Modifying autonomous vehicle driving by recognizing vehicle characteristics

• US 9,731,713 B2
• Filed: 09/10/2014
• Issued: 08/15/2017
• Est. Priority Date: 09/10/2014
• Status: Active Grant

• Current Assignee: Volkswagen AG
• Original Assignee: Volkswagen AG
• Inventors: Horii, Hiroshi
• Primary Examiner(s): Do, Truc M
• Assistant Examiner(s): WHITTINGTON, JESS G
• Application Number:
  US14/482,800
• Publication Number:
  US 20160068156A1
• Time in Patent Office: 1,070 Days
• CPC Class Codes: B60W 30/00 Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
• CPC Class Codes: B60W 30/0956 the prediction being responsive to traffic or environmental parameters
• CPC Class Codes: B60W 30/18163 Lane change; Overtaking manoeuvres
• CPC Class Codes: B60W 40/04 Traffic conditions

Abstract

A system for controlling an autonomous vehicle having cameras for obtaining image data of neighboring vehicles in a proximity of the autonomous vehicle and for identifying physical characteristics of the neighboring vehicles, including makes, models, and trims of neighboring vehicles, colors of the neighboring vehicles, and exposed cargo being carried by the neighboring vehicles. The system maps the identified physical characteristics into predicted potential on-road events in the proximity of the autonomous vehicle, and makes driving decisions based on the predicted potential on-road event.

Claims

1. A system for controlling an autonomous vehicle, comprising:

one or more cameras for obtaining image data of neighboring vehicles in a proximity of the autonomous vehicle;

a processor and a non-transitory data storage on which is stored computer code which, when executed on the processor, causes the autonomous driving system to identify, based on comparing the obtained image data with stored sample image data for a plurality of vehicles of particular makes and models, physical characteristics of the neighboring vehicles, wherein the physical characteristics include;

classes, makes, models, and trims of the neighboring vehicles;

colors of the neighboring vehicles; and

exposed cargo being carried by the neighboring vehicles;

means for mapping the identified physical characteristics into predicted potential on-road events in the proximity of the autonomous vehicle;

means for determining driving decisions based on the predicted potential on-road events; and

vehicle controls that implement the determined driving decisions, wherein the means for mapping the identified physical characteristics into predictions includes a database that;

associates classes, makes, models, and trims of vehicles with respective expected driving patterns for vehicles of those makes, models, and trims;

associates colors of vehicles with respective expected driving patterns for vehicles of those colors; and

associates certain vehicle makes and models with aggressive driving patterns;

wherein the respective expected driving patterns are further defined by road type and road condition;

wherein the means for determining driving decisions includes software that adopts a defensive driving profile upon a determination that a vehicle having an aggressive driving pattern is in the proximity of the autonomous vehicle;

wherein the defensive driving profile is a driving profile that accounts for neighboring vehicles that will distract other drivers and a prediction of how the other drivers will drive in response to the distraction.

2. The system of claim 1, wherein the defensive driving profile includes increasing the space between the autonomous vehicle and the vehicle having one of the aggressive driving patterns.

3. The system of claim 1, wherein the computer code causes the autonomous driving system to identify the physical characteristics of vehicles in left and right neighboring lanes;

and

the means for determining driving decisions includes software that selects either the left or the right neighboring lane for a lane change by assessing, based on the identified physical characteristics, which of the neighboring lanes will be safer.

4. The system of claim 1, wherein classes of the physical characteristics include at least a sports car, a sedan, a sport utility vehicle, and a truck.

5. A system for automatically driving an autonomous vehicle, comprising:

one or more sensors for obtaining data about at least one neighboring vehicle in a proximity of the autonomous vehicle;

a processor and a non-transitory data storage on which is stored computer code which, when executed on the processor, causes the autonomous driving system to;

identify, based on the obtained data, physical characteristics of the neighboring vehicle;

map the identified physical characteristics into predicted potential on-road events in the proximity of the autonomous vehicle; and

determine driving decisions based on the predicted potential on-road events; and

vehicle controls that implement the driving decisions,

wherein the computer code causes the autonomous driving system to;

predict, using the obtained data, neighboring vehicles that will distract other drivers by associating the obtained data with distraction, and

adopt a defensive driving profile in response to the identification of the vehicles that will distract other drivers,

wherein the defensive driving profile is a driving profile that accounts for neighboring vehicles that will distract other drivers and a prediction of how the other drivers will drive in response to the distraction.

6. The system of claim 5, further comprising a database that associates the physical characteristics with respective expected driving patterns for vehicles having those physical characteristics.

7. The system of claim 5, wherein the physical characteristics include a determination of a classification of the neighboring vehicle as one of at least a sports car, a sedan, a sport utility vehicle, and a truck.

8. The system of claim 7, wherein the computer code causes the autonomous driving system to:

predict that a lane containing a truck will move slowly; and

determine that the autonomous vehicle should avoid the lane containing the truck.

9. The system of claim 5, wherein the physical characteristics include at least one of a make of the neighboring vehicle and a model of the neighboring vehicle.

10. The system of claim 5, wherein the computer code causes the autonomous driving system to identify an existence of physical hazards associated with the neighboring vehicle.

11. The system of claim 10, wherein the physical hazards include exposed cargo stored on the neighboring vehicle.

12. The system of claim 10, further comprising weather sensors for determining present weather conditions, wherein

the computer code causes the autonomous driving system to assess driving patterns of the neighboring vehicle given existing weather conditions.

13. The system of claim 12, wherein

the database associates certain physical characteristics with aggressive driving patterns, and

the computer code causes the autonomous driving system to increase a driving distance relative to vehicles having the physical characteristics associated with the aggressive driving patterns.

14. A method of autonomously driving an autonomous vehicle, comprising:

obtaining data regarding neighboring vehicles in a proximity of the autonomous vehicle;

identifying physical characteristics of the neighboring vehicles;

mapping the identified physical characteristics into predicted potential on-road events in the proximity of the autonomous vehicle;

determining driving decisions based on the predicted potential on-road events; and

controlling the vehicle based on the driving decisions,

identifying vehicle makes and vehicle models for the neighboring vehicles; and

mapping the identified vehicle makes into predictions of expected driving behaviors for the neighboring vehicles and vehicle models into predictions of expected driving behaviors for the neighboring vehicles;

determining present weather conditions; and

identifying neighboring vehicles that are not suited for present weather conditions;

wherein one or more vehicles are identified as not suited in response to a determination of a likelihood that the vehicles will slip, skid, or be unable to accelerate in the determined present weather conditions.

15. The method of claim 14, comprising:

classifying neighboring vehicles as one of at least a sports car, a sedan, a sport utility vehicle, and a truck; and

mapping the classifications into predictions of expected driving behaviors for the neighboring vehicles.

16. The method of claim 15, comprising determining lane change timing and location based on vehicle makes or vehicle models of vehicles occupying neighboring lanes.

17. The method of claim 14, comprising

identifying hazards associated with the neighboring vehicles;

and

determining driving decisions that increase a driving distance relative to the hazards.

18. The method of claim 17, comprising identifying exposed cargo on the neighboring vehicle.