Method and system for intention estimation and operation assistance

• US 20080147249A1
• Filed: 02/08/2008
• Published: 06/19/2008
• Est. Priority Date: 12/16/2003
• Status: Active Grant

• Current Assignee: Drexel University
• Original Assignee: Dario Salvucci, Nobuyuki Kuge, Tomohiro Yamamura
• Inventors: Kuge, Nobuyuki, Yamamura, Tomohiro, Salvucci, Dario
• Granted Patent: US 7,778,742 B2
• Time in Patent Office: Days
• US Class Current: 701/1
• CPC Class Codes: B60T 2201/08 Lane monitoring; Lane Keeping Systems
• CPC Class Codes: B60T 2201/086 using driver related features
• CPC Class Codes: B60W 40/09 Driving style or behaviour
• CPC Class Codes: B60W 50/10 Interpretation of driver requests or demands

Abstract

An intention estimation and operation assistance system for estimating an operator'"'"'s intention and providing assistance to operation of a machine. Reference data, such as the operation of a plurality of hypothetical operators, is provided for comparison with the operation of a real operator, to determine an estimated intention of the operator. The estimated operator is utilized to alter the operation of the machine, to provide a safer and/or smoother operation experience in operating the machine.

Claims

1-10. -10. (canceled)

11. A method for estimating a driver'"'"'s intention comprising the steps of:

detecting an operation of a real driver of a vehicle;

calculating an operation of each of a plurality of imaginary drivers to accomplish a respective associated intention according to a real-time vehicle driving condition and a respective associated intention of each imaginary driver;

calculating a respective likelihood value for each of the plurality of imaginary drivers based on the detected operation of the real driver and the operation of each of the plurality of imaginary drivers; and

determining an estimated intention of the real driver based on the respective likelihood value of each of the plurality of imaginary drivers with the respective associated intention.

12. The method of claim 11, wherein the determining step comprises the steps of:

selecting one of the plurality of imaginary drivers based on the respective likelihood values; and

determining the estimated intention of the real driver as the intention of the selected one of the plurality of imaginary drivers.

13. The method of claim 11, wherein the operation of each of the plurality of imaginary drivers is a lane-keeping intention, a lane-changing intention to the right or a lane-changing intention to the left.

14. The method of claim 11, wherein:

the detecting step detects an operation of the real driver at each one of different points in time;

each respective likelihood value for each of the plurality of imaginary drivers is generated based on partial likelihood values of each of the plurality of imaginary drivers at the different points in time; and

each of the partial likelihood values is respectively associated with an operation of one of the plurality of imaginary drivers at each one of the different points in time, and is generated based on the respective detected operation of the real driver at each one of the different points in time and the respective operation of one of the plurality of imaginary drivers at each one of the different points in time.

15. The method of claim 14, wherein each respective likelihood value for each of the plurality of imaginary drivers is a respective summation of partial likelihood values of each of the plurality of imaginary drivers at the different points in time.

16. The method of claim 14 further comprising the step of generating data related to at least one additional imaginary driver at a selected point in time, based on the intention of a first one of the plurality of imaginary drivers at a point of time preceding the selected point of time, wherein each of the at least one additional imaginary driver has an associated intention and operation at each respective one of the different points in time.

17. The method of claim 16, wherein the intention of each of the at least one additional imaginary driver at the selected point of time is different from the intention of the first one of the plurality of imaginary drivers at the selected point of time.

18. The method of claim 16, wherein each additional imaginary driver assumes the associated intention of the first one of the plurality of imaginary drivers at each point in time preceding the selected point in time.

19. The method of claim 16, wherein for each of the imaginary drivers, the second device is configured to generate an intention corresponding to a new point in time based on the respective intention of each of the imaginary drivers at a point in time preceding the new point in time.

20. The method of claim 16, wherein the second device is configured to eliminate a selected one of the plurality of imaginary drivers based on an intention of the selected one of the plurality of imaginary drivers at a specific point in time and an operation status of the vehicle.

21-24. -24. (canceled)

25. A method for estimating an operator'"'"'s intention comprising the steps of:

detecting an operation performed by an operator;

calculating an operation of each of a plurality of imaginary operators to accomplish a respective associated intention according to a real-time operating condition and a respective associated intention of each imaginary driver;

calculating a respective likelihood value for each of the plurality of imaginary operators, based on the detected operation performed by the operator and the operation of each of the plurality of imaginary operators; and

determining an estimated intention of the operator based on the respective likelihood value of each of the plurality of imaginary operators with the respective associated intention.

26. A machine-readable medium bearing instructions for estimating an operator'"'"'s intention, the instructions, upon execution by a data processing system, causing the data processing system to perform the steps of:

detecting an operation performed by an operator;

calculating an operation of each of a plurality of imaginary operators to accomplish a respective associated intention according to a real-time operating condition and a respective associated intention of each imaginary driver;

calculating a respective likelihood value for each of the plurality of imaginary operators based on the detected operation performed by the operator and the operation of each of the plurality of imaginary operators; and

determining an estimated intention of the real operator based on the respective likelihood value of each of the plurality of imaginary operators with the respective associated intention.

27. A method for estimating an intention of a driver to operate a vehicle, the method comprising:

detecting an operation of a real driver that operates a vehicle;

calculating an operation of each of a plurality of imaginary drivers to accomplish a respective associated intention according to a real-time vehicle driving condition and the respective associated intention;

calculating a respective likelihood value for each of the plurality of imaginary drivers based on the calculated operation of the imaginary driver and the detected operation of the real driver;

selecting one of the plurality of imaginary drivers based on the calculated respective likelihood value of each of the plurality of imaginary drivers; and

determining the respective associated intention of the selected one imaginary driver as an intention of the real driver.

28. The method of claim 27, wherein the respective likelihood value is the probability of occurrence population expressed by a normal distribution, and the mean is the detected operation of the real driver and the variance is a predetermined value.