Lane Line Map Construction and Update Method, System, Medium and Device

Abstract

The present application discloses a lane line map construction and update method, system, medium and device, belonging to the field of autonomous driving technology, and can be applied to scenarios such as ports, highways, logistics, mines, closed parks, or urban traffic. The method includes: the server intercepts the current lane lines obtained by the vehicle end to obtain a plurality of current lane line segments; in the lane line database, extracts the corresponding lane line segments at the corresponding positions of the current lane line segments; extracts a plurality of lane line points on the current lane line segments and the corresponding lane line segments respectively; performs quadratic curve fitting on the plurality of lane line points, and intercepts the fitted lane line curve to obtain updated lane line segments; uses the updated lane line segments to update the lane line database with the corresponding lane line segments. The present application processes the lane line data to establish and update the lane line database, and the processing process is more convenient and has higher accuracy at the same time.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular, to a method, system, medium, and device for constructing and updating a lane map. Background Art

[0002] A high-precision map is actually a dedicated map for an autonomous driving system as opposed to an ordinary navigation electronic map. Traditional high-precision map construction methods require expensive dedicated collection for data acquisition and a large amount of manual annotation of the acquired data. Therefore, the traditional map construction method has a high cost and slow update speed. To solve the problems of the cost and efficiency of high-precision map construction and update, a crowdsourcing map construction method has emerged. Crowdsourcing map construction refers to deploying multiple vehicles with environmental perception capabilities, allowing them to collect road information data while driving and upload it to the cloud. The cloud constructs a highly restored and instantaneously updated driving map based on the feedback data. The mainstream crowdsourcing map construction method is to use a large number of passenger vehicles, such as private cars, taxis, etc., and configure low-cost sensors such as dash cams, mobile phones, etc. By collecting a large amount of low-precision data, a fused high-precision-like map is obtained through fusion in the cloud. For the construction of the lane map, generally in places where there is a high-precision map, the collected fragmented data is matched with the original high-precision map to solve the lane map update problem in the high-precision map. In places where there is no high-precision map, by accumulating a sufficient amount of collected data, lane lines are extracted according to the Gaussian model.

[0003] For a vehicle with autonomous driving capabilities, the vehicle itself is already equipped with high-precision sensors, such as RTK, lidar, etc. Generally, the accuracy of the reconstructed lane lines is relatively high, within the meter level. Therefore, in the crowdsourcing map construction method, using vehicles with autonomous driving capabilities for data collection has high data quality, but the number of collections is small. Therefore, the original lane line fusion method is not applicable to such data. Existing technologies generally require a high-precision map as the base map. The crowdsourcing collection method generally solves the problem of high-precision map update. If there is no high-precision map, generally, a sufficient amount of data needs to be collected before cloud fusion can be performed. Currently, there are not many autonomous vehicles, and the number of collections will not be large. However, due to the high-precision positioning of the vehicles, the accuracy of a single collection is relatively high. Summary of the Invention

[0004] This application proposes a method, system, medium, and device for constructing and updating a lane map. By processing the road lane line data collected by high-precision positioning vehicles, a lane line database is established and updated, without the need for a high-precision map as the base map, nor a large amount of perception data. The processing process is more convenient, and at the same time, it has a relatively high accuracy.

[0005] In a first aspect, the present application provides a method for constructing and updating a lane line map, including: the cloud server intercepts the current lane line obtained by the vehicle end according to a first preset length to obtain a plurality of current lane line segments; in the lane line database, extracts the corresponding lane line segments at the corresponding positions of the current lane line segments; extracts a plurality of lane line points on the current lane line segments and the corresponding lane line segments respectively; performs quadratic curve fitting on the plurality of lane line points, and intercepts the fitted lane line curve according to the first preset length to obtain updated lane line segments; uses the updated lane line segments to replace the corresponding lane line segments to update the lane line database.

[0006] Optionally, in the lane line database, extracting the original lane line segments at the corresponding positions of the current lane line segments includes: calculating the distances between the current lane line segments and each of the original lane line segments in the lane line database respectively; if the distance is less than a first preset threshold, determining the corresponding original lane line segment as the corresponding lane line segment.

