Method and apparatus for determining workload of motor vehicle driver

• US 6,061,610 A
• Filed: 10/31/1997
• Issued: 05/09/2000
• Est. Priority Date: 10/31/1997
• Status: Expired due to Term

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan Technical Center North America Incorporated (Nissan Motor Co., Ltd.)
• Inventors: Boer, Erwin R.
• Primary Examiner(s): Chin, Gary
• Application Number:
  US08/962,610
• Time in Patent Office: 921 Days
• Field of Search: 701/1, 701/41, 701/96, 701/301, 180/271, 180/272, 180/280, 340/575, 340/576
• US Class Current: 701/1
• CPC Class Codes: B60K 28/06 responsive to incapacity of driver
• CPC Class Codes: B60W 2050/0006 Digital architecture hierarchy
• CPC Class Codes: B60W 2050/0047 Digital-analogue (D/A) or analogue-digital (A/D) conversion
• CPC Class Codes: B60W 2050/005 Sampling
• CPC Class Codes: B60W 2050/0052 Filtering, filters
• CPC Class Codes: B60W 2520/10 Longitudinal speed
• CPC Class Codes: B60W 2540/10 Accelerator pedal position
• CPC Class Codes: B60W 2540/12 Brake pedal position
• CPC Class Codes: B60W 2540/18 Steering angle
• CPC Class Codes: B60W 2540/30 Driving style
• CPC Class Codes: B60W 40/09 Driving style or behaviour
• CPC Class Codes: G08B 21/06 indicating a condition of sleep, e.g. anti-dozing alarms

Abstract

A steering angle sensor monitors the steering angle of a vehicle as the vehicle is driven. The steering angle signal is input to a processing device which generates steering angle prediction errors based on the received steering angle. The processing device then generates a distribution of steering angle prediction errors. The distribution of prediction errors is compared to the distribution of steering angle prediction errors representing load-free driving. The comparison represents the driver'"'"'s workload.

Claims

1. A method for determining driver workload comprising:

monitoring a steering angle of a vehicle over time;

estimating a predicted steering angle of the vehicle over time based on the monitored steering angle;

generating steering angle prediction errors based on the predicted steering angle and the monitored steering angle at predetermined points in time; and

generating a workload value based on the steering angle prediction errors.

2. The method of claim 1, further comprising:

generating a distribution of said steering angle prediction errors.

3. The method of claim 2, wherein the step of generating a distribution of said steering angle prediction errors further comprises:

establishing a predetermined number of bins, each bin representing a range of possible steering angle prediction errors; and

categorizing the generated steering angle prediction errors into the predetermined number of bins.

4. The method of claim 3, wherein the step of generating a workload value further comprises:

generating, for each bin, a probability that each of the steering angle prediction errors will fall in each bin based on the number of the generated steering angle prediction errors in each bin.

5. The method of claim 4, further comprising:

generating an entropy value based on the probability for each bin.

6. The method of claim 5, further comprising:

comparing the generated entropy value with a predetermined entropy value to generate the workload value.

7. The method of claim 1, wherein the step of estimating the predicted steering angle comprises:

performing a Taylor series expansion of the monitored steering angle at predetermined times to generate predicted steering angles at the predetermined points in time.

8. A system for generating a workload value representing a driver'"'"'s workload comprising:

an input device configured to receive a steering angle signal representing the angle of a vehicle'"'"'s steering wheel; and

a workload monitoring device coupled to the input device and configured to;

receive the steering angle signal;

generate a predicted steering angle at predetermined times based on the received steering angle signal;

compare the predicted steering angle at the predetermined times to the steering angle signal to generate predicted steering angle errors; and

generate the workload value based on said errors.

9. A system for generating a workload value representing a driver'"'"'s workload comprising:

an input device configured to receive a steering angle signal representing the angle of a vehicle'"'"'s steering wheel; and

a workload monitoring device coupled to the input device and configured to;

receive the steering angle signal,

generate a predicted steering angle at predetermined times,

compare the predicted steering angle to the steering angle signal to generate predicted steering anile errors at the predetermined times,

generate a distribution of the predicted steering angle errors, and

generate the workload value based on said distribution.

10. The system of claim 9, wherein the workload monitoring device is further configured to:

establish a predetermined number of bins, each bin representing a range of possible steering angle prediction errors; and

categorize the steering angle prediction errors into the predetermined number of bins.

11. The system of claim 10, wherein the workload monitoring device is further configured to:

generate, for each bin, a probability that each of the steering angle prediction errors will fall in each bin based on the number of the steering angle prediction errors in each bin.

12. The system of claim 11, wherein the workload monitoring device is further configured to:

generate an entropy value based on the probability for each bin.

13. The system of claim 12, wherein the workload monitoring device is further configured to:

compare the generated entropy value with a predetermined entropy value to generate the workload value.

14. A system for generating a workload value representing a driver'"'"'s workload comprising:

an input device configured to receive a steering angle signal representing the angle of a vehicle'"'"'s steering wheel; and

a workload monitoring device coupled to the input device and configured to;

receive the steering angle signal,

perform a Taylor series expansion of the received steering angle signal at predetermined times to generate predicted steering angles at the predetermined times, and

generate a workload value based on the received steering angle signal.

