Data-based driveline estimation and mapping

1 . A method, comprising:

receiving road-level map data and observed driveline data for one or more road users observed in a vehicle transportation network, wherein:

the road-level map data comprises way data comprising one or more ways, a way of the one or more ways comprising a series of nodes, and

the driveline data comprises one or more drivelines, a driveline of the one or more drivelines comprising a series of poses representing an observed location of the one or more road users as the one or more road users traverse the vehicle transportation network;

identifying first sections of the way data as intersections of the vehicle transportation network;

matching second sections of the way data with the observed driveline data by associating a pose of the series of poses of the observed driveline data with a corresponding node of the series of nodes of the way data, to generate multiple way bars, wherein a way bar of the multiple way bars comprises one or more poses of the series of poses from respective drivelines and comprises one node of the series of nodes, and wherein the second sections are different from the first sections;

categorizing a respective way bar as either constant or changing based on a number of lanes inferred from lateral offsets of the one or more poses from the one node, wherein a lane count representing the number of lanes is assigned to the respective way bar when the respective way bar is categorized as constant;

grouping consecutive way bars of the multiple way bars into way bar sections based on the categorization and the lane count of the respective way bar;

generating a data-based driveline map representing lane-level geometry different from the road-level map data using the way bar sections and the first sections; and

operating a vehicle using the data-based driveline map as input to a control system of the vehicle.

2 . The method of claim 1 , wherein an intersection is identified by identifying nodes that correspond to more than one way.

3 . The method of claim 1 , wherein the number of lanes inferred from lateral offsets of the one or more poses from the one node is determined via a first process, wherein the first process comprises:

grouping the one or more poses of the way bar according to directionality;

defining offsets of the one or more poses from the one node;

sorting the offsets within the groups identified in the grouping of the one or more poses;

determining a distance between the offsets as ordered by the sorting of the one or more poses; and

determining the number of lanes based on the distance.

4 . The method of claim 3 , wherein the first process further comprises:

increasing the lane count in response to the distance being greater than a new lane threshold.

5 . The method of claim 3 , wherein the first process further comprises:

characterizing the lane count as unknown in response to the distance being greater than a same lane threshold and less than a new lane threshold.

6 . The method of claim 1 , wherein the generating the data-based driveline map comprises:

assigning link nodes to a beginning and an end of the way bar sections;

determining a first link node of a first way bar section of the way bar sections that connects to a second link node of a second way bar section of the way bar sections; and

connecting a first driveline of the one or more drivelines to a second driveline of the one or more drivelines, wherein the first driveline is associated with the first link node and the second driveline is associated with the second link node.

7 . The method of claim 1 , further comprising:

converting the data-based driveline map to a lane-level map by setting map properties for the data-based driveline map, wherein the map properties comprise:

speed limits, and

lane relationships.

8 . The method of claim 1 , wherein the driveline data is obtained from sensors mounted on an autonomous vehicle.

9 . An apparatus, comprising:

a memory; and

a processor configured to execute instructions stored in the memory to:

receive road-level map data and observed driveline data for one or more road users observed in a vehicle transportation network, wherein:

the road-level map data comprises way data comprising one or more ways, a way of the one or more ways comprising a series of nodes, and

the driveline data comprises one or more drivelines, a driveline of the one or more drivelines comprising a series of poses representing an observed location of the one or more road users as the one or more road users traverse the vehicle transportation network;

identify first sections of the way data as intersections of the vehicle transportation network;

match second sections of the way data with the observed driveline data by associating a pose of the series of poses of the observed driveline data with a corresponding node of the series of nodes of the way data, to generate multiple way bars, wherein a way bar of the multiple way bars comprises one or more poses of the series of poses from respective drivelines and comprises one node of the series of nodes, wherein the second sections are different from the first sections;

categorize a respective way bar as either constant or changing based on a number of lanes inferred from lateral offsets of the one or more poses from the one node, wherein a lane count representing the number of lanes is assigned to the respective way bar when the respective way bar is categorized as constant;

group consecutive way bars of the multiple way bars into way bar sections based on the categorization and the lane count of the respective way bar;

generate a data-based driveline map representing lane-level geometry different from the road-level map data using the way bar sections and the first sections; and

operate a vehicle using the data-based driveline map as input to a control system of the vehicle.