[0007] Optionally, calculating the distances between the current lane line segments and each of the original lane line segments in the lane line database respectively includes: extracting a plurality of judgment points on the current lane line segments according to a second preset length; drawing perpendicular lines perpendicular to the current lane line segments at each judgment point; if there are perpendicular lines intersecting with the original lane line segments, taking the average length of each intersecting line segment as the distance between the current lane line and the corresponding original lane line segment; if there are no perpendicular lines intersecting with the original lane line segments, the distance between the current lane line segment and the original lane line segment is infinite.

[0008] Optionally, extracting a plurality of lane line points on the current lane line segments and the corresponding lane line segments respectively includes: extracting a plurality of lane line points on the corresponding lane line segments according to a third preset length; on the current lane line segments, extracting a plurality of lane line points according to the update weight, where the lower the update weight, the larger the spacing of the lane line points extracted on the current lane line segments.

[0009] Optionally, extracting a plurality of lane line points according to the update weight includes: taking the ratio of the third preset length to the update weight as the spacing of the lane line points extracted on the current lane line segments; extracting a plurality of lane line points according to the spacing.

[0010] Optionally, it further includes: if there is no corresponding lane line segment at the corresponding position of the current lane line segment in the lane line database, directly adding the current lane line segment into the lane line database.

[0011] Optionally, it further includes: in the lane line database, splicing a plurality of lane line segments with an end point distance less than a second preset threshold to obtain a complete lane line.

[0012] In a second aspect, the present application provides a lane line map construction and update system, including: a module for intercepting the current lane line obtained from the vehicle end according to a first preset length in a cloud server to obtain a plurality of current lane line segments; a module for extracting corresponding lane line segments at corresponding positions of the current lane line segments in a lane line database; a module for respectively extracting a plurality of lane line points on the current lane line segments and the corresponding lane line segments; a module for performing quadratic curve fitting on the plurality of lane line points and intercepting the fitted lane line curve according to the first preset length to obtain updated lane line segments; and a module for using the updated lane line segments to replace the corresponding lane line segments to update the lane line database.

[0013] In a third aspect, the present application provides a computer-readable storage medium storing computer instructions, where the computer instructions are operated to execute the lane line map construction and update method in the solution of the first aspect.

[0014] In a fourth aspect, in a technical solution of the present application, a computer device is provided, which includes a processor and a memory, and the memory stores computer instructions, where: the processor operates the computer instructions to execute the lane line map construction and update method in the solution of the first aspect.

[0015] The beneficial effects of the present application are as follows: By processing the road lane line data collected by high-precision positioning vehicles, the present application establishes and updates the lane line database, without the need for a high-precision map as a base map, nor a large amount of perception data. The processing process is simpler, and at the same time, it has higher accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. The drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 is a schematic flowchart of an implementation manner of the lane line map construction and update method of the present application;

[0018] Figure 2 is a schematic diagram for calculating the distance between the current lane line segment and the original lane line segment;

[0019] Figure 3 is a schematic diagram of the matching result between a current lane line segment and an original lane line segment;

[0020] Figure 4 is a schematic diagram of an example of lane line point extraction;

[0021] Figure 5 It is a schematic diagram of an example of lane line curve fitting;

[0022] Figure 6 It is a schematic structural diagram of an implementation manner of the lane line map construction and update system of the present application. Specific implementation manner

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0024] Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims, and above-mentioned accompanying drawings of the present application are used to distinguish similar objects and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a product or device comprising a series of steps or units does not necessarily have to be limited to those units clearly listed, but may include other units not clearly listed or inherent to these products or devices.

