Learning lanes from radar data
Systems, methods, and apparatuses are disclosed for determining lane information of a roadway segment from vehicle probe data. Probe data is received from radar sensors of vehicles at a road segment, where the probe data includes an identification of static objects and dynamic objects in proximity to the respective vehicles at the road segment, and geographic locations of the static objects and the dynamic objects. A reference point, such as a road boundary, at the road segment is determined from the identified static objects. Lateral distances between the identified dynamic objects and the reference point are calculated. A number of lanes at the road segment are ascertained from a distribution of the calculated distances of the identified dynamic objects from the reference point.
1. A method comprising:
receiving probe data from radar sensors of vehicles at a road segment, the probe data comprising an identification of static objects and dynamic objects in proximity to the respective vehicles at the road segment, and geographic locations of the static objects and the dynamic objects;
determining, using a processor, a common reference point at the road segment among the identified static objects received from the vehicles;
calculating lateral distances between the identified dynamic objects and the common reference point using the geographic locations of the identified dynamic objects and the common reference point, each lateral distance measured along an axis laterally perpendicular to lanes of the road segment;
ascertaining a number of lanes at the road segment from a distribution of the calculated distances of the identified dynamic objects from the common reference point, wherein the distribution of the calculated distances of the identified dynamic objects is arranged in a histogram, and the number of lanes at the road segment are ascertained based on a number of peaks in the histogram; and
updating a map database with the ascertained number of lanes at the road segment.
2. The method of claim 1 , further comprising:
defining each lane of the ascertained number of lanes at the road segment with a respective lane number; and
identifying the lane number of a vehicle based on a shortest lateral distance of the vehicle to a location represented by a peak in the histogram.
3. The method of claim 1 , further comprising:
verifying the number of lanes at the road segment using ground truth data.
4. The method of claim 1 , wherein the common reference point is a position at a road boundary.
5. The method of claim 1 , wherein the dynamic objects identified are traveling in a same direction as the respective vehicles.
6. 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:
receive probe data from radar sensors of vehicles at a road segment, the probe data comprising an identification of static objects and dynamic objects in proximity to the respective vehicles at the road segment, and geographic locations of the static objects and the dynamic objects;
determine a common reference point at the road segment among the static objects received from the vehicles;
calculate lateral distances of the identified dynamic objects from the common reference point using the geographic locations of the identified dynamic objects and the common reference point, each lateral distance measured along an axis laterally perpendicular to lanes of the road segment;
ascertain a number of lanes at the road segment from a distribution of the calculated distances of the identified dynamic objects from the common reference point, wherein the distribution of the calculated distances of the identified dynamic objects is arranged in a histogram, and the number of lanes at the road segment are ascertained based on the number of peaks in the histogram; and
update a map database with the ascertained number of lanes at the road segment.
7. The apparatus of claim 6 , wherein the common reference point is a position at a road boundary.
8. The apparatus of claim 6 , wherein the at least one memory and the computer program code are configured to further cause the apparatus to:
define each lane of the ascertained number of lanes at the road segment with a respective lane number; and
identify the lane number of a vehicle based on a shortest distance of the vehicle to a location represented by a peak in the histogram.
9. A method comprising:
receiving probe data from radar sensors of vehicles at a road segment, the probe data comprising an identification of static objects and dynamic objects in proximity to the respective vehicles at the road segment, and geographic locations of the static objects and the dynamic objects;
determining, using a processor, a common reference point at the road segment among the identified static objects received from the vehicles;
calculating lateral distances between the identified dynamic objects and the common reference point using the geographic locations of the identified dynamic objects and the common reference point, each lateral distance measured along an axis laterally perpendicular to lanes of the road segment;
ascertaining a number of lanes at the road segment from a distribution of the calculated distances of the identified dynamic objects from the common reference point, wherein the distribution of the calculated distances of the identified dynamic objects is arranged in a histogram, and the number of lanes at the road segment are ascertained based on a number of peaks in the histogram; and
reporting the ascertained number of lanes at the road segment to a vehicle navigation device over a connected network.
10. The method of claim 9 , further comprising:
defining each lane of the ascertained number of lanes at the road segment with a respective lane number; and
identifying the lane number of a vehicle based on a shortest lateral distance of the vehicle to a location represented by a peak in the histogram.
11. The method of claim 9 , further comprising:
verifying the number of lanes at the road segment using ground truth data.
12. The method of claim 9 , wherein the common reference point is a position at a road boundary.
13. The method of claim 9 , wherein the dynamic objects identified are traveling in a same direction as the respective vehicles.
14. 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:
receive probe data from radar sensors of vehicles at a road segment, the probe data comprising an identification of static objects and dynamic objects in proximity to the respective vehicles at the road segment, and geographic locations of the static objects and the dynamic objects;
determine a common reference point at the road segment among the static objects received from the vehicles;
calculate lateral distances of the identified dynamic objects from the common reference point using the geographic locations of the identified dynamic objects and the common reference point, each lateral distance measured along an axis laterally perpendicular to lanes of the road segment;
ascertain a number of lanes at the road segment from a distribution of the calculated distances of the identified dynamic objects from the common reference point, wherein the distribution of the calculated distances of the identified dynamic objects is arranged in a histogram, and the number of lanes at the road segment are ascertained based on the number of peaks in the histogram; and
report the ascertained number of lanes at the road segment to a vehicle navigation device over a connected network.
15. The apparatus of claim 14 , wherein the common reference point is a position at a road boundary.
16. The apparatus of claim 14 , wherein the at least one memory and the computer program code are configured to further cause the apparatus to:
define each lane of the ascertained number of lanes at the road segment with a respective lane number; and
identify the lane number of a vehicle based on a shortest distance of the vehicle to a location represented by a peak in the histogram.