Modifying vehicle behavior based on confidence in lane estimation

• US 8,948,954 B1
• Filed: 03/15/2012
• Issued: 02/03/2015
• Est. Priority Date: 03/15/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Ferguson, David Ian, Dolgov, Dmitri
• Primary Examiner(s): Cheung, Calvin
• Assistant Examiner(s): Ramesh, Krishnan
• Application Number:
  US13/420,652
• Time in Patent Office: 1,055 Days
• US Class Current: 701/23
• CPC Class Codes: B60W 2556/50 of positioning data, e.g. GPS [Global Positioning System] data
• CPC Class Codes: B60W 30/12 Lane keeping

Abstract

Methods and systems for modifying vehicle behavior based on confidence in lane estimation are described. In an example, a computing device may be configured to receive lane information relating to locations of lane boundaries and may be configured to estimate a lane boundary on a road on which the vehicle is traveling, based on the lane information. The computing device also may be configured to determine a level of confidence for the estimated lane boundary, modify a driving behavior for the vehicle based on the level of confidence, and also may be configured to control the vehicle based on the modified driving behavior.

Claims

1. A method, comprising:

receiving, by a computing device, map information indicative of lane boundaries on a road on which a vehicle is travelling;

determining, by the computing device, that the map information indicative of the lane boundaries on the road is unreliable;

in response to determining that the map information is unreliable, receiving additional lane information from at least one on-board sensor coupled to the vehicle;

estimating, by the computing device, the lane boundaries based on the additional lane information;

determining a level of confidence for the estimated lane boundaries based on the additional lane information; and

determining, by the computing device, a path for the vehicle based on the level of confidence and the estimated lane boundaries.

2. The method of claim 1, wherein the additional lane information comprises one or more of:

(i) position tracking information of one or more vehicles on the road;

(ii) location information of one or more static objects with respect to the road;

(iii) lane marker recognition information; and

(iv) road geometry information.

3. The method of claim 1, wherein determining the path for the vehicle comprises:

causing the vehicle to follow a center of a given lane such that a degree of adherence of the vehicle to the center of the given lane is proportional to the level of confidence; and

taking the estimated lane boundaries into account as a soft constraint that is violable.

4. The method of claim 1, wherein estimating the lane boundaries based on the additional lane information received from the at least one on-board sensor comprises identifying a lane marker based on the additional lane information, and wherein determining the level of confidence for the estimated lane boundaries comprises determining a metric of clarity for the lane marker.

5. The method of claim 1, wherein the additional lane information comprises multiple information types received from respective multiple sources of information, and wherein determining the level of confidence for the estimated lane boundaries based on the additional lane information comprises:

determining, for each information type received from a respective source of information, a corresponding level of confidence for the estimated lane boundaries; and

determining the level of confidence for the estimated lane boundaries based on corresponding levels of confidence determined for the multiple sources of information.

6. The method of claim 1, wherein the additional lane information received from the at least one on-board sensor includes at least position tracking information of a given vehicle in a vicinity of the vehicle, wherein the position tracking information indicates a trajectory of travel of the given vehicle, and wherein estimating the lane boundaries based on the additional lane information comprises:

estimating the lane boundaries based on the trajectory of travel of the given vehicle.

7. The method of claim 1, wherein the computing device is configured to control the vehicle in an autonomous mode of operation.

8. A system, comprising:

at least one sensor coupled to a vehicle;

at least one processor in communication with the at least one sensor; and

data storage comprising instructions that, when executed by the at least one processor, cause the system to perform functions comprising;

receiving map information indicative of lane boundaries on a road on which the vehicle is travelling;

determining that the map information indicative of the lane boundaries on the road is unreliable;

in response to determining that the map information is unreliable, receiving additional lane information from the at least one sensor;

estimating the lane boundaries based on the additional lane information;

determining a level of confidence for the estimated lane boundaries based on the additional lane information; and

determining a path for the vehicle based on the level of confidence and the estimated lane boundaries.

9. The system of claim 8, further comprising a Global Position System (GPS) sensor coupled to the vehicle, the functions further comprising:

receiving location information from the GPS sensor, wherein the map information is based on the location information.