[0025] High-precision maps are actually dedicated maps for autonomous driving systems as opposed to ordinary navigation electronic maps. Traditional high-precision map construction methods require expensive dedicated collection for data collection and a large amount of manual annotation of the collected data. Therefore, the traditional map construction method has high costs and slow updates. To solve the problems of high-precision map construction and update costs and efficiency, a crowdsourcing map construction method has emerged. Crowdsourcing map construction refers to deploying multiple vehicles with environmental perception capabilities, allowing them to collect road information data while driving and upload it to the cloud. The cloud constructs a highly restored and instantaneously updated driving map based on the feedback data. The mainstream crowdsourcing map construction method is to use a large number of passenger cars, such as private cars, taxis, etc., and configure low-cost sensors such as dash cams and mobile phones. By collecting a large amount of low-precision data, a fused high-precision-like map is obtained through fusion in the cloud. For the construction of lane line maps, generally in places where there are high-precision maps, the collected fragmented data is matched with the original high-precision map to solve the problem of lane line map update in high-precision maps. In places where there are no high-precision maps, by accumulating enough collected data, lane lines are extracted according to the Gaussian model.

[0026] For a vehicle with an autonomous driving function, the vehicle itself is already equipped with high-precision sensors such as RTK and lidar. Generally, the accuracy of the reconstructed lane lines is relatively high, within the meter level. Therefore, in the crowdsourcing mapping method, vehicles with autonomous driving functions are used for data collection, and the data quality is high, but the number of collections is small. Therefore, the original lane line fusion method is not applicable to such data. The existing technology generally requires a high-precision map as the base map, and the crowdsourcing collection method generally solves the problem of high-precision map update. If there is no high-precision map, generally enough data needs to be collected to perform cloud fusion. At present, there are not many autonomous driving vehicles, and the number of collections will not be large. However, due to the high-precision positioning of the vehicle, the accuracy of a single collection is relatively high.

[0027] In view of the above problems, the present application proposes a lane line map construction and update method, system, medium and device. The lane line map construction and update method includes: the cloud server performs data fusion on the road lane line data collected by a vehicle with high-precision positioning to obtain the current lane line; intercepts the current lane line according to a first preset length to obtain a plurality of current lane line segments; in the lane line database, extracts the corresponding lane line segments at the corresponding positions of the current lane line segments; extracts a plurality of lane line points on the current lane line segments and the corresponding lane line segments respectively; performs quadratic curve fitting on the plurality of lane line points, and intercepts the fitted lane line curve according to the first preset length to obtain updated lane line segments; uses the updated lane line segments to replace the corresponding lane line segments to update the lane line database. The solution of the present application can be applied to scenarios such as ports, highways, logistics, mines, closed parks, or urban traffic.

[0028] The lane line map construction and update method of the present application takes the road lane line data collected by a vehicle with high-precision positioning as the processing object, directly processes it, establishes a lane line database, and updates it. The processing process is more convenient and has relatively high accuracy.

[0029] The following uses specific embodiments to elaborate in detail on the technical solution of the present application and how the technical solution of the present application solves the above technical problems. These several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following will describe the embodiments of the present application with reference to the accompanying drawings.

[0030] Figure 1 Shows an implementation manner of the lane line map construction and update method of the present application.

[0031] In Figure 1In the illustrated embodiment, the lane line map construction and update method of the present application includes process S101, where the cloud server intercepts the current lane line obtained by the vehicle terminal according to a first preset length to obtain a plurality of current lane line segments.

[0032] In this embodiment, after obtaining the current lane line, the current lane line is intercepted according to the first preset length to obtain a plurality of current lane line segments.

[0033] Specifically, after the high-precision positioning autonomous vehicle collects the road lane line point cloud data and processes it through the cloud server, a 60-meter-long lane line is obtained. For example, if the first preset length is 10 meters, the obtained current lane line can be intercepted into 6 segments to obtain 6 current lane line segments. If a 58-meter-long lane line is obtained and the first preset length is 10 meters, 6 current lane line segments are also obtained, and the length of the tenth current lane line segment is 8 meters. If during the segmentation process, the length of a certain segment is very short, for example, less than 1 meter, then this segment is deleted.

