ADAPTIVE VEHICLE CONTROL SYSTEM WITH INTEGRATED MANEUVER-BASED DRIVING STYLE RECOGNITION

• US 20100023183A1
• Filed: 07/24/2008
• Published: 01/28/2010
• Est. Priority Date: 07/24/2008
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations Incorporated (Motors Liquidation Co. GUC Trust)
• Inventors: Huang, Jihua, Chin, Yuen-Kwok, Lin, William C.
• Granted Patent: US 8,280,601 B2
• Time in Patent Office: Days
• US Class Current: 701/1
• CPC Class Codes: B60W 30/12 Lane keeping
• CPC Class Codes: B60W 40/09 Driving style or behaviour

Abstract

A vehicle control system that classifies a driver'"'"'s driving style based on characteristic maneuvers. The system includes a plurality of vehicle sensors that detect various vehicle parameters. A maneuver identification processor receives the sensor signals to identify a characteristic maneuver of the vehicle and provides a maneuver identifier signal of the maneuver. A style characterization processor receives the maneuver identifier signals, sensor signals from the vehicle sensors and the traffic and road condition signals, and classifies driving style based on the signals to classify the style of the driver driving the vehicle. The classification of the driver style can be provided for a level-1 combination that combines the classification results from different maneuver type classifiers for a single maneuver, a level-2 combination that combines the classification results from multiple maneuvers that are of the same type and a level-3 combination that combines the classification results from different types of characteristic maneuvers.

Claims

1. A system for classifying a driver'"'"'s driving style of a vehicle, said system comprising.

a plurality of vehicle sensors providing sensor measurement signals;

a maneuver identification processor responsive to the sensor signals from the plurality of vehicle sensors, said maneuver identification processor identifying different characteristic maneuvers of the vehicle and providing maneuver identifier signals identifying the characteristic maneuvers; and

a style characterization processor responsive to the maneuver identifier signals and the sensor measurement signals, said style characterization processor including separate classifiers for each different type of maneuver, said classifiers being selected based on the maneuver identifier signal to classify driving style using sensor signals corresponding to a particular maneuver, each classifier providing a driver style classification signal that identifies the type of driver driving the vehicle for the particular maneuver, said style characterization processor further including a first combiner that receives and combines the style classification signals from each of the selected classifiers and provides a combined style classification signal for a single one of several different maneuvers.

2. The system according to claim 1 further comprising a decision fusion processor responsive to the driver style classification signals from the different classifiers in the style characterization processor.

3. The system according to claim 2 wherein the decision fusion processor includes a second combiner that combines the style classification signals for multiple maneuvers that are of the same type.

4. The system according to claim 3 wherein the decision fusion processor further includes a third combiner that combines style classification signals from the second combiner for different types of maneuvers.

5. The system according to claim 1 wherein the first combiner employs a combination process selected from a group comprising voting, sum, mean, median, product, maximum/minimum, fuzzy integral, Dempster-Shafter, mixture of local experts and neural networks.

6. The system according to claim 1 wherein the first combiner employs a Dempster-Shafter process.

7. The system according to claim 1 further comprising a traffic and road condition recognition processor responsive to the sensor signals, said traffic and road condition recognition processor providing traffic and road condition signals to the style characterization processor.

8. The system according to claim 1 further comprising a style profile trip-logger responsive to the maneuver identifier signals and the style classification signals, said trip-logger storing the style classification signals.

9. The system according to claim 1 wherein the characteristic maneuvers include vehicle following maneuvers, lane-change maneuvers, U-turn maneuvers, left/right-turn maneuvers, highway on/off ramp maneuvers, passing maneuvers, vehicle launching maneuvers and curve-handling maneuvers.

10. The system according to claim 1 wherein the style characterization processor includes an original features processor for deriving original features in the maneuver identifier signals, a feature extraction processor for extracting features from the original features, a feature selection processor for selecting certain ones of the extracted features and a classifier for classifying the maneuver using the selected features.

11. The system according to claim 1 wherein the classifiers employ a process of feature extraction and feature selection to classify the driving style.

12. The system according to claim 1 wherein the classifiers use discriminant features to classify the driving style.

13. A system for classifying a driver'"'"'s driving style of a vehicle, said system comprising:

a plurality of vehicle sensors providing sensor measurement signals;

a maneuver identification processor responsive to the sensor signals from the plurality of vehicle sensors, said maneuver identification processor identifying different characteristic maneuvers of the vehicle and providing maneuver identifier signals identifying the characteristic maneuvers;

a style characterization processor responsive to the maneuver identifier signals and the sensor measurement signals, said style characterization processor classifying driving style based on the identifier and sensor signals and providing driving style classification signals that identify the type of driver driving the vehicle; and

a plurality of combiners that combine the classification signals for different maneuvers.

14. The system according to claim 13 wherein the style characterization processor includes a plurality of classifiers where a separate classifier classifies driving style for each different maneuver.

15. The system according to claim 14 wherein the style characterization processor includes a combiner that receives and combines the style classification signals from each of the classifiers and provides a combined style classification signal for a single one of several different maneuvers.

16. The system according to claim 13 wherein the plurality of combiners includes a first combiner that combines the style classification signals for multiple maneuvers that are of the same type.

17. The system according to claim 16 wherein the plurality of combiners includes a second combiner that combines style classification signals from the second combiner for different types of maneuvers.

18. The system according to claim 13 wherein the style characterization processor includes an original features processor for identifying original features from the maneuver identifier signals, a feature extraction processor for extracting features from the original features, a feature selection processor for selecting certain ones of the extracted features and a classifier for classifying the maneuver using the selected features.

19. The system according to claim 13 wherein the classifiers use discriminant features to classify the driving style.