APPARATUS AND METHOD FOR RECOGNIZING DRIVING ENVIRONMENT FOR AUTONOMOUS VEHICLE

• US 20160061612A1
• Filed: 11/28/2014
• Published: 03/03/2016
• Est. Priority Date: 09/02/2014
• Status: Active Grant

• Current Assignee: Hyundai Motor Company (Hyundai Motor Group)
• Original Assignee: Hyundai Motor Company (Hyundai Motor Group)
• Inventors: You, Byung Yong, Heo, Myung Seon, Oh, Young Chul
• Granted Patent: US 9,360,328 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01C 21/26 specially adapted for navigation in a road network
• CPC Class Codes: G01C 21/28 with correlation of data from several navigational instruments
• CPC Class Codes: G01S 19/13 Receivers
• CPC Class Codes: G01S 19/14 specially adapted for specific applications

Abstract

An apparatus and a method are provided for recognizing driving environment for an autonomous vehicle. The apparatus includes a controller configured to receive navigation information from a satellite navigation receiver. The controller is further configured to receive map data from a map storage and image data from an image sensor regarding captured images from around a vehicle and distance information from a distance sensor regarding sensed objects positioned around the vehicle. The controller is also configured to determine a fusion method for information measured by the image sensor and the distance sensor based on a receiving state of the satellite navigation receiver and precision of the map data to recognize the driving environment.

Claims

1. A system for recognizing a driving environment for an autonomous vehicle, comprising:

a memory configured to store program instructions;

a processor configured to execute the program instructions, the program instructions when executed configured to;

receive navigation information from a satellite navigation receiver configured to receive a satellite signal from a satellite;

receive map data from a map storage configured to store map data;

receive image data from an image sensor configured to capture images around the autonomous vehicle;

receive distance information from a distance sensor configured to sense objects positioned around the autonomous vehicle;

determine a fusion method for information measured by the image sensor and the distance sensor based on a receiving state of the satellite navigation receiver and precision of the map data to recognize the driving environment;

determine a reception strength of the navigation information;

determine a precision level of the map data; and

extract and object from information measured by the image sensor and the distance sensor based on the reception strength of the navigation information and the precision level of the map data,

wherein the extracted object is used to recognize the driving environment.

2. The system according to claim 1, wherein the program instructions when executed are further configured to:

extract road bounding spheres using the image data from the image sensor and the distance information from the distance sensor; and

derive a minimum bounding sphere of the extracted road bounding spheres, when the satellite signal is not received or the number of satellites whose signals are directly received on a line of sight (LOS) is equal to or less than a reference value.

3. The system according to claim 2, wherein the program instructions when executed are further configured to:

map the minimum bounding sphere with the map data to generate a possible driving path.

4. The system according to claim 1, wherein the program instructions when executed are further configured to:

extract road facilities using the image data from the image sensor; and

extract surrounding vehicle information using the distance information from the distance sensor when received information included in the satellite signal is in an error range within several meters.

5. The system according to claim 4, wherein the program instructions when executed are further configured to:

perform probability based filtering on a comparison result of the road facilities with a landmark within the map data and the surrounding vehicle information to correct longitudinal and latitudinal positions of the vehicle.

6. The system according to claim 1, wherein the program instructions when executed are further configured to:

extract a substantially long range obstacle using the image data from the image sensor; and

extract a substantially short range obstacle using the distance information from the distance sensor when received information included in the satellite signal is within an error range and the map data is sufficiently precise.

7. The system according to claim 6, wherein the program instructions when executed are further configured to:

map the substantially long range obstacle and the substantially short range obstacle with the map data to recognize a driving situation and predict a behavior of a surrounding vehicle based on the driving situation.

8. The system according to claim 1, wherein the program instructions when executed are further configured to:

extract driving lane information using the image data from the image sensor; and

extract surrounding vehicle information using the distance information from the distance sensor when information included in the satellite signal is within an error range and the precision level of the map data is sufficiently precise.

9. The system according to claim 8, wherein the program instructions when executed are further configured to:

perform probability based filtering on the driving lane information and the surrounding vehicle information to correct a latitudinal position of the vehicle.