[0034] Specifically, when obtaining the current lane line at the vehicle terminal, a vision solution can be selected. During vehicle driving, the front view image is recognized, and the recognized lane line is converted into a three-dimensional lane line in the world coordinate system and represented by connecting scattered points to obtain the final current lane line and upload it to the cloud server for processing. Or a lidar solution can be adopted to extract the lane line from the radar point cloud data and convert it to the world coordinate system to obtain the final current lane line. Among the above processes, all need to be completed at the vehicle terminal, and only the created three-dimensional lane line data is uploaded to the cloud server, which can reduce the amount of uploaded data and improve the data transmission speed.

[0035] In Figure 1 In the illustrated embodiment, the lane line map construction and update method of the present application includes process S102, where in the lane line database, the corresponding lane line segments at the positions corresponding to the current lane line segments are extracted.

[0036] In this embodiment, after determining each current lane line segment of the current lane line, it is necessary to update and correct the lane line database according to the current lane line segments. During the specific update process, the corresponding lane line segments at the positions corresponding to the current lane line segments are extracted to perform the update of the corresponding road sections.

[0037] Specifically, for example, when a highly precise positioning-equipped autonomous vehicle drives through section A, it collects the point cloud of the road lane lines and uploads the point cloud data to the cloud server. The current lane lines of section A are calculated and intercepted to obtain multiple current lane line segments of section A. When updating the lane line map, the corresponding lane line segments of section A are extracted from the lane line database for corresponding updates.

[0038] Optionally, in the lane line database, extracting the corresponding lane line segments at the corresponding positions of the current lane line segments includes: calculating the distances between the current lane line segments and each original lane line segment in the lane line database respectively; if the distance is less than the first preset threshold, determining the corresponding original lane line segment as the corresponding lane line segment.

[0039] In this optional embodiment, when extracting the corresponding lane line segment of a certain current lane line segment, it is judged by the distance relationship between the two. When the distance between the two line segments is less than the first preset threshold, the corresponding original lane line segment in the lane line database is determined as the corresponding vehicle guiding line segment of the current lane line segment. The first preset threshold can be set to 1 meter, and the first preset threshold can be reasonably selected according to the actual situation. The above is only a preferred example.

[0040] Optionally, calculating the distances between the current lane line segments and each original lane line segment in the lane line database respectively includes: extracting multiple judgment points on the current lane line segment according to the second preset length; making perpendicular lines to the current lane line segment at each judgment point; if there is a perpendicular line intersecting with an original lane line segment, taking the average length of each intersecting line segment as the distance between the current lane line and the corresponding original lane line segment; if there is no perpendicular line intersecting with the original lane line segment, the distance between the current lane line segment and the original lane line segment is infinite.

[0041] In this optional embodiment, when calculating the distance between the current lane line segment and the original lane line segment, first, multiple judgment points are determined on the current lane line segment according to the second preset length, and then perpendicular lines to the current lane are made at each judgment point. If there is a perpendicular line intersecting with the original lane line segment in the lane line database, calculate the average length of each intersecting line segment as the distance between the current lane line segment and the original lane line segment. If there is no perpendicular line intersecting with the original lane line segment, it is considered that the distance between the current lane line segment and the corresponding original lane line segment is infinite. The second preset threshold can be set to 1 meter, and the first preset threshold can be reasonably selected according to the actual situation. The above is only a preferred example.

[0042] Specifically, Figure 2 Shows a calculation example of the distance between the current lane line segment and the original lane line segment.

[0043] As Figure 2 shown, line segment A represents the current lane line segment, and line segment B represents a selected segment of an original lane line in the lane line database. Among them, a plurality of judgment points are determined on line segment A at a second preset distance. For example, the length of line segment A is 6.5 meters. The second preset length is 1 meter, and six judgment points a1, a2, a3, a4, a5, and a6 can be obtained respectively. Then perpendicular lines to line segment A are made at the six judgment points respectively. Among them, the perpendicular lines corresponding to judgment points a1, a2, and a3 intersect with line segment B, intersecting at points b1, b2, and b3 respectively. Then the distance between line segment A and line segment B can be expressed as: Then it is judged whether the distance D is less than the first preset threshold. If so, the original lane line segment B is the corresponding lane line segment of the current lane line segment A.

