Touch Position Mapping Using Stochastic Movement Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing touch sensor systems face challenges in accurately mapping measured touch positions to contact identifiers due to the combinatorial assignment problem, often resulting in incorrect assignments when using simple distance metrics like Euclidean or Manhattan distances, which fail to account for the speed, direction, and likelihood of acceleration of object movement.
Innovation Solution
The implementation of a stochastic model-based cost function that accounts for the speed, direction, and likelihood of acceleration parallel and perpendicular to the object's movement, using a processor and memory system to map estimated object locations and determine the optimum assignment through a cost matrix optimized by the Hungarian algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If simple distance metrics (Euclidean or Manhattan distance) are used to map measured touch positions to contact identifiers, then the calculation is computationally simple, but the assignment accuracy deteriorates due to failure to account for speed, direction, and acceleration of object movement
Solution Approach 1:
The patent transforms the cost function from a simple distance metric to a stochastic model-based cost function that incorporates multiple motion parameters including speed, direction, and acceleration likelihood. This parameter enrichment allows the system to account for the dynamic characteristics of moving contacts, significantly improving assignment accuracy while maintaining computational feasibility through efficient algorithmic implementation.
Solution Approach 2:
The patent introduces a stochastic model as an intermediary between the raw distance measurement and the final assignment decision. This model acts as a mediator that processes the distance information through the lens of expected motion patterns, filtering out assignments that contradict the predicted movement behavior of contacts. The intermediary model enables the system to evaluate not just spatial proximity but also temporal consistency with observed motion dynamics.
2Measurement precision
If a stochastic model-based cost function is used to account for speed, direction, and acceleration likelihood, then the assignment accuracy improves, but the device complexity increases
Solution Approach 1:
The patent segments the complex stochastic model into distinct computational components: a cost function module that calculates assignment costs based on motion parameters, a prediction module that estimates future contact positions, and an optimization module that resolves combinatorial assignments. This segmentation allows each component to be optimized independently and implemented efficiently in hardware or software, reducing overall system complexity while maintaining high assignment accuracy.
Solution Approach 2:
The patent implements a dynamic cost function that adapts to the observed motion characteristics of contacts in real-time. The stochastic model parameters are updated based on recent movement patterns, allowing the system to adjust its expectations about contact behavior dynamically. This dynamic approach enables accurate tracking even when motion patterns change, while the modular architecture keeps implementation complexity manageable.
3Productivity
If the combinatorial assignment problem is solved using traditional distance-based methods, then the processing speed is fast, but the reliability of gesture detection deteriorates due to incorrect assignments
Solution Approach 1:
The patent performs preliminary prediction of contact positions and motion parameters before executing the combinatorial assignment. By pre-calculating expected contact locations and motion characteristics, the system narrows down the search space for the assignment algorithm, allowing it to focus computational resources on evaluating plausible assignments rather than all possible combinations. This preliminary action maintains processing speed while significantly improving assignment reliability.
Solution Approach 2:
The patent incorporates feedback mechanisms where the stochastic model continuously learns from assignment outcomes and adjusts its predictions accordingly. The system uses observed contact behavior to refine its understanding of motion patterns, which in turn improves future assignments. This feedback loop enhances gesture detection reliability over time while the efficient algorithmic structure maintains processing speed suitable for real-time operation.
Data Source
AI summary
A position tracking system may include a processor and a memory coupled to the processor. The memory may include instructions that, when loaded and executed by the processor, cause the processor to receive a plurality of measurements from a sensor and map a location of an object to a measurement using movement models. The mapping may be based upon a cost computed from movement models, the measurement, and the location.