issues in estimating finger position and orientation, often resulting in bulky devices with imprecise feedback and natural control feel.
Innovation Solution
A method and device that acquire and transmit position data of a finger relative to an object surface using a finger-worn device with integrated first and second sensor systems, processing the data with a machine learning method on a processor to estimate and transmit the finger's position and orientation, reducing bandwidth and improving precision, allowing for natural and instant control of external computers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If sensor data is transmitted to external computer for processing, then processing power is reduced at device, but bandwidth requirement increases
Solution Approach 1:
The patent extracts and processes only the essential position and orientation data locally using machine learning, transmitting only these extracted features to the external computer rather than transmitting all raw sensor data. This reduces the quantity of transmitted data while maintaining processing capability.
Solution Approach 2:
The patent segments the data processing task into two parts: local extraction of position and orientation features using machine learning on the device, and external processing of these extracted features on the computer. This segmentation reduces bandwidth requirements while distributing processing power.
2Measurement precision
If conventional algorithms are used for position and orientation estimation, then processing accuracy is maintained, but energy consumption increases
Solution Approach 1:
The patent replaces conventional computationally intensive algorithms with machine learning models that are optimized for efficient execution on resource-constrained devices. The machine learning approach provides comparable or superior accuracy while consuming significantly less energy.
Solution Approach 2:
The patent changes the processing parameters by using pre-trained machine learning models with optimized computational requirements. This allows accurate position and orientation estimation with reduced energy consumption compared to conventional algorithms.
3Measurement precision
If cameras with large lenses are used in finger-worn device, then sensor data quality is improved, but device size increases
Solution Approach 1:
The patent extracts essential visual information using compact sensor systems rather than relying on large lenses. The machine learning processing compensates for the smaller sensor size by intelligently analyzing the captured data to determine position and orientation.
Solution Approach 2:
The patent combines multiple sensor types and machine learning processing to achieve high measurement precision with compact hardware. The composite approach of sensors plus intelligent processing replaces the need for large optical components.
4Device complexity
If external devices are used for position and orientation estimation, then device complexity is reduced, but measurement precision decreases
Solution Approach 1:
The patent integrates multiple functions within the finger-worn device itself: sensor data acquisition, machine learning processing, and position/orientation calculation all occur in one device. This eliminates the need for separate external estimation devices while improving precision through direct local measurement.
Data Source
AI summary
The invention relates to a method and a computer program for wireless interactions with an external computer with a finger-worn device configured to acquire and transmit position data of at least one finger relative to an object surface to an external computer, wherein the method comprises the steps of:acquiring (300, 301) sensor data (201, 202) from a first sensor system (9) and a second sensor system (10) comprised in the device (1);estimating (302) a position and/or an orientation of at least one finger (2) with respect to an object surface (7) from the sensor data (201, 202), wherein the estimation (302) of the position and/or the orientation of the at least one finger (2) with respect to the object surface (7) is performed by a machine learning method (100) executed on a processor (4) comprised by the device (1);wirelessly transmitting (304) position data (203) comprising the estimated position and/or orientation of the at least one finger (2) with respect to the subject surface (7) to an external computer (15);relating (305) the estimated position and/or orientation of the at least one finger (2) with respect to the object surface (7) comprised in the position data (203) to a display position and/or a display orientation (204) in a coordinate system of a display (18);indicating the display position and/or the display orientation (204) on the display (18).The invention furthermore relates to a device (1) for executing the method according to the invention.