Steering Wheel Grip Prediction for Driver Dozing Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driver assistance systems are limited in detecting drowsiness as they rely solely on hands-off alerts, failing to identify dozing when the driver grips the steering wheel and often produce unnecessary alerts when the driver is not gripping it.
Innovation Solution
A driver assistance system that includes a gripping detection sensor on the steering wheel to acquire gripping information, such as contact area, position, shape, and pressure, and a controller that matches this information with vehicle behavior data to predict and determine if the driver is dozing, using machine learning algorithms to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If hands-off alert is used to detect driver dozing, then alert function is provided, but false alerts occur when driver is not gripping the steering wheel and dozing is not detected when driver is gripping the steering wheel
Solution Approach 1:
The patent combines multiple detection parameters (grip detection, steering angle, yaw rate, lateral/longitudinal acceleration) into a unified dozing detection system. By merging these different data sources and analyzing them together through pattern recognition, the system achieves more reliable detection than any single parameter could provide alone, thereby reducing false alerts while maintaining high detection accuracy.
2Measurement precision
If gripping detection sensor is used to detect steering wheel grip, then grip information is acquired, but dozing cannot be detected when driver grips the steering wheel
Solution Approach 1:
The patent transitions from one-dimensional grip detection (only detecting presence/absence of grip) to multi-dimensional analysis by incorporating steering angle, yaw rate, and acceleration data. This dimensional expansion allows the system to detect dozing states even when grip is present, as the additional dimensions reveal patterns inconsistent with attentive driving.
Solution Approach 2:
The system continuously monitors multiple parameters and uses feedback loops to compare actual driving behavior against expected patterns. When discrepancies are detected (such as unexpected steering patterns or acceleration variations while gripping the wheel), the system identifies potential dozing states, enabling detection beyond simple grip presence.
3Reliability
If multiple parameters are analyzed to improve dozing detection accuracy, then detection reliability is improved, but system complexity increases
Solution Approach 1:
The controller is designed as a multi-functional unit that handles grip detection, steering angle measurement, yaw rate detection, and acceleration sensing through integrated sensors. This universal approach consolidates multiple detection functions into a single control system, reducing overall system complexity while maintaining the ability to analyze multiple parameters simultaneously for improved dozing detection.
Data Source
AI summary
Disclosed herein is a driver assistance system including a gripping detection sensor provided on a steering wheel of a vehicle and configured to detect a driver gripping the steering wheel and acquire gripping information, and a controller configured to receive the gripping information from the gripping detection sensor, receive behavior information from a behavior detection device configured to acquire the behavior information of the vehicle, and determine whether the driver is dozing, wherein the controller is configured to match the gripping information according to the behavior information, predict gripping information according to current behavior information based on a result of the matching to generate gripping prediction information, and compare current gripping information with the gripping prediction information according to the current behavior information to determine whether the driver is dozing.