EP-4012680-B1 - METHOD, APPARATUS AND COMPUTER PROGRAM PRODUCT FOR DETECTING A LANE CLOSURE USING PROBE DATA

EP4012680B1EP 4012680 B1EP4012680 B1EP 4012680B1EP-4012680-B1

Inventors

FOWE, JAMES ADEYEMI

Dates

Publication Date: 20260506
Application Date: 20211209

Claims (15)

An apparatus (20) comprising at least processing circuitry (22) and at least one non-transitory memory (24) including computer program code instructions, the computer program code instructions configured to, when executed by the processing circuitry, cause the apparatus to: use lateral position indicators to partition subject probe data (400), being real time or near-real time data associated with at least one road segment, into a same number of clusters (410, 412, 414, 416) as historical probe data (500) associated with the at least one road segment, wherein the lateral positional indicators indicate a distance from a predefined basis, the distance being measured at a direction orthogonal to the flow of traffic, and wherein the historical probe data is clustered based on respective lateral positional indicators; for each cluster of the subject probe data, calculate a same statistical measure for the subject lateral position indicators as calculated for the historical lateral positional indicators of a corresponding cluster of the historical probe data, wherein the statistical measures indicate a positioning of a respective cluster including a direction and distance from the predefined basis, the distance being measured at a direction orthogonal to the flow of traffic, and compare the statistical measure of the subject lateral positional indicators to the respective statistical measure of the historical lateral positional indicators; and determine whether any lane of the at least one road segment is closed (812) dependent upon the comparison of the statistical measure of the subject lateral positional indicators to the respective statistical measure of the historical lateral positional indicators.
The apparatus according to claim 1, wherein the computer program code instructions are further configured to, when executed by the processing circuitry, cause the apparatus to: in an instance the statistical measure for at least one cluster of the subject lateral positional indicators differs from the statistical measure for the respective at least one cluster of the historical lateral positional indicators by either of (a) at least a closure threshold, or (b) an amount greater than the closure threshold, determine that at least one lane of the at least one road segment is closed.
The apparatus according to claim 1, wherein the computer program code instructions are further configured to, when executed by the processing circuitry, cause the apparatus to: in an instance it is determined no lanes of the at least one segment are closed, determine whether at least one lane of the at least one road segment is shifted (810).
The apparatus according to claim 1, wherein the computer program code instructions are further configured to, when executed by the processing circuitry, cause the apparatus to: in an instance at least one lane of the at least one road segment is determined as closed, determine a direction of lateral offset (550, 650) of the statistical measure for at least one cluster of the subject lateral positional indicators relative to the statistical measure for the respective at least one cluster of the historical lateral positional indicators; and identify at least one closed lane based upon the direction of the lateral offset.
The apparatus according to claim 1, wherein determining whether any lane of the at least one segment is closed is performed in real-time or near real-time relative to the receipt of the subject probe data.
The apparatus according to claim 1, wherein the subject probe data is associated with a time relative to a week or a day of the week, and the historical probe data is associated with the same time period relative to at least one prior week or at least one prior day of the week.
The apparatus according to claim 1, wherein the computer program code instructions are further configured to, when executed by the processing circuitry, cause the apparatus to: perform a k-means algorithm on the historical probe data to partition the historical probe data and determine the clusters of the historical probe data; and determine the respective statistical measures of the historical lateral positional indicators.
A computer implemented method comprising: using lateral positional indicators to partition subject probe data (400), being real time or near-real time data associated with at least one road segment, into a same number of clusters (410, 412, 414, 416) as historical probe data (500) associated with the at least one road segment, wherein the lateral positional indicators indicate a distance from a predefined basis, the distance being measured at a direction orthogonal to the flow of traffic, and wherein the historical probe data is clustered based on respective lateral positional indicators; for each cluster of the subject probe data, calculating a same statistical measure for the subject lateral position indicators as calculated for the historical lateral positional indicators of a corresponding cluster of the historical probe data, wherein the statistical measures indicate a positioning of a respective cluster including a direction and distance from the predefined basis, the distance being measured at a direction orthogonal to the flow of traffic, and comparing the statistical measure of the subject lateral positional indicators to the respective statistical measure of the historical lateral positional indicators; and determining whether any lane of the at least one road segment is closed (812) dependent upon the comparison of the statistical measure of the subject lateral positional indicators to the respective statistical measure of the historical lateral positional indicators.
The method according to claim 8, further comprising: in an instance the statistical measure for at least one cluster of the subject lateral positional indicators differs from the statistical measure for the respective at least one cluster of the historical lateral positional indicators by either of (a) at least a closure threshold, or (b) an amount greater than the closure threshold, determining that at least one lane of the at least one road segment is closed.
The method according to claim 8, further comprising: in an instance it is determined no lanes of the at least one segment are closed, determining whether at least one lane of the at least one road segment is shifted (810).
The method according to claim 8, further comprising: in an instance at least one lane of the at least one road segment is determined as closed, determining a direction of lateral offset (550, 650) of the statistical measure for at least one cluster of the subject lateral positional indicators relative to the statistical measure for the respective at least one cluster of the historical lateral positional indicators; and identifying at least one closed lane based upon the direction of the lateral offset.
The method according to claim 8, wherein determining whether any lane of the at least one segment is closed is performed in real-time or near real-time relative to the receipt of the subject probe data.
The method according to claim 8, wherein the subject probe data is associated with a time relative to a week or a day of the week, and the historical probe data is associated with the same time period relative to at least one prior week or at least one prior day of the week.
The method according to claim 8, further comprising: performing a k-means algorithm on the historical probe data to partition the historical probe data and determine the clusters of the historical probe data; and determining the respective statistical measures of the historical lateral positional indicators.
A computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein that, when executed by one or more processors, cause the one or more processors to perform the method of any one of the preceding method claims.

