Road boundary determination method and apparatus, electronic device, and storage medium

Abstract

Embodiments of the present disclosure disclose a road boundary determination method and device, electronic equipment and a storage medium. The method comprises: obtaining at least one vehicle driving track and a plurality of lane lines on a road where the vehicle driving track is located; determining a lane where the vehicle driving track is located based on the distance relationship between the vehicle driving track and the lane lines; clustering the vehicle driving tracks based on the distance and driving direction between the vehicle driving tracks; and determining the road boundary of the road corresponding to the lane based on the clustering result and the lane where the vehicle driving tracks of the same class are located. The technical solution can determine the road boundary according to the association between the vehicle driving track and the lane line. Compared with the method of identifying the road boundary from the image by using a complex algorithm, the complexity is low, and in the case that the number of vehicle driving tracks is sufficient and the positioning is accurate, the accuracy of the detected road boundary is high, and the missed detection is not easy to occur.

Description

Technical Field

[0001] This disclosure relates to the field of high-precision map technology, specifically to a method, apparatus, electronic device, and storage medium for determining road boundaries. Background Technology

[0002] Road boundaries are a crucial element in high-definition maps, and accurately detecting them from road data is of great value. Related technologies, based on road images and combined with computer vision or deep learning semantic recognition techniques, identify road boundaries from the images and generate vector data for the map based on these boundaries. However, if the acquired road images are of poor quality, or if the actual road boundaries are worn or missing, identifying road boundaries from images can easily lead to misidentification or omission. Furthermore, the algorithms used in computer vision and deep learning semantic recognition are typically complex, and due to limitations in equipment performance and algorithm capabilities, methods for identifying road boundaries from images are highly complex and inefficient.

[0003] Therefore, a solution is needed to identify road boundaries simply, efficiently, and accurately. Summary of the Invention

[0004] This disclosure provides a method, apparatus, electronic device, and storage medium for determining road boundaries.

[0005] In a first aspect, this disclosure provides a method for determining road boundaries, comprising:

[0006] Obtain at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located;

[0007] Based on the distance relationship between the vehicle's trajectory and the lane lines, the lane in which the vehicle's trajectory is located is determined;

[0008] Based on the vehicle's driving trajectory and direction, the vehicle's driving trajectory is clustered;

[0009] The road boundary of the road corresponding to the lane is determined based on the clustering results and the lanes where the vehicle trajectories are located in the same cluster.

[0010] Secondly, embodiments of the present invention provide a method for updating map vector data, comprising:

[0011] Obtain the road boundary of the target road, wherein the road boundary is determined based on the method described in the first aspect;

[0012] Update the map vector data using the road boundaries.

[0013] Thirdly, embodiments of the present invention provide a road boundary determination device, comprising:

[0014] The acquisition module is configured to acquire at least one vehicle driving trajectory and multiple lane lines on the road where the vehicle driving trajectory is located;

[0015] The lane determination module is configured to determine the lane where the vehicle's trajectory is located based on the distance relationship between the vehicle's trajectory and the lane lines;

[0016] The clustering module is configured to cluster the vehicle trajectories based on the distance and direction of travel between the vehicle trajectories.

[0017] The road boundary determination module is configured to determine the road boundary of the road corresponding to the lane based on the clustering results and the lanes where the vehicle trajectories are located, which are clustered into one class.

[0018] In one possible design, the above-described device includes a memory and a processor. The memory stores one or more computer instructions that support the device in performing the corresponding methods described above, and the processor is configured to execute the computer instructions stored in the memory. The device may also include a communication interface for communicating with other devices or communication networks.

[0019] Fourthly, embodiments of this disclosure provide an electronic device including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in any of the above aspects.

[0020] Fifthly, embodiments of this disclosure provide a computer-readable storage medium for storing computer instructions used by any of the above-described devices, which, when executed by a processor, are used to implement the methods described in any of the above aspects.

[0021] In a sixth aspect, embodiments of this disclosure provide a computer program product comprising computer instructions which, when executed by a processor, are used to implement the methods described in any of the preceding aspects.

[0022] The technical solutions provided in this disclosure can include the following beneficial effects:

[0023] This disclosure proposes a method for determining road boundaries. The method acquires at least one vehicle trajectory and multiple lane lines on the road containing the vehicle trajectory. Based on the distance relationship between the vehicle trajectory and the lane lines, it determines the lane containing the vehicle trajectory. Then, based on the distance and direction of travel between the vehicle trajectories, it clusters the vehicle trajectories. Finally, based on the clustering results and the lanes containing the clustered vehicle trajectories, it determines the road boundary corresponding to each lane. This method determines road boundaries based on the correlation between vehicle trajectories and lane lines. Compared to methods that use complex algorithms to identify road boundaries from images, this method has lower complexity and, when the number of vehicle trajectories is sufficient and the positioning is accurate, it achieves high accuracy in detecting road boundaries and is less prone to missed detections.

[0024] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0025] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0026] Figure 1 A flowchart of a road boundary determination method according to an embodiment of the present disclosure is shown.

[0027] Figure 2 A schematic diagram showing the effect of a road boundary according to an embodiment of the present disclosure is provided.

[0028] Figure 3 A structural block diagram of a road boundary determination device according to an embodiment of the present disclosure is shown.

[0029] Figure 4 This is a schematic diagram of the structure of an electronic device suitable for implementing a road boundary determination method and / or a map vector data update method according to an embodiment of the present disclosure. Detailed Implementation

[0030] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of the exemplary embodiments have been omitted from the drawings.

[0031] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and do not preclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0032] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0033] The user information (including but not limited to user device information such as location information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals shall be provided for users to choose to authorize or refuse.

[0034] The details of the embodiments of this disclosure are described in detail below through specific examples.

[0035] Figure 1 A flowchart illustrating a road boundary determination method according to an embodiment of this disclosure is shown. Figure 1 As shown, the method for determining road boundaries includes the following steps:

[0036] In step S101, at least one vehicle driving trajectory and multiple lane lines on the road where the vehicle driving trajectory is located are obtained;

[0037] In step S102, the lane where the vehicle's trajectory is located is determined based on the distance relationship between the vehicle's trajectory and the lane lines.

[0038] In step S103, the vehicle trajectories are clustered based on the distance and direction of travel between them;

[0039] In step S104, the road boundary of the road corresponding to the lane is determined based on the clustering results and the lane where the vehicle's driving trajectory is located in the same cluster.

[0040] In this embodiment, the road boundary determination method described above can be executed on a server, by a physical device capable of providing server functionality, by a component (such as a chip) configured in the physical device, or by a module capable of implementing some or all of the server functionality, etc., and this disclosure does not limit it in this way. For ease of understanding, this disclosure uses a server as an example to describe the method provided by this disclosure in the following description.

[0041] In this embodiment, the vehicle trajectory can be the path taken by the data acquisition vehicle on the road, and the vehicle trajectory has a corresponding direction of travel. A vehicle trajectory can also be referred to as a vehicle trajectory line. The data acquisition vehicle can collect the vehicle trajectory through an inertial navigation system (INS) or a global positioning system (GPS) deployed on the vehicle, and upload the vehicle trajectory to a server. The vehicle trajectory includes multiple trajectory points, each with corresponding trajectory point coordinates, which can be determined based on a trajectory coordinate system. This trajectory coordinate system is the coordinate system deployed on the data acquisition vehicle to determine the coordinates of the trajectory points. Different data acquisition vehicles may use the same or different trajectory coordinate systems. The data acquisition vehicle can be a professional data acquisition vehicle equipped with high-performance acquisition equipment, or an ordinary vehicle equipped with a camera.

[0042] In some embodiments, the server may obtain at least one vehicle trajectory uploaded by the data collection vehicle.

[0043] In some embodiments, the server may acquire multiple vehicle travel trajectories uploaded by one or more data collection vehicles on the same or different roads. Multiple vehicle travel trajectories on the same road may include vehicle travel trajectories in the same lane or different lanes.

[0044] In some embodiments, when at least one vehicle driving trajectory obtained by the server corresponds to different trajectory coordinate systems, the vehicle driving trajectories corresponding to different trajectory coordinate systems can also be uniformly converted to the same trajectory coordinate system by using the identifier of the trajectory coordinate system corresponding to the vehicle driving trajectory uploaded by the data acquisition vehicle.

