Learning Lanes From Vehicle Probes

• US 20160167582A1
• Filed: 12/16/2014
• Published: 06/16/2016
• Est. Priority Date: 12/16/2014
• Status: Active Grant

• Current Assignee: HERE Global B.V. (HERE Holding Corporation)
• Original Assignee: HERE Global B.V. (HERE Holding Corporation)
• Inventors: Chen, Xin, Ma, Di, Ma, Xiang, Ostrovskiy, Roman, Zhukov, Vladimir, Zou, Xiaotao
• Granted Patent: US 10,262,213 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
• CPC Class Codes: G08G 1/0112 from the vehicle, e.g. floating car data [FCD]

Abstract

Systems, methods, and apparatuses are disclosed for determining lane information of a roadway segment from vehicle probe data. Probe data is received from vehicle camera sensors at a road segment, wherein the probe data includes lane marking data on the road segment. Lane markings are identified, to the extent present, for the left and right boundaries of the lane of travel as well as the adjacent lane boundaries to the left and right of the lane of travel. The identified lane markings are coded, wherein solid lane lines, dashed lane lines, and unidentified or non-existing lane lines are differentiated. The coded lane markings are compiled in a database. A number of lanes are predicted at the road segment from the database of coded lane markings.

Claims

1. A method comprising:

receiving, using a processor, probe data from vehicle camera sensors from a plurality of vehicles at a road segment, the probe data comprising lane marking data for the road segment;

identifying, from the probe data for each vehicle of the plurality of vehicles, any lane markings present for;

(1) a left boundary of a lane of travel, (2) a right boundary of the lane of travel, (3) a left boundary of an adjacent lane to the left of the lane of travel, and (4) a right boundary of an adjacent lane to the right of the lane of travel;

coding the identified lane markings, wherein solid lane lines, dashed lane lines, and unidentified or non-existing lane lines are differentiated;

compiling the coded lane markings from the plurality of vehicles at the road segment in a database; and

predicting a number of lanes at the road segment from the coded lane markings.

2. The method of claim 1, wherein the identifying of the lane markings includes an identification of additional lane markings of additional lane boundaries beyond the adjacent lane to the left of the lane of travel and the adjacent lane to the right of the lane of travel.

3. The method of claim 1, further comprising:

ascertaining a lane number of a vehicle using the predicted number of lanes, lane markings identified by the vehicle, geographic location data of the vehicle, speed of the vehicle, and direction of travel data of the vehicle.

4. The method of claim 3, wherein the lane number is ascertained using a Dead Reckoning algorithm.

5. The method of claim 1, further comprising:

calculating a lane width for a lane of the road segment through a random sample consensus algorithm using estimated distance data between vehicles and the left boundary and the right boundary and modeling the distance data to provide the lane width.

6. The method of claim 1, further comprising:

reporting the predicted number of lanes to a navigation device over a connected network.

7. The method of claim 1, wherein the predicting of the number of lanes comprises using a machine learning algorithm to compare the database of coded lane markings with a probability distribution database that correlates individual coded lane combinations with percentages that the road segment includes a specific number of lanes.

8. The method of claim 7, further comprising:

training the machine learning algorithm through a verification of the predicted number of lanes with ground truth data.

9. The method of claim 8, wherein the machine learning algorithm is a Random Forest algorithm.

10. The method of claim 1, further comprising:

detecting a change in lane count through a comparison of the number of lanes at the road segment with a determined number of lanes at an adjacent road segment.

11. The method of claim 1, further comprising:

ascertaining a lane transition at a road intersection from the predicted number of lanes at the road segment, geographic location data of the plurality of vehicles entering and exiting the road intersection, speed data of the plurality of vehicles entering and exiting the road intersection, and direction of travel data of the plurality of vehicles entering and exiting the road intersection.

12. A method comprising:

receiving, using a processor, probe data from vehicle camera sensors from a plurality of vehicles at a road segment, the probe data comprising lane marking data for the road segment;

identifying, from the probe data for each vehicle of the plurality of vehicles, any lane markings present for;

(1) a left boundary of a lane of travel, (2) a right boundary of the lane of travel, (3) a left boundary of an adjacent lane to the left of the lane of travel, and (4) a right boundary of an adjacent lane to the right of the lane of travel;

coding the identified lane markings, wherein solid lane lines, dashed lane lines, and unidentified or non-existing lane lines are differentiated;

compiling the coded lane markings from the plurality of vehicles at the road segment in a database; and

predicting a number of lanes at the road segment using a Random Forest algorithm comparing the database of coded lane markings with a probability distribution database that correlates individual coded lane combinations with percentages that the road segment includes a specific numbers of lanes.

13. An apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs;

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform;

receive probe data from vehicle camera sensors from a plurality of vehicles at a road segment, the probe data comprising lane marking data for the road segment;

identify any lane markings present from the probe data for each vehicle of the plurality of vehicles for;

(1) a left boundary of a lane of travel, (2) a right boundary of the lane of travel, (3) a left boundary of an adjacent lane to the left of the lane of travel, and (4) a right boundary of an adjacent lane to the right of the lane of travel;

code the identified lane markings, wherein solid lane lines, dashed lane lines, and unidentified or non-existing lane lines are differentiated;

compile the coded lane markings from the plurality of vehicles at the road segment in a database; and

predict a number of lanes at the road segment from the database of coded lane markings.

14. The apparatus of claim 13, wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

verify the predicted number of lanes with ground truth data.

15. The apparatus of claim 13, wherein the identifying of the lane markings includes an identification of additional lane markings of additional lane boundaries beyond the adjacent lane to the left of the lane of travel and the adjacent lane to the right of the lane of travel.

16. The apparatus of claim 13, wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

ascertain a lane number of a vehicle using the predicted number of lanes, lane markings identified by the vehicle, geographic location data of the vehicle, speed of the vehicle, and direction of travel data of the vehicle.

17. The apparatus of claim 16, wherein the lane number is ascertained using a Dead Reckoning algorithm.

18. The apparatus of claim 13, wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

calculate a lane width for a lane of the road segment through a random sample consensus algorithm using estimated distance data between vehicles and the left boundary and the right boundary and modeling the distance data to provide the lane width.

19. The apparatus of claim 13, wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

report the predicted number of lanes to a navigation device over a connected network.

20. The apparatus of claim 13, wherein the predicting of the number of lanes comprises using a machine learning algorithm to compare the database of coded lane markings with a probability distribution database that correlates individual coded lane combinations with percentages that the road segment includes a specific number of lanes.

21. The apparatus of claim 13, wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

detect a change in lane count through a comparison of the number of lanes at the road segment with a determined number of lanes at an adjacent road segment.

22. The apparatus of claim 13, wherein the at least one memory and the computer program code are configured to cause the apparatus to further perform:

ascertain a lane transition at a road intersection from the predicted number of lanes at the road segment, geographic location data of the plurality of vehicles entering and exiting the road intersection, speed data of the plurality of vehicles entering and exiting the road intersection, and direction of travel data of the plurality of vehicles entering and exiting the road intersection.