Driver Drowsiness Detection Using Steering and Lane Position Indices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver monitoring systems for road vehicles face accuracy issues due to calibration based on statistical and non-punctual user data, failing to account for individual differences and boundary conditions, leading to potential false positives or negatives in detecting drowsiness and attention levels.
Innovation Solution
A method that uses sensor devices to detect steering and line parameters, processing a steering index and line index to calculate a driver's tiredness/attention level, incorporating a sensitivity function to consider boundary conditions and calibrate based on individual characteristics, and alerting the driver when the attention level exceeds a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If statistical and non-punctual user data are used for calibration, then the system can be implemented with existing data, but the measurement precision of driver drowsiness detection deteriorates due to individual differences
Solution Approach 1:
The system performs a preliminary characterization phase during which it collects and analyzes the specific driver's steering behavior, lane position patterns, and driving style. This preliminary action creates a personalized baseline that is then used for accurate drowsiness detection, resolving the contradiction by preparing individualized calibration data before actual monitoring begins.
Solution Approach 2:
The system changes the calibration approach from using fixed statistical parameters to dynamically adapting parameters based on the specific driver's behavior. By monitoring individual steering corrections, lane position variations, and driving patterns, the system adjusts its detection thresholds and parameters to match each driver's unique characteristics, thereby improving measurement precision.
2Ease of operation
If standard monitoring thresholds are applied to all drivers, then the system operation is simplified, but false positives and negatives increase due to different driving styles
Solution Approach 1:
The system transitions from static, fixed thresholds to dynamic, adaptive thresholds that automatically adjust based on the driver's observed behavior. The monitoring parameters evolve over time as the system learns the driver's normal patterns, allowing the same operational framework to adapt to individual differences without increasing operational complexity.
Solution Approach 2:
The system implements continuous feedback loops where detection results and driver responses are used to refine and adjust monitoring parameters. This feedback mechanism allows the system to learn from each driving session, improving reliability by correcting false positives and negatives while maintaining simple automated operation.
3Device complexity
If boundary conditions are not considered, then the detection system remains simple, but the reliability of drowsiness assessment deteriorates due to psychophysical variations
Solution Approach 1:
The system performs preliminary characterization of the driver's normal behavior patterns under various boundary conditions such as time of day, traffic conditions, and route characteristics. This preliminary phase establishes a baseline that accounts for psychophysical variations, allowing the system to distinguish between normal variations and actual drowsiness signs.
Solution Approach 2:
The system dynamically adjusts detection parameters based on observed boundary conditions and environmental factors. By monitoring changes in driving patterns relative to external conditions, the system adapts its sensitivity and thresholds to maintain reliable assessment across different operating contexts without requiring complex manual configuration.
Data Source
Figure 1
Figure 2
Figure 3a)~3b)
AI summary
Method for determining the drowsiness/attention level of the driver (DR) of a road vehicle (1); the method comprises the steps of: detecting at least one steering parameter (S) related to the kinematics of a driver-operated device (DR) for steering; where the steering parameter (S) is related to the steering angle (δ) and/or the steering speed (ω) of the road vehicle (1); elaborating a steering index (IS) as a function of the at least one steering parameter (S) detected; detecting at least one line parameter (L) related to the position (P) of the road vehicle (1) within a lane (8); processing a line index (IL) as a function of the at least one line parameter (L) detected; computing a driver (DR) drowsiness/attention index (IDA) as a function of a combination of the steering index (IS) and the line index (IL) .