[0045] While the data collection vehicle generates vehicle trajectories as it travels on the road, it can also collect road images through cameras deployed on the vehicle. Existing simple image recognition technology is then used to identify vector data such as lane lines on the road containing the vehicle's trajectory from the collected road images. This vector data can be uploaded to the server along with the vehicle's trajectory. It should be noted that in some embodiments, the server can also obtain at least one vehicle trajectory and multiple lane lines and other vector data from a third party. Embodiments of this disclosure can obtain the aforementioned vehicle trajectories and lane line vector data from at least one or more third parties, and then determine road boundaries based on these vehicle trajectories and lane lines. When the trajectory coordinate systems corresponding to vehicle trajectories generated by different third parties are different, the vehicle trajectories obtained from different third parties can be uniformly converted to the same trajectory coordinate system based on the identifier and parameters of the trajectory coordinate system corresponding to the vehicle trajectories obtained from the third party.

[0046] In some embodiments, at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located may include vehicle trajectory lines and lane lines on multiple roads within a region whose location is known.

[0047] In some embodiments, lane lines on a road may include, but are not limited to, the position coordinates of various shaped points on the lane lines, and the coordinate system corresponding to the position coordinates of various shaped points on the lane lines and the trajectory coordinate system of the vehicle's driving trajectory may be the same coordinate system. For example, both the lane line coordinate system and the trajectory coordinate system are world coordinate systems.

[0048] In this embodiment of the disclosure, a lane may include a left lane line and a right lane line. It is understood that the vehicle's trajectory line is typically located between the left and right lane lines of the lane. Therefore, the lane containing the vehicle's trajectory can be obtained from multiple lane lines by using the distance relationship between the lane lines and the vehicle's trajectory line on geographic coordinates. In some embodiments, the lane containing the vehicle's trajectory may include the left lane line closest to the vehicle's trajectory and the right lane line closest to the vehicle's trajectory. In some embodiments, the distance between the lane line and the vehicle's trajectory on geographic coordinates can be understood as the distance between the geographical location of the lane line in the actual scenario and the geographical location of the vehicle's trajectory in the actual scenario.

[0049] After obtaining at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located, the server can further determine the lane where each vehicle trajectory is located. When determining the lane where a vehicle trajectory is located, the distance between the vehicle trajectory and each of the multiple lane lines can be determined. Based on the distance relationship, the left lane line closest to the left side of the vehicle trajectory and the right lane line closest to the right side of the vehicle trajectory can be determined from the multiple lane lines. The determined left and right lane lines closest to the vehicle trajectory constitute the lane where the vehicle trajectory is located. In some embodiments, the distance between the vehicle trajectory and the lane lines can be determined by projecting each trajectory point on the vehicle trajectory onto the left and right lane lines, and the average distance between each trajectory point and the corresponding projection point can be used to determine the distance between the vehicle trajectory and the lane lines.

[0050] In some embodiments, the server may also associate the vehicle's trajectory with the lane in which the trajectory lies. In other words, the server may record the correspondence between the vehicle's trajectory and the nearest left lane line and the nearest right lane line.

[0051] In some embodiments, each vehicle trajectory can be associated with a lane, and each lane can be associated with one or more vehicle trajectories.

[0052] In this embodiment of the disclosure, after determining the lane where each vehicle's driving trajectory is located, the server can further cluster the acquired multiple vehicle driving trajectories. When clustering multiple vehicle driving trajectories, clustering can be performed based on the distance between the vehicle driving trajectories and the driving direction of the vehicle driving trajectories. By clustering the vehicle driving trajectories, at least one class of vehicle driving trajectories can be obtained.

[0053] After clustering vehicle trajectories to obtain at least one class of vehicle trajectories, the server can determine the lanes associated with each class of vehicle trajectories. These lanes constitute a road. Furthermore, the road boundary can be determined based on the left and right lane lines corresponding to the multiple lanes contained in the road.

[0054] In some embodiments, the server can associate a lane with the road in which it is located. In other words, the server can record the correspondence between lanes and roads.

[0055] This disclosure proposes a method for determining road boundaries. The method acquires at least one vehicle trajectory and multiple lane lines on the road containing the vehicle trajectory. Based on the distance relationship between the vehicle trajectory and the lane lines, it determines the lane containing the vehicle trajectory. Then, based on the distance and direction of travel between the vehicle trajectories, it clusters the vehicle trajectories. Finally, based on the clustering results and the lanes containing the clustered vehicle trajectories, it determines the road boundary corresponding to each lane. This method determines road boundaries based on the correlation between vehicle trajectories and lane lines. Compared to methods that use complex algorithms to identify road boundaries from images, this method has lower complexity and, when the number of vehicle trajectories is sufficient and the positioning is accurate, it achieves high accuracy in detecting road boundaries and is less prone to missed detections.

[0056] In an optional implementation of this embodiment, step S103, which involves clustering the vehicle trajectories based on the distance and direction of travel between the vehicle trajectories, may further include the following steps:

[0057] Vehicle trajectories that are less than or equal to a preset distance threshold and travel in the same direction are grouped into one category.

[0058] In this optional implementation, when the server clusters the acquired vehicle driving trajectories, it can group vehicle driving trajectories that are close in distance and have the same driving direction into one category based on the distance and driving direction between the vehicle driving trajectories.

[0059] In some embodiments, vehicle trajectories clustered into one class, i.e., a class of vehicle trajectories, can satisfy the following conditions: 1) the vehicle trajectories clustered into one class have the same direction of travel; 2) among the vehicle trajectories clustered into one class, the distance between two adjacent vehicle trajectories is less than or equal to a preset distance threshold, which can be determined by those skilled in the art according to actual needs. It is understood that the server can cluster vehicle trajectories based on the above clustering conditions.

[0060] In some embodiments, if there are individual vehicle trajectories whose driving direction differs from that of the other vehicle trajectories within a cluster of vehicle trajectories, these individual vehicle trajectories can be filtered out. Similarly, if the distance between individual vehicle trajectories exceeds a preset distance threshold, vehicle trajectories that are far from the majority of vehicle trajectories can also be filtered out.

[0061] In some embodiments, the lanes associated with vehicle travel trajectories grouped into one class may be one or more. In some embodiments, each lane corresponds to a road, and each road includes at least one lane.

[0062] In some embodiments, the travel direction of the lane associated with the grouped vehicle travel trajectories can be marked based on the travel direction of the grouped vehicle travel trajectories, and the travel directions of the two can be associated. The travel direction of the grouped vehicle travel trajectories is the same as the travel direction of the lane associated with the grouped vehicle travel trajectories.

[0063] In some embodiments, the driving direction of the vehicle driving trajectories clustered into one group can be the driving direction of any vehicle driving trajectory within that group. Alternatively, a target vehicle driving trajectory can be determined from the clustered vehicle driving trajectories, and the driving direction of that target vehicle driving trajectory can be used as the driving direction of the vehicle driving trajectories clustered into one group. The target vehicle driving trajectory can be the cluster center selected during clustering, or the vehicle driving trajectory with the highest positioning accuracy within that group.

[0064] In an optional implementation of this embodiment, step S104, which is to determine the road boundary of the road corresponding to the lane based on the clustering results and the lanes where the vehicle trajectories are clustered into one class, may further include the following steps:

[0065] The corresponding lane is determined based on the vehicle trajectories that are clustered into one class in the clustering results, and the driving direction of the lane is the same as the driving direction of the vehicle trajectories in the lane.

[0066] Based on one or more lanes with the same direction of travel, determine the road corresponding to the lane;

[0067] The road boundary is determined from the leftmost and rightmost lane lines of the road.

[0068] In this optional implementation, after the server clusters the acquired vehicle driving trajectories based on step S103 to obtain at least one type of vehicle driving trajectory, it can further determine the road corresponding to the lane of each type of vehicle driving trajectory, as well as the road boundary of that road.

[0069] For any one class of vehicle trajectories obtained from clustering, one or more lanes associated with that class of vehicle trajectories can be determined. Since the vehicles in that class of trajectories travel in the same direction, the one or more lanes associated with them also travel in the same direction, thus forming a road. Therefore, this method can be used to determine the road corresponding to the lanes associated with each class of vehicle trajectories. In other words, the road corresponding to each lane can be determined.