10. The system of claim 8, wherein the additional lane information comprises one or more of:

(i) position tracking information of one or more vehicles on the road;

(ii) location information of one or more static objects with respect to the road;

(iii) lane marker recognition information; and

(iv) road geometry information.

11. The system of claim 8, wherein the function of determining the path for the vehicle comprises:

causing the vehicle to follow a center of a given lane such that a degree of adherence of the vehicle to the center of the given lane is proportional to the level of confidence; and

taking the estimated lane boundaries into account as a soft constraint that is violable.

12. The system of claim 8, wherein the at least one sensor includes an image-capture device, wherein the additional lane information comprises image information received from the image-capture device, and wherein the function of estimating the lane boundaries based on the additional lane information comprises:

identifying a lane marker based on the image information, and wherein the function of determining the level of confidence for the estimated lane boundaries comprises determining a metric of clarity of the lane marker.

13. The system of claim 8, wherein the additional lane information comprises multiple information types received from respective multiple sources of information, and wherein the function of determining the level of confidence for the estimated lane boundaries based on the additional lane information comprises:

determining, for each information type received from a respective source of information, a corresponding level of confidence for the estimated lane boundaries; and

determining the level of confidence for the estimated lane boundaries based on corresponding levels of confidence determined for the multiple sources of information.

14. The system of claim 8, wherein the additional lane information includes at least position tracking information of a given vehicle in a vicinity of the vehicle, wherein the position tracking information indicates a trajectory of travel of the given vehicle, and wherein the function of estimating the lane boundaries based on the additional lane information comprises:

estimating the lane boundaries based on the trajectory of travel of the given vehicle.

15. A non-transitory computer readable medium having stored thereon instructions that, when executed by a computing device, cause the computing device to perform functions comprising:

receiving map information indicative of lane boundaries on a road on which a vehicle is travelling;

determining that the map information indicative of the lane boundaries on the road is unreliable;

in response to determining that the map information is unreliable, receiving additional lane information from an on-board sensor coupled to the vehicle;

estimating the lane boundaries based on the additional lane information;

determining a level of confidence for the estimated lane boundaries based on the additional lane information; and

determining a path for the vehicle based on the level of confidence and the estimated lane boundaries.

16. The non-transitory computer readable medium of claim 15, wherein the additional lane information comprises multiple information types received from respective multiple sources of information, and wherein the multiple information types comprise one or more of:

(i) position tracking information of one or more vehicles on the road;

(ii) location information of one or more static objects with respect to the road;

(iii) lane marker recognition information; and

(iv) road geometry information.

17. The non-transitory computer readable medium of claim 15, wherein the function of determining the path for the vehicle comprises:

causing the vehicle to follow a center of a given lane such that a degree of adherence of the vehicle to the center of the given lane is proportional to the level of confidence; and

taking the estimated lane boundaries into account as a soft constraint that is violable.

18. The non-transitory computer readable medium of claim 15, wherein the additional lane information comprises multiple information types received from respective multiple sources of information, and wherein the function of determining the respective level of confidence for the estimated lane boundaries based on the additional lane information comprises:

determining, for each information type received from a respective source of information, a corresponding level of confidence for the estimated lane boundaries; and

determining the level of confidence for the estimated lane boundaries based on corresponding levels of confidence determined for the multiple sources of information.

19. The non-transitory computer readable medium of claim 15, wherein the additional lane information received from the on-board sensor includes at least position tracking information of a given vehicle in a vicinity of the vehicle, wherein the position tracking information indicates a trajectory of travel of the given vehicle, and wherein the function of estimating the lane boundaries based on the additional lane information comprises:

estimating the lane boundaries based on the trajectory of travel of the given vehicle.

20. The non-transitory computer readable medium of claim 15, wherein the additional lane information comprises image information captured by the on-board sensor, wherein the function of estimating the lane boundaries on the road comprises identifying a lane marker based on the image information, and wherein the function of determining the level of confidence for the estimated lane boundaries comprises determining a metric of clarity of the lane marker.