UNOBTRUSIVE DRIVER DROWSINESS DETECTION SYSTEM AND METHOD

• US 20100109881A1
• Filed: 11/05/2009
• Published: 05/06/2010
• Est. Priority Date: 11/05/2008
• Status: Active Grant

• Current Assignee: George Washington University
• Original Assignee: George Washington University
• Inventors: Eskandarian, Azim, Mortazavi, Ali
• Granted Patent: US 8,519,853 B2
• Time in Patent Office: Days
• US Class Current: 340/575
• CPC Class Codes: A61B 5/163 by tracking eye movement, gaze, or pupil change
• CPC Class Codes: A61B 5/18 for vehicle drivers or machine operators
• CPC Class Codes: A61B 5/6887 mounted on external non-worn devices, e.g. non-medical devices
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: B60K 28/066 actuating a signalling device B60K28/063 takes precedence
• CPC Class Codes: B60W 2540/18 Steering angle
• CPC Class Codes: B60W 2556/10 Historical data
• CPC Class Codes: G08B 21/06 indicating a condition of sleep, e.g. anti-dozing alarms
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A drowsiness detection system and method uses Empirical Mode Decomposition (EMD) signal processing to detect whether a vehicle driver is drowsy. The system uses a decomposed component of the steering wheel signal to extract specific features representing the steering control degradation phases. The system classifies the measured features into alert or drowsy state. The detection system is independent of the road geometry and automatically compensates steering control performance variability between drivers. The system is accurate in detecting the drowsy periods and drowsy-related lane departures. The detection system is unobtrusive and can be applied on-line.

Claims

1. A system for determining whether a driver of a vehicle is drowsy, the system comprising:

a steering wheel sensor for generating a steering wheel signal;

a processor for extracting at least one feature of the steering wheel signal based on an empirical mode decomposition, determining whether the driver is drowsy based on the at least one extracted feature, and generating an alarm signal in response to determining that the driver is drowsy; and

,

an alarm device for indicating that the driver is drowsy, or a safety mechanism for controlling the vehicle, wherein the alarm device or safety mechanism receives the alarm signal from said processor and is activated in response to the alarm signal.

2. The system of claim 1, wherein the steering wheel signal includes an angle of the steering wheel.

3. The system of claim 1, wherein the at least one extracted feature is independent of road curvature.

4. The system of claim 1, wherein the at least one extracted feature is representative of steering control degradation phases.

5. The system of claim 1, wherein the steering wheel signal includes a lane keeping waveform.

6. The system of claim 1, wherein the steering wheel signal includes a curve negotiation waveform.

7. The system of claim 1, wherein the at least one extracted feature comprises two features.

8. The system of claim 1, wherein the processor further automatically normalizes the at least one extracted feature based on a normal steering control behavior of the driver.

9. The system of claim 1, wherein the processor records a normal steering control behavior of the driver.

10. The system of claim 9, further comprising a manually-operated switch for activating the processor to record the normal steering control behavior of the driver.

11. The system of claim 1, wherein the processor generates the alarm signal if it is determined that the driver is drowsy.

12. The system of claim 1, wherein the determining is independent of road geometry.

13. The system of claim 1, wherein the determining automatically compensates between drivers.

14. The system of claim 1, wherein the processor uses empirical mode decomposition to decompose steering wheel signals into intrinsic oscillation modes enabling the analysis of the steering signal behavior independent of road geometry.

15. The system of claim 1, wherein if said processor detects a large amplitude steering angle correction followed by sporadic short intervals with no significant changes in steering angle, said processor determines that the driver is drowsy.

16. A system for determining whether a driver of a vehicle is drowsy, the system comprising:

a steering wheel sensor for generating a steering wheel angle signal;

a feature extraction module extracting a first feature and a second feature of the steering wheel signal based on an empirical mode decomposition intrinsic mode function 1;

a classifier for determining whether the driver is drowsy based on said first extracted feature and said second extracted feature; and

an alarm device for indicating that the driver is drowsy, or a safety mechanism for controlling the vehicle, wherein the alarm device or safety mechanism receives the alarm signal from said classifier and is activated in response to the alarm signal.

17. The system of claim 16, wherein said first feature is a standard deviation of the empirical mode decomposition intrinsic mode function 1 extrema, and said second feature is a standard deviation of a distance of zero crossing.

18. The system of claim 16, wherein the at least one extracted feature is representative of steering control degradation phases.

19. The system of claim 18, wherein the steering control degradation phases include drowsy and impaired.

20. The system of claim 16, wherein said feature extraction module and said classifier are implemented by a processor.

21. A system for determining whether a driver of a vehicle is drowsy, the system comprising:

a steering wheel sensor for generating a steering wheel signal;

a processor having a training mode and a detect mode, said processor extracting at least one feature of the steering wheel signal based on an empirical mode decomposition in both the training mode and the detect mode, said processor comparing the at least one feature extracted during the training phase with the at least one feature extracted during the detect phase, determining whether the driver is drowsy based on said comparison, and generating an alarm signal in response to determining that the driver is drowsy; and

,

an alarm device for indicating that the driver is drowsy, or a safety mechanism for controlling the vehicle, wherein the alarm device or safety mechanism receives the alarm signal from said processor and is activated in response to the alarm signal.