[0070] When a road includes more than one lane, the road can meet the following conditions: 1) the lanes of the more than one lane have the same direction of travel; 2) the lateral distances between the lanes of the more than one lane are similar.

[0071] In some embodiments, the lateral distance between lanes may be within a road width threshold range.

[0072] In some embodiments, the lateral distance between lanes can be the distance between the center lines of the two lanes, which can be the center line of the lane lines on the left and right sides of the lane.

[0073] In some embodiments, the lateral distance between lanes can also be the lateral distance between the centerlines of the two lanes. For example, based on the projection of various shape points on the centerline of one lane onto the centerline of another lane, the average of the projection distances between the projection points corresponding to each shape point and the shape point itself can be used as the lateral distance between the centerlines of the two lanes, that is, the lateral distance between the two lanes. In other words, based on the above two conditions, multiple lanes with the same travel direction and similar lateral distances can be identified as multiple lanes corresponding to the same road.

[0074] In some embodiments, when the lanes containing the travel trajectories of two types of vehicles have the same travel direction and the lateral distance between the lanes is similar, a corresponding road can be determined based on the lanes containing the travel trajectories of these two types of vehicles. That is, if the travel direction of the lane containing the travel trajectories of one type of vehicle is the same as that of the lane containing the travel trajectories of another type of vehicle, and the lateral distance between the lanes containing the travel trajectories of one type of vehicle and the lane containing the travel trajectories of another type of vehicle is similar, then a corresponding road can be determined based on the lanes containing the travel trajectories of these two types of vehicles as a whole.

[0075] Once the road is identified, its boundaries can be determined based on the leftmost and rightmost lane lines. The leftmost lane line corresponds to the leftmost lane of the road, and the rightmost lane line corresponds to the rightmost lane of the road.

[0076] Understandably, if a lane has no other adjacent lanes, the road corresponding to that lane may be a single-lane road. In such cases, the lane can be associated with that road, and the left and right boundaries of the road can be determined based on the left and right lane lines of that lane. In some embodiments, the left lane line of the lane can be determined as the left boundary of the road, and the right lane line of the lane can be determined as the right boundary of the road.

[0077] In an optional implementation of this embodiment, the step of determining the road boundary from the leftmost and rightmost lane lines of the road further includes the following steps:

[0078] When there is no lane line within a preset distance outside the leftmost lane line, the leftmost lane line is determined as the left boundary of the road.

[0079] When there is no lane line within a preset distance outside the rightmost lane line, the rightmost lane line is determined as the right boundary of the road.

[0080] In this optional implementation, the left and right boundaries of a road can be determined from the leftmost and rightmost lane lines associated with one or more lanes corresponding to the same road. To ensure the accuracy of the road boundaries, the server can further determine whether there are other lane lines within a preset distance range outside the leftmost lane line. If no other lane lines exist, the leftmost lane line can be used as the left boundary of the road; if other lane lines exist, the leftmost lane line may not be the left boundary of the road. Furthermore, the server can also determine whether there are other lane lines within a preset distance range outside the rightmost lane line. If no other lane lines exist, the rightmost lane line can be used as the right boundary of the road; if other lane lines exist, the rightmost lane line may not be the right boundary of the road. This is because the lane lines associated with the lanes of the road are determined based on the vehicle's driving trajectory. If the collected vehicle driving trajectory does not include the trajectory of driving in the leftmost and / or rightmost lanes of the road, then the leftmost and / or rightmost lanes of the road are not detected and are not associated with the road. Therefore, the leftmost and / or rightmost lane lines of the associated lanes cannot be determined as the left and / or right boundaries of the road.

[0081] It should be understood that those skilled in the art can set a preset distance range according to actual needs. For example, the preset distance range can be a distance range greater than or equal to the width of one lane. This disclosure does not limit this range, and all such ranges are within the scope of protection.

[0082] In an optional implementation of this embodiment, the method may further include the following steps:

[0083] Based on the clustering results, the different lanes where the vehicle trajectories of the vehicles that were clustered into different classes were located were determined.

[0084] If two lanes with different driving directions are associated with the same lane line, then the same lane line is determined as the up and down road boundaries of the roads corresponding to the two lanes with different driving directions.

[0085] In this optional implementation, after the server clusters the acquired vehicle driving trajectories, if multiple types of vehicle driving trajectories are obtained, the road corresponding to the lane of each type of vehicle driving trajectory can be further determined, and the road boundaries for both directions can also be determined.

[0086] The boundaries of the up and down roads can meet the following conditions: 1) The boundary of the up and down roads is a lane line shared by two adjacent roads, that is, one lane on one of the two adjacent roads and one lane on the other road share the same lane line; 2) The driving trajectories of vehicles on both sides of the boundary of the up and down roads are opposite, that is, the driving trajectories of vehicles on the two adjacent roads are opposite in direction, or the driving directions of lanes on the two adjacent roads are opposite.

[0087] After the server clusters all vehicle trajectories and obtains multiple types of vehicle trajectories, one type of vehicle trajectory can correspond to one lane, and multiple types of vehicle trajectories can determine multiple different lanes. Once the lane is determined, the lane markings can also be determined. If there are two lanes with different travel directions among the multiple lanes, they are associated with the same lane marking. Since these two lanes with different travel directions belong to different roads, the same lane marking can be considered as the up and down road boundaries of the roads corresponding to the two lanes respectively.

[0088] In some embodiments, the server may record a mapping table between vehicle trajectory, lane lines, lanes, and roads. Therefore, to improve efficiency, the up and down lane boundaries of a road can be determined using this mapping table.

[0089] In an optional implementation of this embodiment, step S102, which determines the lane where the vehicle's trajectory is located based on the distance relationship between the vehicle's trajectory and the lane line, further includes the following steps:

[0090] Based on the distance relationship between the trajectory points on each vehicle's driving trajectory and each lane line in geographical coordinates, determine the left lane line and the right lane line that are closest to the left and right sides of the vehicle's driving trajectory from the multiple lane lines;

[0091] The lane in which each vehicle's trajectory is located is determined based on the left lane line and the right lane line.

[0092] In this optional implementation, after the server obtains at least one vehicle driving trajectory and multiple lane lines on the road where the vehicle driving trajectory is located through step S101, it can determine the nearest left lane line and the nearest right lane line to each vehicle driving trajectory by the distance between the trajectory points on each vehicle driving trajectory and each lane line. Then, based on the determined nearest left lane line and the nearest right lane line, the lane where each vehicle driving trajectory is located can be determined.

[0093] For any vehicle trajectory, the nearest left lane line and the nearest right lane line can be determined through the following processes 1) to 3).

[0094] 1) Based on the distance relationship between the trajectory points and lane lines on the vehicle's driving trajectory in the geographical coordinates, determine the left lane line and the right lane line that are closest to the trajectory point from multiple lane lines.

[0095] In this embodiment of the disclosure, since the vehicle's driving trajectory is composed of multiple trajectory points, the server can determine the nearest left lane line and the nearest right lane line to each trajectory point, that is, determine the lane where each trajectory point is located.

[0096] In some embodiments, for any trajectory point on the vehicle's driving trajectory, the nearest left lane line and the nearest right lane line to the trajectory point can be determined as follows: determine the distance between the trajectory point and each of the multiple lane lines; based on the distance, determine the nearest left lane line and the nearest right lane line to the trajectory point from the multiple lane lines.

[0097] In some embodiments, for any trajectory point on the vehicle's driving trajectory and any lane line among multiple lane lines, the distance between the trajectory point and the lane line can be determined as follows: determine the perpendicular from the trajectory point on the vehicle's driving trajectory to the lane line, and obtain the geographic coordinates corresponding to the perpendicular; determine the distance between the trajectory point and the lane line based on the trajectory point coordinates and the geographic coordinates of the perpendicular.

[0098] In some embodiments, the location of the lane line to the left or right of the trajectory point can be determined based on the relative position of the lane line and the trajectory point on the vehicle's driving trajectory in geographic coordinates.

