Method and system for recognizing driver impairment

• US 7,463,157 B2
• Filed: 05/30/2006
• Issued: 12/09/2008
• Est. Priority Date: 11/30/2003
• Status: Expired due to Fees

• Current Assignee: Volvo Technology Corporation
• Original Assignee: Volvo Technology Corporation
• Inventors: Victor, Trent, Larsson, Petter
• Primary Examiner(s): La; Anh V
• Application Number:
  US11/420,864
• Publication Number:
  US 20070008151A1
• Time in Patent Office: 924 Days
• Field of Search: 340/439, 340/575, 340/573.1, 340/576, 340/425.5, 180/272, 180/270, 180/269, 180/282, 280/735, 280/734, 280/736, 600/372, 600/390, 600/503, 701/1, 701/301
• US Class Current: 340/576
• CPC Class Codes: A61B 5/11 Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb for measuring pulse A61B5/02 ;A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721
• CPC Class Codes: A61B 5/1122 of movement trajectories
• 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: B60K 28/06 responsive to incapacity of driver
• CPC Class Codes: B60W 2556/50 of positioning data, e.g. GPS [Global Positioning System] data
• CPC Class Codes: G08B 21/06 indicating a condition of sleep, e.g. anti-dozing alarms

Abstract

A method and system for recognizing and/or detecting impairment and/or inattention of a person, especially a driver of a vehicle, and especially caused by drowsiness, distraction and/or workload, is disclosed by detecting and evaluating head movements of the person and/or head movement reactions of the person upon a disturbance exerted onto the vehicle.

Claims

1. Method for recognizing and/or detecting impairment and/or inattention of a person by detecting and evaluating head movements of the person on the basis of at least one component, the at least one component being at east one of three head rotation components, pitch, yaw, roll, and three head translation components, each in one of three orthogonal directions, the method comprising the following steps:

calculating within at least one predetermined time interval an actual density pattern of the at least one component in a form of an actual distribution of numbers of samples of the at least one component between at least one of a normal region and a normal region center point, which includes a peak of the density pattern, and a plurality of pre-set outer normal regions surrounding the at least one of the normal region and normal region center point;

comparing the actual density pattern with a reference density pattern of the at least one component in a form of a reference distribution of samples of the at least one component in the at least one of the normal region and normal region center point; and

evaluating a difference between at least one one of actual and reference patterns and actual and reference distributions and outputting a signal if the difference exceeds a predetermined value.

2. Method according to claim 1, wherein the reference density pattern is evaluated during at least one predetermined time interval in which the person has a reference level of impairment and/or inattention.

3. Method according to claim 1, wherein the reference density pattern is provided in a form of at least one threshold distribution which is pre-defined independently from the person as a standard for any person.

4. Method according to claim 1, wherein the regions have a form of a end of one of bins and funnels and the distribution is a normalized histogram of the signal components based on sizes of the one of the bins and funnels.

5. Method according to claim 1, wherein the at least one predetermined time interval is a sliding time window and wherein the actual density pattern of the at least one component is continuously calculated based on the samples within the sliding time window.

6. Method according to claim 1, wherein at least one threshold is predetermined for defining at least one normal region, respectively.

7. Method according, to claim 6, wherein the at least one threshold is determined with respect to a standard deviation from the normal region center point.

8. Method according to claim 6, wherein comparing the actual density pattern with the reference density pattern includes evaluating a difference between both the actual density pattern and the reference density pattern in at least one of different normal regions and segments of at least one normal region.

9. Method according to claim 1, comprising recognizing and/or detecting workload of a person by detecting and evaluating head movements of the person based on at least one of an amount and the time dependent derivative of at least one of three head rotation components and three head translation components.

10. System for conducting, a method according to claim 1.

11. A computer medium that contains a program comprising computer program code means adapted to perform a method according to claim 1 when the program is run on a programmable microcomputer.

12. A computer medium that contains a program according to claim 11 adapted to be downloaded to a system when run on a computer which is connected to the internet, the system being adapted to perform a method for recognizing and/or detecting impairment and/or inattention of a person by detecting at and evaluating head movements of the person on the basis of at least one component, the at least one component being at least one of three head rotation components, pitch, yaw, roll, and three head translation components, each in one of three orthogonal directions, the method comprising the following steps:

calculating within at least one predetermined time interval an actual density pattern of the at least one component in a form of an actual distribution of numbers of samples of the at least one component between at least one of a normal region and a normal region center point, which includes a peak of the density pattern, and a plurality of pre-set outer normal regions surrounding the at least one of the normal region and normal region center point;

comparing the actual density pattern with a reference density pattern of the at least one component in a form of a reference distributions of samples of the at least one component in the at least one of the normal region and normal region center point; and

evaluating difference between at least one one of actual and reference patterns and actual and reference distributions and outputting a signal if the difference exceeds a predetermined value.

