Driver Fatigue Warning System Using Road Curvature

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing driver-fatigue warning systems often issue false warnings when a driver is navigating curved roadways, as they incorrectly interpret the biased steering angle as driver fatigue, leading to unnecessary alerts and potential deactivation of the system.

Innovation Solution

A warning system that includes an angle-detector, a curvature-detection means (such as a digital map or camera), and a controller-circuit to determine an expected steering angle based on the road curvature, calculating steering-angle deviations and activating alerts only when the frequency of these deviations exceeds a threshold indicative of driver fatigue, thereby reducing false warnings.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the warning system monitors steering-angle deviations without considering road curvature, then it can detect driver fatigue, but it generates false warnings when driving on curved roadways

Engineering Contradiction:
Improveaccuracy of driver-fatigue detectionVSAvoidfalse warnings
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent introduces road curvature as an intermediary variable that mediates between the steering angle and the fatigue detection algorithm. By detecting road curvature separately (via GPS, map data, or vision systems) and using it to adjust the expected steering angle, the system distinguishes between steering deviations caused by road geometry versus those caused by driver fatigue, thereby eliminating false warnings while maintaining reliable fatigue detection

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically changes the reference parameter for steering angle evaluation based on road curvature. Instead of using a fixed zero-center reference, the expected steering angle is adjusted as a variable parameter that reflects the actual road geometry. This parameter change allows the system to adapt to different driving conditions and accurately distinguish between intentional steering corrections and fatigue-induced deviations

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If the system uses a fixed deviation threshold for steering angle, then the implementation is simple, but it cannot adapt to different road curvatures and speeds

Engineering Contradiction:
Improvesimplicity of threshold implementationVSAvoidadaptability to road conditions
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent transforms the static deviation threshold into a dynamic parameter that adapts to changing road conditions. The expected steering angle is calculated as a function of road curvature and vehicle speed, allowing the threshold to automatically adjust without requiring complex manual calibration. This dynamic approach maintains system simplicity while significantly improving adaptability to various driving scenarios

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs preliminary detection of road curvature and speed before evaluating steering angle deviations. By pre-calculating the expected steering angle based on upcoming road geometry and current vehicle state, the system prepares the adaptive threshold in advance, enabling real-time fatigue detection without adding computational complexity during the critical evaluation phase

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3528225B1Warning system
Publication Date: 2021.02.17 APTIV TECHNOLOGIES LTD
  • EP3528225B1 patent drawingFigure 1
  • EP3528225B1 patent drawingFigure 2
  • EP3528225B1 patent drawingFigure 3

AI summary

A warning system (10) includes an angle-detector (14), a curvature-detection-means (24), an alert-device (38), and a controller-circuit (32). The angle-detector (14) indicates a steering-angle (16) of a hand-wheel (18) of a host-vehicle (12). The curvature-detection-means (24) detects a curvature (26) of a portion of a roadway (28) traveled by the host-vehicle (12) characterized as ahead of the host-vehicle (12). The alert-device (38) is operable to alert an operator (20) of the host-vehicle (12) of driver-fatigue. The controller-circuit (32) is in communication with the angle-detector (14), the curvature-detection-means (24), and the alert-device (38). The controller-circuit (32) is configured to determine an average-steering-angle (40), determine an expected-steering-angle (42) based on the curvature (26) detected by the curvature-detection-means (24), determine a steering-angle-deviation (44) based on the average-steering-angle (40) and the expected-steering-angle (42), determine whether the steering-angle-deviation (44) is greater than a predetermined deviation-threshold (48), determine a frequency (50) of the steering-angle-deviation (44), and activate the alert-device (38) when the frequency (50) of the steering-angle-deviation (44) exceeds a change-threshold (52).