[0099] 2) When multiple left lane lines closest to multiple trajectory points on the vehicle's trajectory are the same lane line, and multiple right lane lines closest to multiple trajectory points on the vehicle's trajectory are the same lane line, the left lane line shall be regarded as the left lane line closest to the vehicle's trajectory, and the right lane line shall be regarded as the right lane line closest to the vehicle's trajectory.

[0100] In this embodiment of the disclosure, after the server determines the nearest left lane line and the nearest right lane line to each trajectory point on the vehicle's trajectory, if the nearest left lane line to all trajectory points on the vehicle's trajectory is the same left lane line, and the nearest right lane line to all trajectory points on the vehicle's trajectory is also the same right lane line, it indicates that the data collection vehicle is traveling in the same lane and has not changed lanes. Therefore, the same left lane line closest to all trajectory points can be taken as the nearest left lane line to the vehicle's trajectory, and the same right lane line closest to all trajectory points can be taken as the nearest right lane line to the vehicle's trajectory, thus determining the lane where the vehicle's trajectory is located.

[0101] 3) When the two left lane lines closest to two adjacent trajectory points on the vehicle's trajectory are different left lane lines, and / or the two right lane lines closest to two adjacent trajectory points on the vehicle's trajectory are different right lane lines, the vehicle's trajectory is divided into two vehicle trajectories, each vehicle trajectory corresponding to one of the two left lane lines, and / or each vehicle trajectory corresponding to one of the two right lane lines.

[0102] In one embodiment of this disclosure, after the server determines the nearest left lane line to each trajectory point on the vehicle's driving trajectory, if any two adjacent trajectory points among the multiple trajectory points are not matched by the same nearest left lane line, the vehicle's driving trajectory can be divided into two independent vehicle driving trajectories based on the location of these two adjacent trajectory points, with each vehicle driving trajectory matched by a nearest left lane line.

[0103] In another scenario, after the server determines the nearest right lane line to each trajectory point on the vehicle's trajectory, if any two adjacent trajectory points are matched with a right lane line that is not the same right lane line, the vehicle's trajectory can be divided into two independent vehicle trajectories based on the location of these two adjacent trajectory points, with each vehicle trajectory matched with a nearest right lane line.

[0104] In both of the above situations, since the vehicle's driving trajectory is divided into two independent vehicle driving trajectories, with one side of these two vehicle driving trajectories corresponding to the same lane line and the other side corresponding to different lane lines, the corresponding lane line can also be divided into two independent lane lines, with each lane line corresponding to one vehicle driving trajectory. In this way, the lane corresponding to each of these two vehicle driving trajectories can be determined.

[0105] In another scenario, after the server determines the nearest left lane line and the nearest right lane line to each trajectory point on the vehicle's trajectory, if any two adjacent trajectory points are matched with a left lane line that is not the same, and the matched right lane line that is not the same, then the vehicle's trajectory is divided into two independent vehicle trajectories based on the positions of these two adjacent trajectory points. Each of these two vehicle trajectories is matched with a nearest left lane line and a nearest right lane line. The nearest left lane line and the nearest right lane line matched by each vehicle trajectory can then constitute the lane of that vehicle's trajectory.

[0106] In short, when the lane lines matching consecutive trajectory points on a vehicle's trajectory change, the vehicle's trajectory can be truncated.

[0107] In some embodiments, the vehicle trajectory needs to be truncated when the lane lines matching consecutive trajectory points on the vehicle trajectory change. Therefore, regardless of whether the vehicle trajectory is truncated, the server can still fit consecutive trajectory points on each vehicle trajectory into a new vehicle trajectory and use a KD-tree (K-dimensional tree) or R-tree (Rectangle Tree) to establish an index between the vehicle trajectory and the lane.

[0108] In some embodiments, the server may also record the correspondence between trajectory points on each vehicle's driving trajectory and the nearest left lane line and the nearest right lane line to the trajectory point. That is, it records the correspondence between trajectory points on each vehicle's driving trajectory and the lanes where the trajectory points are located.

[0109] In an optional implementation of this embodiment, before step S102, i.e., determining the lane where the vehicle's trajectory is located based on the distance relationship between the vehicle's trajectory and the lane line, the method further includes:

[0110] Obtain the location data of the vehicle's driving trajectory;

[0111] Based on the positioning data, vehicle trajectories that do not meet the positioning accuracy requirements are filtered out.

[0112] In this optional implementation, in order to improve the efficiency and accuracy of road boundary determination, after step S101 and before step S102, the server can also determine the trajectory positioning quality of each vehicle trajectory by acquiring the positioning data of the vehicle trajectory; based on the trajectory positioning quality, low-quality vehicle trajectories that do not meet the positioning accuracy requirements in at least one acquired vehicle trajectory are filtered out, and high-quality vehicle trajectories that meet the positioning accuracy requirements are retained.

[0113] In some embodiments, the location data of the vehicle's driving trajectory can be uploaded to the server by the data acquisition vehicle.

[0114] In some embodiments, the location data may include real-time kinematic (RTK) location status and / or trajectory location reliability.

[0115] RTK positioning, based on carrier phase measurement, is a real-time differential GPS measurement that provides real-time positioning results for vehicle trajectories. Different operating modes can be used for RTK positioning of vehicle trajectories. For example, single-point positioning can be used. The RTK positioning status characterizes the positioning accuracy of the vehicle trajectory under a specific operating mode. Trajectory positioning reliability can be used to characterize the positioning accuracy of the vehicle trajectory.

[0116] In some embodiments, different RTK positioning states correspond to different trajectory positioning qualities, and / or different trajectory positioning confidence levels correspond to different trajectory positioning qualities. It should be understood that the correspondence between positioning data and trajectory positioning quality can be predetermined, and this disclosure does not limit the specific correspondence, all of which are within the protection scope of this disclosure.

[0117] When the positioning data includes RTK positioning status, the trajectory positioning quality can be the positioning accuracy of RTK positioning of the vehicle's driving trajectory in a certain working mode; when the positioning data includes trajectory positioning reliability, the trajectory positioning quality can be the positioning accuracy of the vehicle's driving trajectory; when the positioning data includes both RTK positioning status and trajectory positioning reliability, the trajectory positioning quality can be comprehensively determined by combining weights.

[0118] In some embodiments, the positioning accuracy requirement can be that the positioning accuracy of the vehicle's driving trajectory is at the decimeter or meter level or higher. It is understood that the server can retain vehicle driving trajectories with positioning accuracy at the decimeter or meter level or higher as high-quality vehicle driving trajectories, and filter out vehicle driving trajectories with positioning accuracy at the decimeter or meter level or lower as low-quality vehicle driving trajectories.

[0119] In some embodiments, to further improve the efficiency and accuracy of road boundary determination, the server may also include the following process: acquiring the vehicle speed of the vehicle's driving trajectory; and filtering out vehicle driving trajectories that do not conform to the vehicle's driving pattern based on the vehicle speed. For example, the server may filter out vehicle driving trajectories with excessively fast or slow speeds as low-quality vehicle driving trajectories, and retain vehicle driving trajectories with uniform speeds or certain accelerations as high-quality vehicle driving trajectories.

[0120] Figure 2 A schematic diagram showing the effect of a road boundary according to an embodiment of the present disclosure is provided. Figure 2 The map contains Road 1 and Road 2, each with multiple lanes. Each lane contains multiple vehicle trajectory lines, each with a corresponding direction of travel. As shown in the diagram, each vehicle trajectory has a nearest left lane line on its left and a nearest right lane line on its right; these two lane lines can form a lane. If there are no other lane lines outside the leftmost lane line of Road 1, then this leftmost lane line is the left boundary of Road 1. If there are other lane lines outside the rightmost lane line of Road 1, then this rightmost lane line cannot be considered the right boundary of Road 1. Similarly, if there are no other lane lines outside the rightmost lane line of Road 2, then this rightmost lane line is the right boundary of Road 2. If there are other lane lines outside the leftmost lane line of Road 2, then this leftmost lane line cannot be considered the left boundary of Road 2. If the rightmost lane line of Road 1 and the leftmost lane line of Road 2 are the same lane line, and the vehicle trajectories on both sides of this lane line have opposite directions of travel, then this lane line can be considered the boundary of the up and down lanes.

