Granted Patent B2

US 12,283,114 · App. 18/147,031 · Granted Apr 22, 2025

Inventors: Christian Wegner (Grosse Ile, MI); Mahmoud Yousef Ghannam (Canton, MI); Bradford Scott Bondy (St. Clair Shores, MI); Muhannad Hamdan (Canton, MI)

Assignee: Ford Global Technologies, LLC

G06V20/588G01C21/34G06V20/584

Abstract

Claims (68)

1. A system comprising:

A processor and a memory, the memory storing instructions executable by the processor, including instructions to:

Receive forward image data of a roadway from a forward-facing camera of a vehicle, the roadway including a driving lane for the vehicle;

Determine visible lane boundaries of the driving lane based on lane marking features in the forward image data, the forward image data including sufficient lane marking features to determine the visible lane boundaries, wherein a predicted lane boundary is located between the visible lane boundaries on the roadway;

Determine the predicted lane boundary of the driving lane by an in-between predicting model based, at least in part, on forward features in the forward image data, the forward image data including insufficient lane marking features to determine the predicted lane boundary;

Determine a driving path of the vehicle through the visible lane boundaries and the predicted lane boundary; and

Actuate the vehicle based on the driving path.

2. The system of claim 1 , wherein the instructions include instructions to:

receive rearward image data of the roadway;

determine a location accuracy of the predicted lane boundary, the location accuracy based on rearward features from, at least in part, the rearward image data; and

update the in-between predicting model based on the location accuracy of the predicted lane boundary.

3. The system of claim 1 ,

wherein the instructions include instructions to detect lane marking edges from lane markings in the forward image data; and

wherein the forward features include the lane marking edges.

4. The system of claim 1 ,

wherein the instructions include instructions to detect track edges of vehicle tracks in the forward image data; and

wherein the forward features include the track edges.

5. The system of claim 1 ,

wherein the instructions include instructions to detect tire edges of vehicle tires in the forward image data; and

wherein the forward features include the tire edges.

6. The system of claim 1 ,

wherein the instructions include instructions to detect taillight edges of vehicle taillights in the forward image data; and

wherein the forward features include the taillight edges.

7. The system of claim 1 , wherein the forward features include a lane width of the driving lane.

8. The system of claim 1 , further comprising:

instructions to receive LIDAR data from a LIDAR sensor of the vehicle;

wherein the visible lane boundaries are based on, at least in part, the LIDAR data.

9. The system of claim 1 , wherein the instructions include instructions to:

determine a lane deviation of the vehicle when a position of the vehicle in the roadway departs from the driving path; and

provide an indication of lane deviation when the position of the vehicle in the roadway departs from the driving path.

10. A method comprising:

Receiving forward image data of a roadway from a forward-facing camera of a vehicle, the roadway including a driving lane for vehicle;

Determining visible lane boundaries of the driving lane based on lane marking features in the forward image data, the forward image data including sufficient lane marking features to determine the visible lane boundaries, wherein a predicted lane boundary is located between the visible lane boundaries on the roadway;

Determining the predicted lane boundary of the driving lane by an in-between predicting model based, at least in part, on forward features in the forward image data, the forward image data including insufficient lane marking features to determine the predicted lane boundary;

Determining a driving path of the vehicle through the visible lane boundaries and the predicted lane boundary; and

Actuating the vehicle based on the driving path.

11. The method of claim 10 , further comprising: receiving rearward image data of the roadway;

determining a location accuracy of the predicted lane boundary, the location accuracy based on rearward features from, at least in part, the rearward image data;

and updating the in-between predicting model based on the location accuracy of the predicted lane boundary.

12. The method of claim 10 , further comprising:

detecting lane marking edges, vehicle track edges, vehicle tail light edges, or vehicle tire edges in the forward image data; and

wherein the forward features include the lane marking edges, vehicle track edges, vehicle tail light edges, or vehicle tire edges.

13. The method of claim 10 , wherein the forward features include a lane width of the driving lane.

14. The method of claim 10 , further comprising:

receiving LIDAR data from a LIDAR sensor of the vehicle;

wherein the visible lane boundaries are based on, at least in part, the LIDAR data.

15. The method of claim 10 , further comprising: determining a lane deviation of the vehicle when a position of the vehicle in the roadway departs from the driving path;

and providing an indication of lane deviation when the position of the vehicle in the roadway departs from the driving path.

16. A system comprising:

a processor and a memory, the memory storing instructions executable by the processor, including instructions to:

receive forward image data of a roadway from a forward-facing camera of a vehicle, the roadway including a driving lane for the vehicle;

receive rearward image data of the roadway;

determine visible lane boundaries of the driving lane based on lane marking features in the forward image data, the forward image data including sufficient lane marking features to determine the visible lane boundaries;

determine a predicted lane boundary of the driving lane by a forward predicting model based, at least in part, on forward features in the forward image data, the forward image data including insufficient lane marking features to determine the predicted lane boundary, wherein the predicted lane boundary is located forward of the visible lane boundaries on the roadway;

determine a location accuracy of the predicted lane boundary, the location accuracy based on rearward features from, at least in part, the rearward image data;

update the forward predicting model based on the location accuracy of the predicted lane boundary;

determine a driving path of the vehicle through the visible lane boundaries and the predicted lane boundary; and

actuate the vehicle based on the driving path.

17. The system of claim 16 :

wherein the instructions include instructions to detect lane marking edges, vehicle track edges, vehicle tail light edges, or vehicle tire edges in the forward image data; and

wherein the forward features include the lane marking edges, vehicle track edges, vehicle tail light edges, or vehicle tire edges.

18. The system of claim 16 , wherein the forward features include a lane width of the driving lane.

19. The system of claim 16 , further comprising:

instructions to receive LIDAR data from a LIDAR sensor of the vehicle;

wherein the visible lane boundaries are based on, at least in part, the LIDAR data.

20. The system of claim 16 , wherein the instructions include instructions to:

determine a lane deviation of the vehicle when a position of the vehicle in the roadway departs from the driving path; and

provide an indication of lane deviation when the position of the vehicle in the roadway departs from the driving path.

Assignments (1)

Continuity (1)

References Cited (18)