10 . The apparatus of claim 9 , wherein an intersection is identified by identifying nodes that correspond to more than one way.

11 . The apparatus of claim 9 , wherein the number of lanes inferred from lateral offsets of the one or more poses from the one node is determined via a first process, wherein the first process comprises:

grouping the one or more poses of the way bar according to directionality;

defining offsets of the one or more poses from the one node;

sorting the offsets within the groups identified in the grouping of the one or more poses;

determining a distance between the offsets as ordered by the sorting of the one or more poses; and

determining the number of lanes based on the distance.

12 . The apparatus of claim 11 , wherein the first process further comprises:

increasing the lane count in response to the distance being greater than a new lane threshold; and

characterizing the lane count as unknown in response to the distance being greater than a same lane threshold and less than a new lane threshold.

13 . The apparatus of claim 9 , wherein the generating the data-based driveline map comprises:

assigning link nodes to a beginning and an end of the way bar sections;

determining a first link node of a first way bar section of the way bar sections that connects to a second link node of a second way bar section of the way bar sections; and

connecting a first driveline of the one or more drivelines to a second driveline of the one or more drivelines, wherein the first driveline is associated with the first link node and the second driveline is associated with the second link node.

14 . The apparatus of claim 9 , wherein the processor is further configured to:

convert the data-based driveline map to a lane-level map by setting map properties for the data-based driveline map, wherein the map properties comprise:

speed limits, and

lane relationships.

15 . The apparatus of claim 9 , wherein the driveline data is obtained from sensors mounted on an autonomous vehicle.

16 . A non-transitory computer-readable medium storing instructions operable to cause one or more processors to perform operations comprising:

receiving road-level map data and observed driveline data for one or more road users observed in a vehicle transportation network, wherein:

the road-level map data comprises way data comprising one or more ways, a way of the one or more ways comprising a series of nodes, and

the observed driveline data comprises one or more drivelines, a driveline of the one or more drivelines comprising a series of poses representing an observed location of the one or more road users as the one or more road users traverse the vehicle transportation network;

identifying first sections of the way data as intersections of the vehicle transportation network;

matching second sections of the way data with the observed driveline data by associating a pose of the series of poses of the observed driveline data with a corresponding node of the series of nodes of the way data, to generate multiple way bars, wherein a way bar of the multiple way bars comprises one or more poses of the series of poses from respective drivelines and comprises one node of the series of nodes, and wherein the second sections are different from the first sections;

categorizing a respective way bar as either constant or changing based on a number of lanes inferred from lateral offsets of the one or more poses from the one node, wherein a lane count representing the number of lanes is assigned to the respective way bar when the respective way bar is categorized as constant;

grouping consecutive way bars of the multiple way bars into way bar sections based on the categorization and the lane count of the respective way bar;

generating a data-based driveline map representing lane-level geometry different from the road-level map data using the way bar sections and the first sections; and

operating a vehicle using the data-based driveline map as input to a control system of the vehicle.

17 . The non-transitory computer-readable medium storing instructions of claim 16 , wherein the number of lanes inferred from lateral offsets of the one or more poses from the one node is determined via a first process, wherein the first process comprises:

grouping the one or more poses of the way bar according to directionality;

defining offsets of the one or more poses from the one node;

sorting the offsets within the groups identified in the grouping of the one or more poses;

determining a distance between the offsets as ordered by the sorting of the one or more poses; and

determining the number of lanes based on the distance.

18 . The non-transitory computer-readable medium storing instructions of claim 17 , wherein the first process further comprises:

increasing the lane count in response to the distance being greater than a new lane threshold; and

characterizing the lane count as unknown in response to the distance being greater than a same lane threshold and less than a new lane threshold.

19 . The non-transitory computer-readable medium storing instructions of claim 16 , wherein the generating the data-based driveline map comprises:

assigning link nodes to a beginning and an end of the way bar sections;

determining a first link node of a first way bar section of the way bar sections that connects to a second link node of a second way bar section of the way bar sections; and

connecting a first driveline of the one or more drivelines to a second driveline of the one or more drivelines, wherein the first driveline is associated with the first link node and the second driveline is associated with the second link node.

20 . The non-transitory computer-readable medium storing instructions of claim 16 , the operations further comprising:

converting the data-based driveline map to a lane-level map by setting map properties for the data-based driveline map, wherein the map properties comprise:

speed limits, and

lane relationships.