13. A computer medium that contains a program as set forth in claim 11, wherein the program is stored on a computer usable medium.

14. Method for recognizing and/or detecting impairment and/or inattention of a person who is at least one of actively and reactively connected with an object, wherein the object is at least one of influenced and affected by an external disturbance which causes a head movement reaction of the person, comprising the following steps:

detecting the disturbance exerted onto the object by external forces;

detecting head movement reaction of the person upon the disturbance based on at least one component, the at least one component being at least one of three head rotation components and three head translation components;

evaluating a difference between the head movement reaction and at least one of a predefined reference head movement reaction and a stored previous head movement reaction upon a previous disturbance; and

outputting a disturbance signal if the difference exceeds a predefined value.

15. Method according to claim 14, wherein the object is a vehicle and wherein detecting the disturbance includes measuring of at least one of the following values:

dynamic properties of objects within the vehicle;

an acceleration force acting on the person;

cabin air-suspension and tilt angle compared to a rest of the vehicle; and

steering wheel movements.

16. Method according to claim 14, wherein detecting the head movement reaction includes at least one of the following measures and evaluations:

pre-defining a threshold value for the disturbance signal and evaluating head movement reactions if the disturbance exceeds the threshold value;

evaluating head movements if the vehicle is travelling substantially in a straight direction for substantially at least one of a predefined period of time and above a predefined speed threshold value; and

ignoring head movements if the vehicle is not travelling substantially in a straight direction for substantially the at least one of the predefined period of time and at above the predefined speed threshold value.

17. Method according to claim 16, wherein whether the vehicle travels substantially in a straight direction is evaluated based on a navigation/localization system and data.

18. Method according to claim 17, wherein whether the vehicle travels substantially in a straight direction is evaluated based on GPS- and map-matching data.

19. Method according to claim 16, wherein speed of the vehicle is evaluated based on a navigation/localization system and data.

20. Method according to claim 19, wherein speed of the vehicle is evaluated based on GPS- and map-matching data.

21. Method according to claim 16, wherein detecting head movement reaction is at least one of activated and deactivated based on a navigation/localization system and data.

22. Method according to claim 21, wherein detecting head movement reaction is at least one of activated and deactivated based on GPS- and map-matching data.

23. Method according to claim 14, wherein evaluation of the difference includes at least one of the following measures:

a difference between the reaction and the normal region center point;

a RMS (root mean square) value of a difference between the reaction and the normal region center point; and

an error between a model for a head movement in response to a disturbance and the measured head movement in response to the same disturbance.

24. Method according to claim 14, comprising recognizing and/or detecting workload of a person by detecting and evaluating head movements of the person based on at least one of an amount and the time dependent derivative of at least one of three head rotation components and three head translation components.

25. System for conducting a method according to claim 14.

26. Method for recognizing and/or detecting impairment and/or inattention of a person based on a workload of the person by detecting and evaluating head movements of the person based on at least one of an amount and the time-dependent derivative of at least one component, the at least one component being at least one of three head rotation components and three head translation components, comprising the following steps:

calculating a standard deviation and at least one of a root mean square value of the at least one component and a root mean square value of a time dependent derivative of the at least one component over a sliding time interval for producing a continuous quantification of workload, and

comparing the quantified workload with at least one threshold and outputting a signal if the difference exceeds the at least one of the threshold.

27. System for conducting a method according to claim 26.

28. Method according to claim 26, for recognizing and/or detecting distraction of the person on the basis the quantified workload.

29. Method according to claim 26, wherein the person in a driver of a vehicle.

30. Computer medium comprising computer program code means adapted to perform a method according to claim 26 when the program is run on a programmable microcomputer.

31. Computer medium according to claim 30, wherein the program is adapted to be downloaded to a system when run on a computer which is connected to the internet, the system being adapted to perform a method for recognizing and/or detecting impairment and/or inattention of a person by detecting and evaluating head movements of the person on the basis of at least one component, the at least one component being at least one of three head rotation components, pitch, yaw, roll, and three head translation components, each in one of three orthogonal directions, the method comprising the following steps:

calculating within at least one predetermined time interval an actual density pattern of the at least one component in a form of an actual distribution of numbers of samples of the at least one component between at least one of a normal region and a normal region center point, which includes a peak of the density pattern, and a plurality of pre-set outer normal regions surrounding the at least one of the normal region and normal region center point;

comparing the actual density pattern with a reference density pattern of the at least one component in a form of a reference distribution of samples of the at least one component in the at least one of the normal region and normal region center point; and

evaluating a difference between at least one one of actual and reference patterns and actual and reference distributions and outputting a signal if the difference exceeds a predetermined value.

32. A computer medium that contains a program as set forth in claim 30, wherein the program is stored on a computer usable medium.