[0121] According to one embodiment of this disclosure, a method for updating map vector data is also proposed, the method comprising:

[0122] Obtain the road boundary of the target road, which is determined based on the road boundary determination method described above;

[0123] Update the map vector data using the road boundaries.

[0124] In this embodiment, the server, through... Figure 1 The road boundary determination method shown can determine the road boundary. Since the road boundary is a known lane line, the vector data of the road boundary, including its position coordinates and direction, is also known. Therefore, existing map vector data can be updated based on the acquired road boundary vector data for subsequent map generation or updates. For example, when the position coordinates of a road boundary change, the map vector data can be updated based on the identified position coordinates of the road boundary. Because this disclosure uses vehicle trajectory analysis to determine road boundaries, changes in real-world road boundaries can be detected promptly and at low cost, ensuring the freshness of map vector data.

[0125] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein.

[0126] Figure 3This diagram illustrates a structural block diagram of a road boundary determination device according to an embodiment of the present disclosure. This device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 3 As shown, the road boundary determination device includes:

[0127] The acquisition module 301 is configured to acquire at least one vehicle driving trajectory and multiple lane lines on the road where the vehicle driving trajectory is located;

[0128] The lane determination module 302 is configured to determine the lane where the vehicle's driving trajectory is located based on the distance relationship between the vehicle's driving trajectory and the lane lines;

[0129] Clustering module 303 is configured to cluster the vehicle driving trajectories based on the distance and driving direction between the vehicle driving trajectories;

[0130] The road boundary determination module 304 is configured to determine the road boundary of the road corresponding to the lane based on the clustering results and the lane where the vehicle's driving trajectory is located, which is clustered into one class.

[0131] The aforementioned road boundary determination device may be a server, a physical device capable of providing server functions, a component (such as a chip) configured in a physical device, or a module capable of implementing some or all of the server functions. This disclosure does not limit the specific device.

[0132] In this embodiment, the vehicle trajectory can be the path taken by the data acquisition vehicle on the road, and the trajectory has a corresponding direction of travel. The data acquisition vehicle can collect the vehicle trajectory through INS or GPS devices deployed on the vehicle and upload the trajectory to the road boundary determination device. The vehicle trajectory includes multiple trajectory points, each with corresponding coordinates, which can be determined based on a trajectory coordinate system. This trajectory coordinate system is a coordinate system deployed on the data acquisition vehicle to determine the coordinates of the trajectory points. Different data acquisition vehicles may use the same or different trajectory coordinate systems. The data acquisition vehicle can be a professional data acquisition vehicle equipped with high-performance acquisition equipment or a regular data acquisition vehicle.

[0133] In some embodiments, the road boundary determination device can acquire at least one vehicle trajectory uploaded by a data collection vehicle. In some embodiments, the road boundary determination device can acquire multiple vehicle trajectories uploaded by one or more data collection vehicles on the same road or different roads. Multiple vehicle trajectories on the same road may include vehicle trajectories in the same lane or different lanes.

[0134] In some embodiments, when at least one vehicle trajectory acquired by the road boundary determination device corresponds to different trajectory coordinate systems, the vehicle trajectories corresponding to different trajectory coordinate systems can also be uniformly converted to the same trajectory coordinate system by using the identifier of the trajectory coordinate system corresponding to the vehicle trajectory uploaded by the data acquisition vehicle.

[0135] While the data collection vehicle generates vehicle trajectories as it travels on the road, it can also collect road images through cameras deployed on the vehicle. Existing simple image recognition technology is used to identify vector data such as lane lines on the road containing the vehicle's trajectory from the collected road images. This vector data can be uploaded to the road boundary determination device along with the vehicle's trajectory. It should be noted that in some embodiments, the road boundary determination device can also obtain at least one vehicle trajectory and multiple lane lines and other vector data from a third party. Embodiments of this disclosure can obtain the aforementioned vehicle trajectories and lane line vector data from at least one or more third parties, and then determine the road boundary based on these vehicle trajectories and lane lines. When the trajectory coordinate systems corresponding to vehicle trajectories generated by different third parties are different, the vehicle trajectories obtained from different third parties can be uniformly converted to the same trajectory coordinate system based on the identifier and parameters of the trajectory coordinate system corresponding to the vehicle trajectories obtained from the third party.

[0136] In some embodiments, at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located may include vehicle trajectory lines and lane lines on multiple roads within a region whose location is known.

[0137] In some embodiments, lane lines on a road may include, but are not limited to, the position coordinates of various shaped points on the lane lines, and the coordinate system corresponding to the position coordinates of various shaped points on the lane lines and the trajectory coordinate system of the vehicle's driving trajectory may be the same coordinate system. For example, both the lane line coordinate system and the trajectory coordinate system are world coordinate systems.

[0138] In this embodiment of the disclosure, a lane may include a left lane line and a right lane line. It is understood that the vehicle's trajectory line is typically located between the left and right lane lines of the lane. Therefore, the lane containing the vehicle's trajectory can be obtained from multiple lane lines by using the distance relationship between the lane lines and the vehicle's trajectory line on geographic coordinates. In some embodiments, the lane containing the vehicle's trajectory may include the left lane line closest to the vehicle's trajectory and the right lane line closest to the vehicle's trajectory. In some embodiments, the distance between the lane line and the vehicle's trajectory on geographic coordinates can be understood as the distance between the geographical location of the lane line in the actual scenario and the geographical location of the vehicle's trajectory in the actual scenario.

[0139] After acquiring at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located, the road boundary determination device can further determine the lane where each vehicle trajectory is located. When determining the lane where a vehicle trajectory is located, the distance between the vehicle trajectory and each of the multiple lane lines can be determined. Based on the distance relationship, the left lane line closest to the left side of the vehicle trajectory and the right lane line closest to the right side of the vehicle trajectory can be determined from the multiple lane lines. The determined left and right lane lines closest to the vehicle trajectory constitute the lane where the vehicle trajectory is located. In some embodiments, the distance between the vehicle trajectory and the lane lines can be determined by projecting each trajectory point on the vehicle trajectory onto the lane lines on the left and right sides, and the average distance between each trajectory point and the corresponding projection point can be used to determine the distance between the vehicle trajectory and the lane lines.

[0140] In some embodiments, the road boundary determination device can also associate the vehicle's trajectory with the lane in which the vehicle's trajectory lies. In other words, the road boundary determination device can record the correspondence between the vehicle's trajectory and the nearest left lane line and the nearest right lane line.

[0141] In some embodiments, each vehicle trajectory can be associated with a lane, and each lane can be associated with one or more vehicle trajectories.

[0142] In this embodiment of the disclosure, after determining the lane where each vehicle trajectory is located, the road boundary determination device can further cluster the acquired multiple vehicle trajectories. When clustering multiple vehicle trajectories, clustering can be performed based on the distance between the vehicle trajectories and the direction of travel of the vehicle trajectories. By clustering the vehicle trajectories, at least one class of vehicle trajectories can be obtained.

[0143] After clustering vehicle trajectories to obtain at least one class of vehicle trajectories, the road boundary determination device can determine the lanes associated with each class of vehicle trajectories. These lanes constitute a road. Furthermore, the road boundary can be determined based on the left and right lane lines corresponding to the multiple lanes contained in the road.

[0144] In some embodiments, the road boundary determination device can associate a lane with the road in which the lane is located. In other words, the road boundary determination device can record the correspondence between lanes and roads.

[0145] This disclosure proposes a road boundary determination device. The device acquires at least one vehicle trajectory and multiple lane lines on the road containing the vehicle trajectory. Based on the distance relationship between the vehicle trajectory and the lane lines, it determines the lane containing the vehicle trajectory. Then, based on the distance and direction of travel between the vehicle trajectories, it clusters the vehicle trajectories. Finally, based on the clustering results and the lanes containing the clustered vehicle trajectories, it determines the road boundary corresponding to each lane. This disclosure determines road boundaries based on the correlation between vehicle trajectories and lane lines. Compared to methods that use complex algorithms to identify road boundaries from images, this method has lower complexity and, when the number of vehicle trajectories is sufficient and the positioning is accurate, it achieves high accuracy in detecting road boundaries and is less prone to missed detections.

[0146] In an optional implementation of this embodiment, the clustering module 303 is configured to cluster vehicle trajectories that are less than or equal to a preset distance threshold and have the same driving direction into one class.