Description

TECHNOLOGICAL FIELD An example embodiment relates generally to a method, apparatus and computer program product for detecting lane statuses of road segments, such as a lane closures or shifting, using probe data. BACKGROUND Some traffic-aware routing and navigation systems detect road closure events or slowdowns caused by accidents, road construction, and/or the like by detecting the thru traffic in road segments. Such systems detect average speeds of traffic moving through a particular segment, and provide alerts relating to slowdowns when average speeds of traffic stop or differ substantially from average or normal traffic speeds for the same segment at the same time of day or time of week. In some instances, the alert may include a general cause of the slowdown, such as road construction or accident, and may provide a time estimate of additional time it will take to travel through the segment, in comparison to average or normal traffic conditions. In many cases, information related to the general cause of slowdowns, such as construction or an accident, is provided to a service from another user who has passed through the area, or may be generated from integrated accident reporting systems, road construction reporting systems, and/or the like. The information or alerts provided to drivers using such systems may be helpful for determining general traffic speeds for the segment, but may not provide details pertaining to lane-level information. For example, in instances in which a road segment includes a plurality of lanes, the segment information is oftentimes not specific as to individual lanes nor the side of the road, or side of the road segment affected. Even if information regarding lane closure or shoulder closure is provided, such systems rely on user input, such as by another driver, or an administrator or customer service representative associated with construction activities, accident responses, and/or the like, to provide the lane-level details to a system prior to dissemination to drivers. In many cases the information may be outdated, inaccurate, and/or not provided in a timely and efficient manner to alert drivers. US patent application, Pub. No. "US 2020/0202708 A1", discloses an approach for speed aggregation of probe data for high-occupancy-vehicle (HOV) or other road lanes. The approach involves, for example, determining a line that is parallel to a road segment and divides the road segment along a longitudinal axis. The approach also involves determining a spatial distribution of probe data collected from the road segment with respect to the line. The approach further involves clustering the probe data into a first cluster and a second cluster based on speed. US patent application, Pub. No. "US 2019/0311613 A1", discloses a method for identifying lane obstructions. The method includes collecting, in an electronic data store, position data of surrounding vehicles observed by reporting vehicles that travel over a roadway segment. The method includes analyzing the position data to identify whether observed positions correlate with an obstruction pattern that is indicative of a lane obstruction in at least one lane of the roadway segment. US patent application, Pub. No. "US 2015/0170514 A1", discloses a methods for determining real time traffic conditions. A candidate road is divided into road segments by perpendicular bisectors. A spatial sliding window is positioned over at least a portion of a road segment, wherein the spatial sliding window corresponds to a front end of the road segment in a direction of travel of the road segment. Real time probe data is received from mobile devices in probe vehicles or on travelers of the at least portion of the road segment within the spatial sliding window. The real time probe data is analyzed, and a computer program assists in determining the real time traffic conditions of the at least portion of the road segment within the spatial sliding window. US patent, Patent number10,140,856, discloses A plurality of instances of probe data are received. Each instance is received from a probe apparatus of a plurality of probe apparatuses each comprising a plurality of sensors and being onboard a vehicle. An instance comprises location information indicating a location of the corresponding probe apparatus. For each of one or more instances, a distance parameter is determined based on the location information and a road segment corresponding to the location. A set of distance parameters is defined based on the distance parameter determined for each of the one or more instances. The set of distance parameters is analyzed to identify clusters of probe data. The number of clusters identified is determined and compared to a historical number of clusters. If the number of clusters identified is less than the historical number of clusters, it is determined that there is a lane closure corresponding to the road segment. BRIEF SUMMARY A method, apparatus and computer p