10. A method for recognizing a driving environment for an autonomous vehicle, comprising:

determining, by a controller, a reception strength of a satellite signal received from a satellite;

determining, by the controller, a precision level of map data received from a map storage configured to store map data; and

extracting, by the controller, an object from information measured by an image sensor configured to capture images around the autonomous vehicle and a distance sensor configured to sense objects positioned around the autonomous vehicle based on the reception strength of the satellite signal and the precision level of the map data,

wherein the extracted object is used to recognize the driving environment.

11. The method according to claim 10, further comprising:

confirming, by the controller, whether the reception strength of the satellite signal is less than a reference value;

extracting, by the controller, a road facility and surrounding vehicle information, respectively, using the image sensor and the distance sensor when the reception strength of the satellite signal is less than the reference value; and

correcting, by the controller, longitudinal and latitudinal positions of the vehicle by performing probability based filtering on the road facility and the surrounding vehicle information.

12. The method according to claim 11, further comprising:

determining, by the controller, that the map data is sufficiently precise when the reception strength of the satellite signal is equal to or greater than the reference value;

extracting, by the controller, a substantially long range obstacle and a substantially short range obstacle using the image sensor and the distance sensor, respectively, when the map data is sufficiently precise; and

predicting, by the controller, a behavior of the surrounding vehicle by mapping the long range obstacle and the substantially short range obstacle with the map data.

13. The method according to claim 12, further comprising:

extracting, by the controller, driving lane information and the surrounding vehicle information using the image sensor and the distance sensor, respectively, when the map data is not sufficiently precise; and

performing, by the controller, probability based filtering on the driving lane information and the surrounding vehicle information to correct a latitudinal position of the vehicle.

14. The method according to claim 10, further comprising:

extracting, by the controller, a road bounding sphere using the image sensor and the distance sensor when reception of the satellite signal is substantially unsatisfactory;

extracting, by the controller, a minimum bounding sphere of the road bounding spheres; and

generating, by the controller, a possible driving path based on the minimum bounding sphere.

15. A non-transitory computer readable medium containing program instructions executed by a processor or controller for recognizing driving environment for an autonomous vehicle, the computer readable medium comprising:

program instructions that determine a reception strength of a satellite signal received from a satellite;

program instructions that determine a precision level of map data received from a map storage configured to store map data; and

program instructions that extract an object from information measured by an image sensor configured to capture images around the autonomous vehicle and a distance sensor configured to sense objects positioned around the autonomous vehicle based on the reception strength of the satellite signal and the precision level of the map data,

wherein the extracted object is used to recognize the driving environment.

16. The non-transitory computer readable medium according to claim 15, further comprising:

program instructions that confirm whether the reception strength of the satellite signal is less than a reference value;

program instructions that extract a road facility and surrounding vehicle information, respectively, using the image sensor and the distance sensor when the reception strength of the satellite signal is less than the reference value; and

program instructions that correct longitudinal and latitudinal positions of the vehicle by performing probability based filtering on the road facility and the surrounding vehicle information.

17. The non-transitory computer readable medium according to claim 16, wherein the program instructions that recognize driving environment further include:

program instructions that determine whether the map data is sufficiently precise when the reception strength of the satellite signal is equal to or greater than the reference value;

program instructions that extract a substantially long range obstacle and a substantially short range obstacle using the image sensor and the distance sensor, respectively, when the map data is sufficiently precise; and

program instructions that predict a behavior of the surrounding vehicle by mapping the substantially long range obstacle and the substantially short range obstacle with the detailed map.

18. The non-transitory computer readable medium according to claim 17,

wherein the program instructions that recognize driving environment further include:

program instructions that extract driving lane information and the surrounding vehicle information using the image sensor and the distance sensor, respectively, when the map data is not sufficientvly precise; and

program instructions that perform probability based filtering on the driving lane information and the surrounding vehicle information to correct a latitudinal lateral position of the vehicle.

19. The non-transitory computer readable medium according to claim 15, further comprising:

program instructions that extract a road bounding sphere using the image sensor and the distance sensor when reception of the satellite signal is substantially unsatisfactory;

program instructions that extract a minimum bounding sphere of the road bounding spheres; and

program instructions that generate a possible driving path based on the minimum bounding sphere.