Vehicle path guiding apparatus and method

1. A vehicle path guiding method, comprising:

determining, based on first sensor data, two-dimensional (2D) coordinate location information of a vehicle;

determining, based on a landmark included in an ambient image of the vehicle, a distance from the landmark to the vehicle in a direction of travel of the vehicle;

correcting, based on the determined distance, the 2D coordinate location information in the direction of travel of the vehicle;

generating, based on the corrected 2D coordinate location information, three-dimensional (3D) vehicle path information for a path along which the vehicle is to be driven;

determining, based on the ambient image, a driving lane in which the vehicle is travelling;

correcting, based on a location of the determined driving lane, the generated 3D vehicle path information; and

displaying the corrected 3D vehicle path information,

wherein determining of the driving lane comprises generating a lane model by modeling lanes around the vehicle based on the ambient image of the vehicle and location information of the driving lane determined using second sensor data, and determining the driving lane using the generated lane model.

2. The vehicle path guiding method of claim 1 , wherein the correcting of the vehicle path information comprises

matching the vehicle path information and the generated lane model, and

correcting the vehicle path information based on the location of the driving lane corresponding to the lane model.

3. The vehicle path guiding method of claim 2 , wherein the lane model comprises a model from which an ambient environment in the ambient image is removed.

4. The vehicle path guiding method of claim 1 , wherein the generating of the vehicle path information comprises

extracting distance information of the vehicle from the ambient image of the vehicle, including the position of the vehicle in the direction of travel, and

correcting the location information of the vehicle based on the distance information.

5. The vehicle path guiding method of claim 4 , wherein the correcting of the location information comprises

generating a nearby map based on the distance information,

comparing the nearby map and the location information of the vehicle, and

correcting the location information of the vehicle based on a comparison result.

6. The vehicle path guiding method of claim 1 , wherein the generating of the lane model comprises

extracting distance information of the vehicle from the ambient image of the vehicle, and

generating the lane model based on the distance information and location information of a lane determined using second sensor data.

7. The vehicle path guiding method of claim 6 , wherein the second sensor data is acquired from a light detection and ranging (LiDAR) sensor or a radio detection and ranging (RADAR) sensor.

8. The vehicle path guiding method of claim 1 , wherein the generating of the vehicle path information comprises

generating vehicle path information of a two-dimensional (2D) map,

converting the 2D map into a three-dimensional (3D) map, and

converting the vehicle path information of the 2D map into vehicle path information of the 3D map based on a result of the converting of the 2D map.

9. The vehicle path guiding method of claim 1 , wherein the first sensor data is acquired from an inertial measurement unit (IMU) or a global positioning system (GPS).

10. The vehicle path guiding method of claim 1 , wherein the generating of the vehicle path information comprises

detecting the landmark in the ambient image of the vehicle,

correcting the location information of the vehicle based on location information of the detected landmark, and

generating the vehicle path information based on the corrected location information.

11. A vehicle path guiding method, comprising:

generating, based on two-dimensional (2D) location information of a vehicle determined using first sensor data, three-dimensional (3D) vehicle path information for a path along which the vehicle is to be driven;

determining, based on an ambient image of the vehicle, a driving lane in which the vehicle is travelling;

correcting, based on a location of the determined driving lane, the generated 3D vehicle path information; and

displaying the corrected 3D vehicle path information,

wherein the generating of the vehicle path information comprises

detecting a landmark in the ambient image of the vehicle,

correcting the 2D location information of the vehicle based on location information of the detected landmark, and

generating the 3D vehicle path information based further on the corrected 2D location information,

wherein determining of the driving lane comprises generating a lane model by modeling lanes around the vehicle based on the ambient image of the vehicle and lane location information, and determining the driving lane using the generated lane model.

12. The vehicle path guiding method of claim 11 , wherein the detecting of the landmark comprises

extracting distance information of the vehicle from the ambient image of the vehicle,

generating a nearby map based on the distance information, and

detecting an intersection as the landmark using the nearby map.

13. 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 .

14. A vehicle path guiding apparatus comprising:

a processor configured to:

determine, based on first sensor data, two-dimensional (2D) coordinate location information of a vehicle,

determine, based on a landmark included in an ambient image of the vehicle, a distance from the landmark to the vehicle in a direction of travel of the vehicle,

correct, based on the determined distance, the 2D coordinate location information in the direction of travel of the vehicle,

generate, based on the corrected 2D coordinate location information, three-dimensional (3D) vehicle path information for a path along which the vehicle is to be driven,

determine, based on the ambient image, a driving lane in which the vehicle is travelling,

correct, based on a location of the determined driving lane, the generated 3D vehicle path information based on a location of the driving lane, and

display the corrected 3D vehicle path information,

wherein the processor is further configured to generate a lane model by modeling lanes around the vehicle based on the ambient image of the vehicle and location information of a lane determined using second sensor data, and to determine the driving lane using the generated lane model.

15. The vehicle path guiding apparatus of claim 14 , wherein the processor is further configured to

match the vehicle path information and the generated lane model, and

correct the vehicle path information based on the location of the driving lane corresponding to the lane model.

16. The vehicle path guiding apparatus of claim 14 , wherein the processor is further configured to

detect the landmark in the ambient image of the vehicle,

correct the location information of the vehicle based on location information of the detected landmark, and

generate the vehicle path information based on the corrected location information.

17. The vehicle path guiding apparatus of claim 14 , wherein the processor is further configured to

extract distance information of the vehicle from the ambient image of the vehicle, and

correct the location information of the vehicle based on the distance information.

18. The vehicle path guiding apparatus of claim 14 , further comprising:

an inertial measurement unit (IMU) or a global positioning system (GPS) configured to acquire the first sensor data.

19. The vehicle path guiding apparatus of claim 14 , further comprising:

a light detection and ranging (LiDAR) sensor or a radio detection and ranging (RADAR) sensor configured to acquire the second sensor data.

20. A vehicle path guiding method, comprising:

determining, using global positioning system (GPS) data, two-dimensional (2D) location information of a vehicle;

determining, based on a landmark included in an ambient image of the vehicle, a distance from the landmark to the vehicle in a direction of travel of the vehicle;

correcting, based on the determined distance, the 2D location information in the direction of travel of the vehicle;

generating, based on the corrected 2D location information, three-dimensional (3D) vehicle path information for a driving path of the vehicle, wherein the vehicle path information comprises a road-level accuracy;

determining, based on the ambient image, a driving lane in which the vehicle is travelling;

correcting, based on a location of the determined driving lane, the generated 3D vehicle path information; and

outputting the corrected vehicle path information through any one or both of a display and a speaker,

wherein determining of the driving lane comprises generating a lane model by modeling lanes around the vehicle based on the ambient image of the vehicle and lane location information, and determining the driving lane using the generated lane model.

21. The method of claim 20 , wherein the image comprises a multi-view image.

22. The method of claim 21 , further comprising:

detecting an intersection as the landmark in the multi-view image; and

correcting the location information of the vehicle based on a distance between the vehicle and the intersection.

23. The method of claim 20 , wherein the generating of the vehicle path information comprises converting vehicle path information of a 2D map into vehicle path information of a 3D map.