Method and apparatus providing information for driving vehicle

1. A method comprising:

detecting a lane region including lane demarcation lines from a driving image;

determining curvature information of a road on which a vehicle is driving based on map information;

estimating a lateral offset of the vehicle based on the lane region and a reference line having the curvature information; and

outputting a signal including information for driving the vehicle based on the curvature information and the lateral offset,

wherein the determining curvature information comprises:

performing a linear regression analysis on a plurality of waypoints located on an edge of the road, wherein the plurality of waypoints are located within a point adjacent to a current location of the vehicle and another point away from the point adjacent to the current location of the vehicle by a look-ahead distance;

obtaining a regression function corresponding to a shape of the road based on the linear regression analysis; and

determining the curvature information corresponding to coefficients of the regression function, using the regression function.

2. The method of claim 1 , wherein the detecting of the lane region comprises:

classifying an object included in the driving image;

generating a segmentation image in which the object included in the driving image is classified; and

detecting the lane region from the segmentation image.

3. The method of claim 1 , wherein the determining of the curvature information comprises:

determining the current location of the vehicle.

4. The method of claim 1 , wherein the look-ahead distance is determined based on any one or any combination of a speed of the vehicle, a visible distance, a weather condition, a driving condition, and a condition of the road.

5. The method of claim 1 , wherein the determining of the curvature information comprises:

selecting, when the road on which the vehicle is driving includes a plurality of paths, a path based on destination information of the vehicle; and

determining the curvature information of the road corresponding to the path.

6. The method of claim 1 , wherein the estimating of the lateral offset of the vehicle comprises:

converting an image of the lane region into a top-view image;

calculating polynomial fitting scores of lines included in the top-view image with respect to the reference line; and

estimating the lateral offset of the vehicle based on the polynomial fitting scores.

7. The method of claim 6 , wherein the calculating of the polynomial fitting scores comprises:

calculating, while moving the reference line by a predetermined number of pixel units, candidate fitting scores in response to the reference line being moved, by counting a number of matching pixels between the lines included in the top-view image and the reference line.

8. The method of claim 7 , wherein the calculating of the candidate fitting scores comprises:

moving the reference line by the predetermined number of pixel units;

counting matching pixels between the lines included in the top-view image and the reference line moved by the predetermined number of pixel units; and

calculating a candidate fitting score associated with the reference line based on the matching pixels.

9. The method of claim 6 , wherein the estimating of the lateral offset of the vehicle comprises:

detecting a left line and a right line of a lane of the road on which the vehicle is driving based on the polynomial fitting scores; and

calculating the lateral offset based on a lateral position of the vehicle between the left line and the right line of the lane of the road on which the vehicle is driving in the top-view image.

10. The method of claim 9 , wherein the detecting the left line and the right line comprises:

determining at least one polynomial fitting score corresponding to each line among the polynomial fitting scores based on a threshold; and

detecting a corresponding line based on the at least one polynomial fitting score corresponding to each line.

11. The method of claim 10 , wherein the detecting of the corresponding line comprises:

detecting a polynomial fitting score having a highest value from at least one polynomial fitting score of the corresponding line; and

determining the corresponding line based on offset information corresponding to the polynomial fitting score.

12. The method of claim 10 , wherein the detecting of the corresponding line comprises:

obtaining a weighted sum of at least one item of offset information corresponding to the at least one polynomial fitting score based on at least one polynomial fitting score of the corresponding line as a weight; and

determining the corresponding line based on the weighted sum.

13. The method of claim 9 , wherein the calculating of the lateral offset comprises:

determining a center of the lane of the road on which the vehicle is driving between the left line and the right line of the lane of the road on which the vehicle is driving adjacent to the lateral position of the vehicle; and

calculating the look-ahead distance between the center of the lane of the road on which the vehicle is driving and the center of the vehicle.

14. The method of claim 9 , wherein the lateral position of the vehicle in the top-view image includes a laterally central position of the top-view image.

15. The method of claim 9 , wherein the lateral position of the vehicle in the top-view image is determined based on a lateral position of a camera attached to the vehicle to capture the driving image.

16. The method of claim 1 , wherein the outputting of the signal including the information for the driving the vehicle comprises:

determining a local path of the vehicle based on the curvature information and the lateral offset; and

generating the signal for controlling the driving of the vehicle based on the local path.

17. The method of claim 1 , wherein the outputting of the signal including the information for the driving of the vehicle comprises:

determining a local path of the vehicle based on the curvature information and the lateral offset; and

providing driving information including a driving path of the vehicle based on the local path.

18. The method of claim 1 , further comprising:

acquiring the driving image; and

acquiring the map information.

19. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the method of claim 1 .

20. An apparatus comprising:

a processor configured to:

detect a lane region including lane demarcation lines from a driving image,

perform a linear regression analysis on a plurality of waypoints located on an edge of a road, wherein the plurality of waypoints are located within a point adjacent to a current location of a vehicle and another point away from the point adjacent to the current location of the vehicle by a look-ahead distance,

obtain a regression function corresponding to a shape of the road based on the linear regression analysis,

determine curvature information corresponding to coefficients of the regression function, using the regression function,

estimate a lateral offset of the vehicle based on the lane region and a reference line having the curvature information, and

output a signal including information for driving the vehicle based on the curvature information and the lateral offset.

21. A method comprising:

detecting a lane region on a road with a camera sensor;

determining a current location of the vehicle with a GPS sensor;

performing a linear regression analysis on a plurality of waypoints located on an edge of the road, wherein the plurality of waypoints are located within a point adjacent to the current location of a vehicle and another point away from the point adjacent to the current location of the vehicle by a look-ahead distance;

obtaining a regression function corresponding to a shape of the road based on the linear regression analysis;

determining curvature information corresponding to coefficients of the regression function, using the regression function;

determining an offset of the vehicle from a center of the lane region based on the curvature information; and

controlling driving of the vehicle based on the determined offset and the curvature information.

22. The method of claim 21 , wherein the center of the lane region is a center line of the road.

23. The method of claim 21 , wherein the curvature information is obtained by performing a linear regression analysis on the plurality of waypoints of the road.