[0147] In this optional implementation, when clustering the acquired vehicle driving trajectories, the clustering module 303 can group vehicle driving trajectories that are close in distance and have the same driving direction into one category based on the distance and driving direction between the vehicle driving trajectories.

[0148] In some embodiments, vehicle trajectories clustered into one class, i.e., a class of vehicle trajectories, can satisfy the following conditions: 1) the vehicle trajectories clustered into one class have the same direction of travel; 2) among the vehicle trajectories clustered into one class, the distance between two adjacent vehicle trajectories is less than or equal to a preset distance threshold, which can be determined by those skilled in the art according to actual needs. It is understood that vehicle trajectories can be clustered based on the above clustering conditions.

[0149] In some embodiments, if there are individual vehicle trajectories whose driving direction differs from that of the other vehicle trajectories within a cluster of vehicle trajectories, these individual vehicle trajectories can be filtered out. Similarly, if the distance between individual vehicle trajectories exceeds a preset distance threshold, vehicle trajectories that are far from the majority of vehicle trajectories can also be filtered out.

[0150] In some embodiments, the lanes associated with vehicle travel trajectories grouped into one class may be one or more. In some embodiments, each lane corresponds to a road, and each road includes at least one lane.

[0151] In some embodiments, the travel direction of the lane associated with the grouped vehicle travel trajectories can be marked based on the travel direction of the grouped vehicle travel trajectories, and the travel directions of the two can be associated. The travel direction of the grouped vehicle travel trajectories is the same as the travel direction of the lane associated with the grouped vehicle travel trajectories.

[0152] In some embodiments, the driving direction of the vehicle driving trajectories clustered into one group can be the driving direction of any vehicle driving trajectory within that group. Alternatively, a target vehicle driving trajectory can be determined from the clustered vehicle driving trajectories, and the driving direction of that target vehicle driving trajectory can be used as the driving direction of the vehicle driving trajectories clustered into one group. The target vehicle driving trajectory can be the cluster center selected during clustering, or the vehicle driving trajectory with the highest positioning accuracy within that group.

[0153] In an optional implementation of this embodiment, the road boundary determination module 304 includes:

[0154] The first lane determination submodule is configured to determine the corresponding lane based on the vehicle driving trajectory that is clustered into one class in the clustering results, wherein the driving direction of the lane is the same as the driving direction of the vehicle driving trajectory on the lane.

[0155] The road determination submodule is configured to determine the road corresponding to a lane based on one or more lanes with the same direction of travel.

[0156] The road boundary determination submodule is configured to determine the road boundary from the leftmost lane line and the rightmost lane line of the road.

[0157] In this optional implementation, after clustering the acquired vehicle driving trajectories based on the clustering module 303 to obtain at least one type of vehicle driving trajectory, the road corresponding to the lane of each type of vehicle driving trajectory and the road boundary of that road can be further determined.

[0158] For any one class of vehicle trajectories obtained from clustering, one or more lanes associated with that class of vehicle trajectories can be determined. Since the vehicles in that class of trajectories travel in the same direction, the one or more lanes associated with them also travel in the same direction, thus forming a road. Therefore, this method can be used to determine the road corresponding to the lanes associated with each class of vehicle trajectories. In other words, the road corresponding to each lane can be determined.

[0159] When a road includes more than one lane, the road can meet the following conditions: 1) the lanes of the more than one lane have the same direction of travel; 2) the lateral distances between the lanes of the more than one lane are similar.

[0160] In some embodiments, the lateral distance between lanes may be within a road width threshold range.

[0161] In some embodiments, the lateral distance between lanes can be the distance between the center lines of the two lanes, which can be the center line of the lane lines on the left and right sides of the lane.

[0162] In some embodiments, the lateral distance between lanes can also be the lateral distance between the centerlines of the two lanes. For example, based on the projection of various shape points on the centerline of one lane onto the centerline of another lane, the average of the projection distances between the projection points corresponding to each shape point and the shape point itself can be used as the lateral distance between the centerlines of the two lanes, that is, the lateral distance between the two lanes. In other words, based on the above two conditions, multiple lanes with the same travel direction and similar lateral distances can be identified as multiple lanes corresponding to the same road.

[0163] In some embodiments, when the lanes containing the travel trajectories of two types of vehicles have the same travel direction and the lateral distance between the lanes is similar, a corresponding road can be determined based on the lanes containing the travel trajectories of these two types of vehicles. That is, if the travel direction of the lane containing the travel trajectories of one type of vehicle is the same as that of the lane containing the travel trajectories of another type of vehicle, and the lateral distance between the lanes containing the travel trajectories of one type of vehicle and the lane containing the travel trajectories of another type of vehicle is similar, then a corresponding road can be determined based on the lanes containing the travel trajectories of these two types of vehicles as a whole.

[0164] Once the road is identified, its boundaries can be determined based on the leftmost and rightmost lane lines. The leftmost lane line corresponds to the leftmost lane of the road, and the rightmost lane line corresponds to the rightmost lane of the road.

[0165] It is understood that if a lane has no other adjacent lanes, the road corresponding to that lane may be a single-lane road. In this case, the lane can be associated with that road, and the left and right boundaries of the road can be determined based on the left and right lane lines of that lane. In some embodiments, the left lane line of the lane can be determined as the left boundary of the road, and the right lane line of the lane can be determined as the right boundary of the road. In an optional implementation of this embodiment, the road boundary determination submodule includes:

[0166] The left boundary determination submodule is configured to determine the leftmost lane line as the left boundary of the road when there is no lane line within a preset distance outside the leftmost lane line.

[0167] The right boundary determination submodule is configured to determine the rightmost lane line as the right boundary of the road when there is no lane line within a preset distance outside the rightmost lane line.

[0168] In this optional implementation, the left and right boundaries of a road can be determined from the leftmost and rightmost lane lines associated with one or more lanes corresponding to the same road. To ensure the accuracy of the road boundaries, the third determining submodule can further determine whether there are other lane lines within a preset distance range outside the leftmost lane line. If no other lane lines exist, the leftmost lane line can be used as the left boundary of the road; if other lane lines exist, the leftmost lane line may not be the left boundary of the road. Furthermore, the fourth determining submodule can further determine whether there are other lane lines within a preset distance range outside the rightmost lane line. If no other lane lines exist, the rightmost lane line can be used as the right boundary of the road; if other lane lines exist, the rightmost lane line may not be the right boundary of the road. This is because the lane lines associated with the lanes of the road are determined based on the vehicle's driving trajectory. If the collected vehicle driving trajectory does not include the trajectory of driving in the leftmost and / or rightmost lanes of the road, then the leftmost and / or rightmost lanes of the road are not detected and are not associated with the road. Therefore, the leftmost and / or rightmost lane lines of the associated lanes cannot be determined as the left and / or right boundaries of the road.

[0169] It should be understood that those skilled in the art can set a preset distance range according to actual needs. For example, the preset distance range can be a distance range greater than or equal to the width of one lane. This disclosure does not limit this range, and all such ranges are within the scope of protection.

[0170] In an optional implementation of this embodiment, the apparatus further includes:

[0171] The up and down road boundary determination module is configured to determine multiple lanes corresponding to two types of vehicle travel trajectories with different travel directions based on clustering results; if two lanes with different travel directions are associated with the same lane line, then the same lane line is determined as the up and down road boundaries of the roads corresponding to the two lanes with different travel directions.

[0172] In this optional implementation, after clustering the acquired vehicle driving trajectories, if multiple types of vehicle driving trajectories are obtained, after determining the road corresponding to the lane of each type of vehicle driving trajectory, the up and down road boundaries of the road can be further determined based on the up and down road boundary determination module.

[0173] The road boundary for both directions can meet the following conditions: 1) The road boundary is a lane line shared by two adjacent roads, that is, one lane on one of the two adjacent roads and one lane on the other road share the same lane line; 2) The vehicle trajectories on both sides of the road boundary are opposite, that is, the vehicle trajectories on the two adjacent roads are in opposite directions, or the lanes on the two adjacent roads are in opposite directions.

