Method, apparatus, and computer program product for determining vehicle lanes of a road segment based on received probe data

1. A map services provider system comprising:

a memory configured to receive probe data points from a plurality of probes, wherein the probe data comprises heading information and location information for each probe data point; and

processing circuitry configured to:

match the probe data to a road segment to generate map-matched probe data;

divide the road segment into sub-segments;

determine distances between each probe data point map-matched to the road segment and a point on each of the sub-segments of the road segment that is closest to the probe data point;

for each probe data point, select the shortest distance from among the distances between that probe data point and the point closest to that probe data point on the sub-segments of the road segment;

determine a value for the shortest distance of each probe data point;

analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment;

determine a number of vehicle travel lanes of the road segment based on peaks in the established multi-modal distribution being associated with individual lanes; and

provide the determined number of vehicle travel lanes and the associated road segment to a map services database for lane-level route guidance.

2. The system of claim 1 , wherein the multi-modal distribution of probe data is determined by a multimodality detection and clustering algorithm.

3. The system of claim 1 , wherein the processing circuitry configured to analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment uses a multimodality detection and clustering algorithm with first magnitude parameter; and

wherein the processing circuitry is further configured to:

analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment using the multimodality detection and clustering algorithm with a second magnitude parameter, different from the first magnitude parameter.

4. The system of claim 1 , wherein

each sub-segment is defined between two consecutive shape points; and

wherein the value for the shortest distance is positive in response to being disposed on a first side of the sub-segment and negative in response to being disposed on the opposite side of the sub-segment.

5. The system of claim 1 , wherein the processing circuitry configured to analyze the probe data relative to the road segment comprises processing circuitry configured to:

generate a distribution of probe data points from the road segment, wherein the shortest distance from the road segment for each probe data point is a lane distance from the centerline of the road segment; and

establish multi-modal trends in the distribution of probe data points.

6. The system of claim 1 , wherein the processing circuitry configured to analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment is further configured to:

divide the road segment into a plurality of road sub-segments; and

for each road sub-segment, analyze the probe data relative to the road sub-segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road sub-segment.

7. The system of claim 6 , wherein the processing circuitry configured to analyze the probe data relative to the road sub-segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road sub-segment comprises processing circuitry configured to:

analyze each road sub-segment using a multimodal distribution clustering algorithm and a series of different magnitude values to establish a number of lanes and a confidence value for each combination of road sub-segment and magnitude value; and

establish a number of lanes of the road segment based on a most represented value for the number of lanes from among all combinations of the road sub-segments and the magnitude values.

8. An apparatus comprising at least one processor and at least one memory including computer program code, the at least one memory and computer program code configured to, with the processor, cause the apparatus to at least:

receive probe data from a plurality of probes, wherein the probe data comprises location and heading;

match the probe data to a road segment to generate map-matched probe data comprising a plurality of map-matched probe data points;

divide the road segment into a plurality of sub-segments, wherein each sub-segment comprises a predefined reference position;

determine, for each probe data point, a lane distance from the probe data point to the predefined reference position of each of the plurality of sub-segments;

establish, for each probe data point, a shortest lane distance as the probe data point lane distance for the respective probe data point relative to the road segment;

analyze the probe data relative to the road segment to generate a multi-modal distribution of the probe data points representing a distance of the probe data from the respective predefined reference position of the road segment based on the probe data point lane distance for each probe data point;

determine a number of vehicle travel lanes of the road segment based on the probe data point lane distance from the predefined reference and the multi-modal distribution; and

provide the determined number of vehicle travel lanes and the associated road segment to a map services database for lane-level route guidance.

9. The apparatus of claim 8 , wherein causing the apparatus to match each probe data point to an individual lane of the road segment based on the lane distance from the predefined reference and the multi-modal distribution comprises establishing individual lanes of the road segment based on the peaks of the multi-modal distribution using a multimodality detection and clustering algorithm.

10. The apparatus of claim 8 , wherein causing the apparatus to divide the road segment into a plurality of sub-segments comprises dividing the road segment into a plurality of sub-segments, where each sub-segment is defined between two consecutive shape points along the road segment.

11. The apparatus of claim 8 , wherein causing the apparatus to analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment uses a multimodality detection and clustering algorithm with first magnitude parameter; and

wherein the apparatus is further caused to:

analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment using the multimodality detection and clustering algorithm with a second magnitude parameter, different from the first magnitude parameter.

12. The apparatus of claim 8 , wherein causing the apparatus to analyze the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment comprises:

for each road sub-segment, analyze the probe data relative to the road sub-segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road sub-segment.

13. The apparatus of claim 12 , wherein causing the apparatus to analyze the probe data relative to the road sub-segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road sub-segment comprises causing the apparatus to:

analyze each road sub-segment using a multimodal distribution clustering algorithm and a series of different magnitude values to establish a number of lanes and a confidence value for each combination of road sub-segment and magnitude value; and

establish a number of lanes of the road segment based on a most represented value for the number of lanes from among all combinations of the road sub-segments and the magnitude values.

14. A method comprising:

receiving probe data from a plurality of probes, wherein the probe data comprises probe data points having location and heading;

matching the probe data to a road segment to generate map-matched probe data;

dividing the road segment into sub-segments;

determining distances between each probe data point map-matched to the road segment and a point on each of the sub-segments of the road segment that is closest to the probe data point;

for each probe data point, selecting the shortest distance from among the distances between that probe data point and the point closest to that probe data point on the sub-segments of the road segment;

determining a value for the shortest distance of each probe data point;

analyzing the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment;

determining a number of vehicle travel lanes of the road segment based on peaks in the established multi-modal distribution being associated with individual lanes; and

providing the determined number of vehicle travel lanes and the associated road segment to a map services database for lane-level route guidance.

15. The method of claim 14 , wherein the multi-modal distribution of probe data is determined by a multimodality detection and clustering algorithm.

16. The method of claim 14 , wherein analyzing the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment uses a multimodality detection and clustering algorithm with first magnitude parameter;

the method further comprising:

analyzing the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment using the multimodality detection and clustering algorithm with a second magnitude parameter, different from the first magnitude parameter.

17. The method of claim 14 , wherein

each sub-segment is defined between two consecutive shape points; and

wherein the value for the shortest distance is positive in response to being disposed on a first side of the sub-segment and negative in response to being disposed on the opposite side of the sub-segment.

18. The method of claim 14 , wherein analyzing the probe data relative to the road segment comprises:

generating a distribution of probe data points from the road segment, wherein the shortest distance from the road segment for each probe data point is a lane distance from the centerline of the road segment; and

establishing multi-modal trends in the distribution of probe data points.

19. The method of claim 14 , wherein analyzing the probe data relative to the road segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road segment comprises:

dividing the road segment into a plurality of road sub-segments; and

for each road sub-segment, analyzing the probe data relative to the road sub-segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road sub-segment.

20. The method of claim 19 , wherein analyzing the probe data relative to the road sub-segment to establish a multi-modal distribution of probe data representing a distance of the probe data from a predefined reference position of the road sub-segment comprises:

analyzing each road sub-segment using a multimodal distribution clustering algorithm and a series of different magnitude values to establish a number of lanes and a confidence value for each combination of road sub-segment and magnitude value; and

establishing a number of lanes of the road segment based on a most represented value for the number of lanes from among all combinations of the road sub-segments and the magnitude values.