Two-Eye Tracking with Electronic Contact Lens Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing eye tracking systems using electronic contact lenses face inaccuracies due to noise in sensor measurements, as they often rely on single-eye models that ignore correlations and constraints between left and right eye movements, leading to suboptimal tracking and noise filtration.
Innovation Solution
A system employing a pair of electronic contact lenses that utilize a two-eye orientation estimation model, applying filters like Kalman filters to jointly estimate orientations based on sensor measurements, correlations, and constraints, thereby improving accuracy and noise filtration by considering the limited and correlated motions of both eyes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single-eye models are used for eye tracking, then device complexity is reduced, but measurement precision deteriorates due to noise in sensor measurements
Solution Approach 1:
The patent combines data from both eyes into a unified tracking model. The system integrates sensor measurements from left and right electronic contact lenses, applying constraints that model the correlated motions between eyes to produce a single consistent estimate of head orientation, rather than treating each eye independently.
Solution Approach 2:
The system uses feedback from the known physiological constraints of eye movement to continuously refine tracking accuracy. By incorporating constraints about how eyes move together during head rotation, the model continuously corrects for sensor noise and provides more accurate head orientation estimates.
2Reliability
If single-eye models are used, then ease of operation is improved, but reliability deteriorates due to inability to filter noise effectively
Solution Approach 1:
The patent merges information from both eyes to improve tracking reliability. By combining sensor data from left and right lenses and applying physiological constraints, the system achieves more stable and noise-resistant tracking compared to single-eye models.
Solution Approach 2:
The system introduces physiological constraints as intermediary rules that mediate between raw sensor measurements and final tracking output. These constraints act as a filter that eliminates noise while preserving genuine eye movement patterns.
3Measurement precision
If two-eye models are used, then measurement precision is improved, but device complexity increases due to need to model correlations and constraints
Solution Approach 1:
The patent transforms the problem by changing parameters from independent eye orientations to head-centered coordinates. By modeling eye movements relative to head orientation and incorporating physiological constraints, the system achieves accurate head tracking while managing computational complexity through parameter transformation.
4Reliability
If two-eye models are used, then reliability is improved through noise filtration, but device complexity increases due to joint estimation requirements
Solution Approach 1:
The system uses feedback from physiological constraints to continuously refine noise filtration. By incorporating known relationships between eye movements and head orientation, the model provides ongoing correction that improves reliability while managing complexity through biologically-inspired constraints.
Data Source
AI summary
A system includes a pair of electronic contact lenses that obtain respective motion sensor measurements in response to eye movements. A tracking module derives estimated orientations for both eyes based on the sensor measurements and a set of correlations and constraints that describe human eye movement. The model describes the limited number of ways that an individual eye can move and relationships between relative movement of the left and right eye. The tracking module performs filtering based on the measurements and the eye model to suppress noise and generate orientation estimates for both eyes.