[0174] The server clusters all vehicle trajectories to obtain multiple categories of vehicle trajectories. One category of vehicle trajectory corresponds to one lane, while multiple categories of vehicle trajectories can determine multiple different lanes. Once the lanes are determined, their lane lines can also be determined. If there are two lanes with different travel directions among the multiple lanes, they are associated with the same lane line. Since these two lanes with different travel directions belong to different roads, the same lane line can be considered as the up and down road boundaries of the roads corresponding to the two lanes respectively.

[0175] In some embodiments, the road boundary determination device may record a correspondence table between vehicle travel trajectory, lane lines, lanes, and roads. Therefore, to improve efficiency, the up and down road boundaries can be determined using this correspondence table.

[0176] In an optional implementation of this embodiment, the lane determination module 302 includes:

[0177] The lane line determination submodule is configured to determine, based on the distance relationship between the trajectory points on each of the vehicle's driving trajectory and each of the lane lines, the left lane line and the right lane line that are closest to the left and right sides of each of the multiple lane lines;

[0178] The second lane determination submodule is configured to determine the lane where each of the vehicle's travel trajectories is located based on the left lane line and the right lane line.

[0179] In this optional implementation, after obtaining at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located through the acquisition module 301, the lane line determination submodule and the second lane determination submodule can determine the lane where each vehicle trajectory is located. That is, the lane line determination submodule can determine the nearest left lane line and the nearest right lane line to each vehicle trajectory based on the distance between the trajectory points on each vehicle trajectory and each lane line, and the second lane determination submodule can determine the lane where each vehicle trajectory is located based on the determined left lane line and right lane line.

[0180] For any vehicle trajectory, the lane determination submodule can determine the nearest left lane line and the nearest right lane line to the vehicle trajectory through the following processes 1) to 3).

[0181] 1) Based on the distance relationship between the trajectory points and lane lines on the vehicle's driving trajectory in the geographical coordinates, determine the left lane line and the right lane line that are closest to the trajectory point from multiple lane lines.

[0182] In this embodiment of the disclosure, since the vehicle's driving trajectory is composed of multiple trajectory points, the lane line determination submodule can determine the nearest left lane line and the nearest right lane line to each trajectory point, that is, determine the lane where each trajectory point is located.

[0183] In some embodiments, for any trajectory point on the vehicle's driving trajectory, the nearest left lane line and the nearest right lane line to the trajectory point can be determined as follows: determine the distance between the trajectory point and each of the multiple lane lines; based on the distance, determine the nearest left lane line and the nearest right lane line to the trajectory point from the multiple lane lines.

[0184] In some embodiments, for any trajectory point on the vehicle's driving trajectory and any lane line among multiple lane lines, the distance between the trajectory point and the lane line can be determined as follows: determine the perpendicular from the trajectory point on the vehicle's driving trajectory to the lane line, and obtain the geographic coordinates corresponding to the perpendicular; determine the distance between the trajectory point and the lane line based on the trajectory point coordinates and the geographic coordinates of the perpendicular.

[0185] In some embodiments, the location of the lane line to the left or right of the trajectory point can be determined based on the relative position of the lane line and the trajectory point on the vehicle's driving trajectory in geographic coordinates.

[0186] 2) When multiple left lane lines closest to multiple trajectory points on the vehicle's trajectory are the same lane line, and multiple right lane lines closest to multiple trajectory points on the vehicle's trajectory are the same lane line, the left lane line shall be regarded as the left lane line closest to the vehicle's trajectory, and the right lane line shall be regarded as the right lane line closest to the vehicle's trajectory.

[0187] In this embodiment of the disclosure, after the lane line determination submodule determines the nearest left lane line and the nearest right lane line to each trajectory point on the vehicle's travel trajectory, if the nearest left lane line to all trajectory points on the vehicle's travel trajectory is the same left lane line, and the nearest right lane line to all trajectory points on the vehicle's travel trajectory is also the same right lane line, it indicates that the data collection vehicle is traveling in the same lane and has not changed lanes. Therefore, the same left lane line closest to all trajectory points can be taken as the nearest left lane line to the vehicle's travel trajectory, and the same right lane line closest to all trajectory points can be taken as the nearest right lane line to the vehicle's travel trajectory, thus determining the lane where the vehicle's travel trajectory is located.

[0188] 3) When the two left lane lines closest to two adjacent trajectory points on the vehicle's trajectory are different left lane lines, and / or the two right lane lines closest to two adjacent trajectory points on the vehicle's trajectory are different right lane lines, the vehicle's trajectory is divided into two vehicle trajectories, each vehicle trajectory corresponding to one of the two left lane lines, and / or each vehicle trajectory corresponding to one of the two right lane lines.

[0189] In one embodiment of this disclosure, after the lane line determination submodule determines the nearest left lane line to each trajectory point on the vehicle's driving trajectory, if any two adjacent trajectory points among the multiple trajectory points are not matched by the same nearest left lane line, the vehicle's driving trajectory can be divided into two independent vehicle driving trajectories based on the location of these two adjacent trajectory points, with each vehicle driving trajectory matched by a nearest left lane line.

[0190] In another scenario, after the lane line determination submodule determines the nearest right lane line to each trajectory point on the vehicle's trajectory, if any two adjacent trajectory points are not matched by the same nearest right lane line, the vehicle's trajectory can be divided into two independent vehicle trajectories based on the locations of these two adjacent trajectory points, with each vehicle trajectory matched by a nearest right lane line.

[0191] In both of the above situations, since the vehicle's driving trajectory is divided into two independent vehicle driving trajectories, with one side of these two vehicle driving trajectories corresponding to the same lane line and the other side corresponding to different lane lines, the corresponding lane line can also be divided into two independent lane lines, with each lane line corresponding to one vehicle driving trajectory. In this way, the lane corresponding to each of these two vehicle driving trajectories can be determined.

[0192] In another scenario, after the lane determination submodule determines the nearest left lane line and the nearest right lane line to each trajectory point on the vehicle's trajectory, if any two adjacent trajectory points are matched with a left lane line that is not the same left lane line, and the matched right lane line that is not the same right lane line, then the vehicle's trajectory is divided into two independent vehicle trajectories based on the positions of these two adjacent trajectory points. Each of these two vehicle trajectories is matched with a nearest left lane line and a nearest right lane line. The nearest left lane line and the nearest right lane line matched by each vehicle trajectory can constitute the lane where the vehicle's trajectory is located.

[0193] In short, when the lane lines matching consecutive trajectory points on a vehicle's trajectory change, the vehicle's trajectory can be truncated.

[0194] In some embodiments, the vehicle trajectory needs to be truncated when the lane lines matched by consecutive trajectory points on the vehicle trajectory change. Therefore, regardless of whether the vehicle trajectory is truncated, consecutive trajectory points on each vehicle trajectory can be fitted into a new vehicle trajectory, and an index between the vehicle trajectory and the lane can be established using a K-dimensional tree or a Rectangle Tree.

[0195] In some embodiments, the correspondence between trajectory points on each vehicle's driving trajectory and the nearest left lane line and the nearest right lane line to the trajectory point can also be recorded. That is, the correspondence between trajectory points on each vehicle's driving trajectory and the lanes where the trajectory points are located can be recorded.

[0196] In an optional implementation of this embodiment, the apparatus further includes:

[0197] The filtering module is configured to acquire the positioning data of the vehicle's driving trajectory; and based on the positioning data, filter out vehicle driving trajectories that do not meet the positioning accuracy requirements.

[0198] In this optional implementation, to improve the efficiency and accuracy of road boundary determination, a filtering module can be used to filter out vehicle trajectories. Specifically, the filtering module acquires the positioning data of the vehicle trajectories and determines the trajectory positioning quality of each trajectory. Based on the trajectory positioning quality, low-quality vehicle trajectories that do not meet the positioning accuracy requirements are filtered out from at least one acquired vehicle trajectory, while high-quality vehicle trajectories that meet the positioning accuracy requirements are retained.

[0199] In some embodiments, the positioning data of the vehicle's driving trajectory can be uploaded by the data acquisition vehicle to the road boundary determination device.

[0200] In some embodiments, the location data may include real-time kinematic (RTK) location status and / or trajectory location reliability.

