Adaptive Sleepiness Detection Using Dynamic VOR Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in accurately determining sleepiness in real environments, particularly in vehicles, due to complex eye movements and varying head and eye movement patterns, which are not constant and often do not induce vestibulo-ocular reflex movements effectively.
Innovation Solution
A data processing device that calculates pupil and head movements, evaluates the suitability of the environment for vestibulo-ocular reflex movement calculation, and selects between different techniques for sleepiness determination based on a suitability degree, including saccadic movement and eyelid movement calculations when vestibulo-ocular reflex is not suitable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vestibulo-ocular reflex movement is used to calculate sleepiness, then sleepiness determination can be performed in controlled experimental environments, but measurement precision deteriorates in actual vehicle environments due to complex eye movements and varying head positions
Solution Approach 1:
The system dynamically switches between different sleepiness calculation methods (vestibulo-ocular reflex-based method and alternative method) based on the detected environment. The evaluation unit continuously monitors environmental characteristics and selects the appropriate calculation method, allowing the system to adapt to varying conditions in real-time rather than using a fixed single method.
Solution Approach 2:
The system changes the operational parameters by selecting different calculation methods based on environmental suitability. When the vestibulo-ocular reflex movement cannot be reliably detected due to complex eye movements or lack of head movement, the system switches to alternative parameters (such as eyelid movement or other eye movement characteristics) to maintain accurate sleepiness determination.
2Adaptability or versatility
If multiple eye movement types are present in actual vehicle environment, then environmental complexity increases, but sleepiness calculation accuracy decreases due to difficulty in isolating vestibulo-ocular reflex movement
Solution Approach 1:
The evaluation unit segments the eye movement components by identifying and separating the vestibulo-ocular reflex movement from other eye movements (saccadic movements, convergence movements, etc.). By analyzing the characteristics of different eye movement types and isolating the relevant VOR components, the system can accurately calculate sleepiness even when multiple movement types are present simultaneously.
Solution Approach 2:
The evaluation unit acts as an intermediary that mediates between the raw eye movement data and the sleepiness calculation. It processes and filters the complex eye movement signals, identifying suitable segments for VOR-based calculation and switching to alternative methods when necessary, thereby maintaining accuracy despite environmental complexity.
3Measurement precision
If head movement is not sufficiently induced in actual vehicle environment, then vestibulo-ocular reflex movement cannot be reliably detected, but sleepiness determination cannot be performed with high accuracy
Solution Approach 1:
The system dynamically adapts its operation based on head movement detection. When head movement is insufficient to induce reliable vestibulo-ocular reflex movement, the system switches to alternative sleepiness calculation methods that do not depend on head movement, maintaining operational capability without requiring artificial head vibration or specific head movement conditions.
Solution Approach 2:
The system uses the natural head movements and eye movements that occur during normal driving conditions as the basis for sleepiness determination, rather than requiring external induction of head movement. The evaluation unit leverages the existing movement data from the sensors to perform accurate sleepiness calculation without additional experimental manipulation.
Data Source
AI summary
A data processing device that performs data processing of monitoring a person, the data processing device includes: a calculator configured to calculate pupil movement and head movement of the person; an evaluator configured to evaluate a suitability degree of a situation in calculating the VOR; a provision unit configured to provide the suitability degree evaluated by the evaluator to data calculated by the calculator; a selector configured to select a first technique of calculating the sleepiness based on the VOR of the person or a second technique different from the first technique based on the suitability degree provided to the data; a first sleepiness calculator configured to calculate the sleepiness based on the first technique when the selector selects the first technique; and a second sleepiness calculator configured to calculate the sleepiness based on the second technique when the selector selects the second technique.