Drowsiness Assessment Using Base Value Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for assessing driver drowsiness rely solely on instantaneous features, such as PERCLOS and blinking-related values, which can lead to abrupt changes in estimated sleepiness values due to measurement inaccuracies or sudden situations, failing to account for the continuous development of drowsiness over time.
Innovation Solution
A method that determines drowsiness values cyclically over defined time intervals by considering drowsiness characteristics, including eyelid-opening signals and other features, and incorporates a drowsiness base value from previous time intervals to provide a more accurate and reliable assessment, using processes like multiple linear regression and artificial neural networks for evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If drowsiness assessment is based solely on instantaneous features (PERCLOS, blinking-related values), then the system responds quickly to current state, but the estimated sleepiness values exhibit abrupt changes due to measurement inaccuracies or sudden situations
Solution Approach 1:
The system performs preliminary assessment in prior time intervals to establish a drowsiness base value, which is then carried forward to the current time interval. This preliminary action provides a stable reference that prevents abrupt changes in the estimated sleepiness value, resolving the contradiction between quick response and stability.
Solution Approach 2:
The drowsiness base value from prior time intervals is fed back into the current assessment as an input feature. This feedback mechanism ensures that the current estimated sleepiness value is influenced by previous assessments, smoothing out abrupt changes while maintaining accuracy through the combination of instantaneous features and historical context.
2Measurement precision
If drowsiness value is determined only from current time interval data, then the assessment reflects immediate state, but it fails to account for continuous development of drowsiness over time
Solution Approach 1:
The system maintains continuous assessment across multiple time intervals by carrying forward the drowsiness base value. This continuity ensures that the temporal development of drowsiness is captured, as each assessment builds upon previous ones, preventing loss of information about the evolving drowsiness state while still reflecting the immediate condition.
3Reliability
If multiple time intervals are considered with drowsiness base value, then the assessment becomes more stable and reliable, but the system complexity increases
Solution Approach 1:
The system uses its own previous output (drowsiness base value) as an input for the current assessment. This self-service approach allows the system to incorporate temporal information without requiring external complex processing, maintaining robustness while avoiding excessive complexity by leveraging its own historical data.
Data Source
AI summary
A method for determining a drowsiness value that represents the drowsiness of a driver of a motor vehicle and is determined cyclically over defined time intervals on the basis of at least one drowsiness characteristic, includes the task of determining a drowsiness value of a time interval to be assessed, in addition to the drowsiness characteristic, at least one drowsiness base value is taken into account.