[0201] RTK positioning, based on carrier phase measurement, is a real-time differential GPS measurement that provides real-time positioning results for vehicle trajectories. Different operating modes can be used for RTK positioning of vehicle trajectories. For example, single-point positioning can be used. The RTK positioning status characterizes the positioning accuracy of the vehicle trajectory under a specific operating mode. Trajectory positioning reliability can be used to characterize the positioning accuracy of the vehicle trajectory.

[0202] In some embodiments, different RTK positioning states correspond to different trajectory positioning qualities, and / or different trajectory positioning confidence levels correspond to different trajectory positioning qualities. It should be understood that the correspondence between positioning data and trajectory positioning quality can be predetermined, and this disclosure does not limit the specific correspondence, all of which are within the protection scope of this disclosure.

[0203] When the positioning data includes RTK positioning status, the trajectory positioning quality can be the positioning accuracy of RTK positioning of the vehicle's driving trajectory in a certain working mode; when the positioning data includes trajectory positioning reliability, the trajectory positioning quality can be the positioning accuracy of the vehicle's driving trajectory; when the positioning data includes both RTK positioning status and trajectory positioning reliability, the trajectory positioning quality can be comprehensively determined by combining weights.

[0204] In some embodiments, the positioning accuracy requirement can be that the positioning accuracy of the vehicle's driving trajectory is at the decimeter or meter level or higher. It is understood that the filtering module can retain vehicle driving trajectories with positioning accuracy at the decimeter or meter level or higher as high-quality vehicle driving trajectories, and filter out vehicle driving trajectories with positioning accuracy at the decimeter or meter level or lower as low-quality vehicle driving trajectories.

[0205] In some embodiments, to further improve the efficiency and accuracy of road boundary determination, the filtering module may further include the following process: obtaining the vehicle speed of the vehicle's driving trajectory; and filtering out vehicle driving trajectories that do not conform to the vehicle's driving pattern based on the vehicle speed. For example, vehicle driving trajectories with excessively fast or slow speeds can be filtered out as low-quality vehicle driving trajectories, while vehicle driving trajectories with constant speed or certain acceleration can be retained as high-quality vehicle driving trajectories.

[0206] According to one embodiment of this disclosure, a map vector data updating apparatus is also provided, the apparatus comprising:

[0207] The road boundary acquisition module is configured to acquire the road boundary of the target road, the road boundary being determined based on the aforementioned road boundary determination device;

[0208] The update module is configured to update map vector data using the road boundaries.

[0209] In this embodiment, by Figure 3 The road boundary determination device shown can determine the road boundary. Since the road boundary is a known lane line, the vector data of the road boundary, including its position coordinates and direction, is also known. Therefore, existing map vector data can be updated based on the acquired road boundary vector data for subsequent map generation or updates. For example, when the position coordinates of a road boundary change, the map vector data can be updated based on the identified position coordinates of the road boundary. Because this disclosure uses vehicle travel trajectories to analyze road boundaries, changes in real-world road boundaries can be detected in a timely and cost-effective manner, ensuring the freshness of map vector data.

[0210] Figure 4 This is a schematic diagram of the structure of an electronic device suitable for implementing a road boundary determination method and / or a map vector data update method according to an embodiment of the present disclosure.

[0211] like Figure 4 As shown, the electronic device 400 includes a processing unit 401, which can be implemented as a CPU, GPU, FPGA, NPU, or other processing unit. The processing unit 401 can execute various processes according to any of the methods described above in this disclosure, based on a program stored in the read-only memory (ROM) 402 or a program loaded from the storage portion 408 into the random access memory (RAM) 403. The RAM 403 also stores various programs and data required for the operation of the electronic device 400. The processing unit 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0212] The following components are connected to I / O interface 405: an input section 406 including a keyboard, mouse, etc.; an output section 407 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card, modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to I / O interface 405 as needed. A removable medium 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 410 as needed so that computer programs read from it can be installed into storage section 408 as needed.

[0213] In particular, according to embodiments of this disclosure, any of the methods described above in the embodiments of this disclosure can be implemented as a computer software program. For example, embodiments of this disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program containing program code for performing any of the methods in the embodiments of this disclosure. In such an embodiment, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411.

[0214] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0215] The units or modules described in the embodiments of this disclosure can be implemented in software or hardware. The described units or modules can also be located in a processor, and the names of these units or modules do not necessarily constitute a limitation on the unit or module itself.

[0216] In another aspect, this disclosure also provides a computer-readable storage medium, which may be a computer-readable storage medium included in the apparatus described in the above embodiments; or it may be a standalone computer-readable storage medium not assembled into a device. The computer-readable storage medium stores one or more programs that are used by one or more processors to perform the methods described in this disclosure.

[0217] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A method for determining road boundaries, wherein, include: Obtain at least one vehicle trajectory and multiple lane lines on the road where the vehicle trajectory is located, wherein the lane lines are known vector data; Based on the distance relationship between the vehicle's trajectory and the lane lines, the lane in which the vehicle's trajectory is located is determined; The vehicle trajectories are clustered based on the distance and direction of travel between them; The road boundary of the road corresponding to the lane is determined based on the clustering results and the lanes where the vehicle trajectories are located in the same cluster.

2. The method according to claim 1, wherein, Based on the distance and direction of travel between the vehicle trajectories, the vehicle trajectories are clustered, including: Vehicle trajectories that are less than or equal to a preset distance threshold and travel in the same direction are grouped into one category.

3. The method according to claim 1 or 2, wherein, Based on the clustering results and the lanes where the vehicle trajectories clustered into one class are located, the road boundary of the road corresponding to the lane is determined, including: The corresponding lane is determined based on the vehicle trajectories that are clustered into one class in the clustering results, and the driving direction of the lane is the same as the driving direction of the vehicle trajectories in the lane. Based on one or more lanes with the same direction of travel, determine the road corresponding to the lane; The road boundary is determined from the leftmost and rightmost lane lines of the road.

4. The method according to claim 3, wherein, Determining the road boundary from the leftmost and rightmost lane lines of the road includes: When there is no lane line within a preset distance outside the leftmost lane line, the leftmost lane line is determined as the left boundary of the road. When there is no lane line within a preset distance outside the rightmost lane line, the rightmost lane line is determined as the right boundary of the road.

5. The method according to any one of claims 1-2 and 4, wherein, The method further includes: Based on the clustering results, the different lanes where the vehicle trajectories of the vehicles that were clustered into different classes were located were determined. If two lanes with different driving directions are associated with the same lane line, then the same lane line is determined as the up and down road boundaries of the roads corresponding to the two lanes with different driving directions.

6. The method according to any one of claims 1-2, 4, wherein, Determining the lane where the vehicle's trajectory is located based on the distance relationship between the vehicle's trajectory and the lane lines includes: Based on the distance relationship between the trajectory points on each vehicle's driving trajectory and each lane line, the left lane line and the right lane line that are closest to the left and right sides of each vehicle's driving trajectory are determined from the plurality of lane lines; The lane in which each vehicle's trajectory is located is determined based on the left lane line and the right lane line.

7. The method according to any one of claims 1-2 and 4, wherein, Before determining the lane where the vehicle's trajectory is located based on the distance relationship between the vehicle's trajectory and the lane lines, the method further includes: Obtain the location data of the vehicle's driving trajectory; Based on the positioning data, vehicle trajectories that do not meet the positioning accuracy requirements are filtered out.

8. A method for updating map vector data, wherein, include: Obtain the road boundary of the target road, wherein the road boundary is determined based on the method of any one of claims 1 to 7; Update the map vector data using the road boundaries.

9. A road boundary determination device, wherein, include: The acquisition module is configured to acquire at least one vehicle driving trajectory and multiple lane lines on the road where the vehicle driving trajectory is located, wherein the lane lines are known vector data. The lane determination module is configured to determine the lane where the vehicle's trajectory is located based on the distance relationship between the vehicle's trajectory and the lane lines; The clustering module is configured to cluster the vehicle's driving trajectory based on the vehicle's driving trajectory and driving direction; The road boundary determination module is configured to determine the road boundary of the road corresponding to the lane based on the clustering results and the lanes where the vehicle trajectories are located, which are clustered into one class.

10. An electronic device, wherein, The method includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method of any one of claims 1-8.

11. A computer-readable storage medium having computer instructions stored thereon, wherein, When executed by a processor, the computer instructions implement the method described in any one of claims 1-8.

Images