Orientation Locus Device for Embedded Gesture Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for analyzing movements using orientation measuring sensors face challenges in reducing memory space and calculation time while preserving useful information, particularly in embedded applications like mobile devices, where raw sensor signals are cumbersome and inefficient.
Innovation Solution
A device that determines the orientation of a moving coordinate frame with respect to a reference frame and calculates a locus of points on a surface, such as a sphere or polyhedron, to represent movements, reducing memory usage and enabling easier gesture recognition by compressing information into a more manageable and recognizable format.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If raw sensor signals are stored and processed directly, then complete movement information is preserved, but memory space and calculation time increase significantly
Solution Approach 1:
The patent extracts only the essential information from raw sensor signals by determining the orientation of the moving coordinate frame and generating a simplified locus representation. This extraction process removes redundant data while preserving the core movement characteristics needed for analysis, thereby reducing memory requirements without significant information loss.
Solution Approach 2:
Instead of storing raw sensor data and then processing it, the patent inverts the approach by immediately transforming sensor readings into a compressed representation (orientation + locus points) at the point of data generation. This inversion occurs at the data acquisition stage rather than during subsequent processing, achieving compression before storage.
2Measurement precision
If raw sensor signals are processed directly, then accurate movement analysis is achieved, but calculation time increases
Solution Approach 1:
The patent extracts only the critical movement parameters (orientation angles and key locus points) from the complete sensor signal spectrum. This selective extraction focuses computational resources on calculating only the essential features needed for movement analysis, significantly reducing calculation time while maintaining measurement precision for the extracted parameters.
Solution Approach 2:
The patent applies partial action by computing only the necessary subset of movement characteristics (orientation and locus representation) rather than processing the entire raw signal dataset. This partial processing approach achieves sufficient analysis accuracy for embedded applications without the computational burden of complete signal processing.
3Measurement precision
If complete sensor data is used for gesture recognition, then recognition accuracy is improved, but device complexity increases
Solution Approach 1:
The patent extracts the essential gesture-defining features from complete sensor data by representing movements as sequences of orientation states and locus points. This extraction creates a simplified data structure that maintains gesture recognition accuracy while significantly reducing the complexity of the processing system required to analyze the data.
Solution Approach 2:
The patent changes the parameter representation from raw sensor signals (multiple axes, high frequency) to a transformed parameter set (orientation angles, locus coordinates). This parameter transformation simplifies the data structure and reduces processing complexity while preserving the information necessary for accurate gesture recognition.
4Adaptability or versatility
If raw movement data is stored, then detailed movement analysis is possible, but memory requirements increase
Solution Approach 1:
The patent extracts the fundamental movement characteristics (orientation trajectory and locus points) that enable detailed movement analysis. This extracted representation maintains adaptability for various analysis tasks while occupying minimal memory space, as it stores only the essential geometric features rather than complete raw signal waveforms.
Solution Approach 2:
The patent creates a simplified copy or representation of the movement data in the form of orientation sequences and locus points. This copied representation preserves the essential movement patterns needed for flexible analysis while requiring far less memory than storing the original high-resolution sensor signals.
Data Source
AI summary
A device for analyzing the movement of at least one moving element (EM), provided, for at least one moving element, with first means (DET1) for determining the orientation of a moving coordinate frame (Rm) connected in motion to the moving element (EM), with respect to a reference coordinate frame (Rr), including second means (DET2) for determining at least one locus of points (Tx, Ty, Tz) of at least one surface from at least one respective direction of an oriented axis (x, y, z) of the moving coordinate frame (Rm) connected in motion to the moving element (EM) and said surface.