15. The system of claim 14, wherein the workload monitoring device is further configured to:

compare the predicted steering angles to the steering angle signal at the predetermined times to generate the steering angle prediction errors.

16. A method for determining driver workload comprising:

monitoring a steering angle of a vehicle over a period of time;

estimating a predicted steering angle of the vehicle at least at predetermined intervals of time;

generating steering angle prediction errors based on the predicted steering angle and the monitored steering angle at the predetermined intervals of time;

generating a distribution of the steering angle prediction errors;

generating a first entropy value based on the distribution of the steering angle prediction errors;

comparing the first entropy value to a second entropy value based on steering angle prediction errors obtained under load-free driving conditions; and

generating a workload index based on the result of the comparing step.

17. The method of claim 16, further comprising:

updating the second entropy value based on the road geometry and environmental conditions.

18. The method of claim 16, wherein the step of generating a first entropy value includes:

computing the first entropy value based on a predetermined number of the steering angle prediction errors over a predetermined time.

19. A method for determining driver workload comprising:

monitoring a control variable of a vehicle over time;

estimating a predicted control variable of the vehicle over time based on the monitored control variable;

generating prediction errors, at predetermined intervals of time, based on the monitored control variable and the predicted control variable; and

generating a value representative of the driver workload based on the prediction errors.

20. The method of claim 19, wherein the monitoring step comprises monitoring at least one of a plurality of control variables comprising variables representing the vehicle'"'"'s acceleration, the vehicle'"'"'s braking, the vehicle'"'"'s wheel angle and the vehicle'"'"'s steering angle.

21. The method of claim 19, wherein the step of generating a workload value further comprises:

sorting said control variable prediction errors to create a distribution of control variable prediction errors; and

comparing the distribution to a predetermined distribution to generate the workload value.

22. A method for determining driver workload comprising:

monitoring the steering angle of the vehicle over time;

estimating a predicted steering angle of the vehicle over time;

generating steering angle prediction errors based on the predicted steering angle and the monitored steering angle at predetermined points in time; and

generating a workload value in response to the steering angle prediction errors.

23. A method as in claim 22, wherein the step of generating the workload value comprises calculating an entropy value using the steering angle prediction errors.

24. A method as in claim 23, wherein the step of generating the workload value further comprises comparing the calculated entropy value to a reference value for entropy.

25. A method as in claim 22, wherein the step of estimating comprises performing a Taylor series expansion of the monitored steering angle at the predetermined points in time.

26. A system for generating a workload value representing a driver'"'"'s workload comprising:

an input device configured to generate a steering angle signal representing the angle of a vehicle'"'"'s steering wheel; and

a workload monitoring device coupled to the input device and configured to;

receive the steering angle signal;

generate a predicted steering angle at predetermined intervals,

compare the predicted steering angle at the predetermined intervals to the steering angle signal to determine predicted steering angle errors, and

generate the workload value based on the predicted steering angle errors.

27. A system as in claim 26, wherein the workload monitoring device comprises a processor, a working memory and a program memory storing a program for the processor defining the configuration of the workload monitoring device.

28. A system as in claim 26, wherein the workload monitoring device is further configured to generate the predicted steering angle at predetermined intervals through processing the received steering angle signal.

29. A system as in claim 28, wherein the processing of the received steering angle signal comprises performing a Taylor series expansion of values of the steering angle signal at the predetermined intervals.

30. A system as in claim 26, wherein the workload monitoring device is further configured to calculate an entropy value from the predicted steering angle errors.

31. A system as in claim 30, wherein the workload monitoring device is further configured to generate the workload value from a comparison of the calculated entropy value to an entropy reference value.

32. A method for determining driver workload comprising:

monitoring a control variable on a vehicle;

estimating predictions of the control variable of the vehicle at intervals of time;

generating prediction errors at the intervals of time based on the predictions of the control variable and the monitored control variable; and

generating a value representative of the driver workload based on the prediction errors.

33. A system for generating a workload value representing a driver'"'"'s workload comprising:

an input device configured to monitor a control variable on a vehicle and generate a signal in response to the monitored control variable; and

a workload monitoring device coupled to the input device and configured to;

receive the control variable signal;

generate predictions of the control variable of the vehicle at intervals of time;

generate prediction errors at the intervals of time based on the predictions of the control variable and the control variable signal; and

generate a value representative of the driver workload based on the prediction errors.

34. A system as in claim 33, wherein the workload monitoring device is further configured to generate the predictions of the control variable based on processing of the control variable signal.

35. A system as in claim 33, wherein the workload monitoring device comprises a processor, a working memory and a program memory storing a program for the processor defining the configuration of the workload monitoring device.

36. A method for determining driver workload comprising:

monitoring a parameter of a state of the vehicle, over time;

estimating a predicted variable related to steering of the vehicle, over time, based on the monitored parameter;

generating prediction errors based on the predicted variable at intervals of time and actual values of a variable related to the steering of the vehicle; and

generating a value representative of the driver workload based on the prediction errors.

37. A method as in claim 36, wherein the step of generating prediction errors comprises:

deriving the actual values of the variable related to the steering of the vehicle from the monitoring of the parameter; and

comparing the derived actual values to values of the predicted variable at the intervals of time.