[0044] Specifically, Figure 3 shows a schematic diagram of the matching result between a current lane line segment and an original lane line segment.

[0045] As Figure 3 shown, line segment A represents the current lane line segment, line segments B1 and B2 represent the corresponding lane line segments matched with line segment A, and other line segments are other original lane line segments in the lane line database.

[0046] In Figure 1 the embodiment shown, the lane line map construction and update method of the present application includes process S103, and a plurality of lane line points are respectively extracted on the current lane line segment and the corresponding lane line segment.

[0047] In this embodiment, after determining the current lane line segment and the corresponding lane line segment, the subsequent lane line update process is carried out. First, a plurality of lane line points are respectively extracted on the current lane line segment and the corresponding lane line segment for subsequent line fitting.

[0048] Optionally, respectively extracting a plurality of lane line points on the current lane line segment and the corresponding lane line segment includes: extracting a plurality of lane line points on the corresponding lane line segment at a third preset length; on the current lane line segment, extracting a plurality of lane line points according to the update weight, where the lower the update weight, the greater the spacing of the lane line points extracted on the current lane line segment.

[0049] In this optional embodiment, when extracting lane line points on the corresponding lane line segment, a plurality of lane line points are evenly extracted on the corresponding lane line at a third preset length. When extracting lane line points on the current lane line segment, a plurality of lane line points are extracted on the current lane line segment according to the weight of this lane line update. Among them, the lower the update weight, the greater the spacing of the lane line points extracted on the current lane line segment, and the fewer the number of the extracted lane line points.

[0050] Specifically, the current lane line segment is obtained by collecting road lane line point clouds through radar during the driving of the autonomous vehicle and processing them on the cloud server. Since there are certain errors in the process of lane line data collection, when updating the lane lines in the lane line database, corresponding update weights will be set.

[0051] Optionally, according to the update weight, multiple lane line points are extracted, including: taking the ratio of the third preset length to the update weight as the spacing for extracting lane line points on the current lane line segment; extracting multiple lane line points according to the spacing.

[0052] In this optional embodiment, when extracting lane line points on the current lane line segment, the third preset length on the corresponding lane line segment is divided by the update weight to obtain the spacing for extracting lane line points on the current lane line segment. Then, lane line extraction is performed according to this spacing.

[0053] Specifically, assume the update weight is w, where the update weight is a value between 0 and 1 and can be reasonably set according to actual update requirements. For example, the update weight w takes a value of 0.5. The third preset length on the corresponding lane line segment is represented as t, for example, 2 meters. Then the spacing for extracting lane line points on the current lane line segment is That is, 2 / 0.5 = 4 meters.

[0054] Specifically, Figure 4 Fig. shows an example of lane line point extraction.

[0055] As shown in Fig. 4, on the current lane line segment A, multiple lane line points are extracted on the line segments B1 and the corresponding lane line segments B1 and B2, where the spacing between the lane line points on the current lane line segment A is greater than the spacing between the lane line points on the corresponding lane line segments B1 and B2.

[0056] In Figure 1 In the shown embodiment, the lane line map construction and update method of the present application includes process S104, performing quadratic curve fitting on multiple lane line points, and intercepting the fitted lane line curve according to the first preset length to obtain an updated lane line segment.

[0057] In this embodiment, after obtaining multiple lane line points of the corresponding lane line segment and the current lane line segment, quadratic curve fitting is performed on these multiple lane line points to obtain the fitted lane line curve, and then the fitted lane line curve is intercepted according to the first preset length to obtain an updated lane line segment.

[0058] Specifically, Figure 5 Fig. shows an example of lane line curve fitting.

[0059] As Figure 5 shown, after obtaining multiple lane line points, through the fitting of the lane line curve, the final lane line curve is obtained, and then it is intercepted to obtain the updated lane line segment.

[0060] In Figure 1 the embodiment shown, the lane line map construction and update method of the present application includes process S105, using the updated lane line segment to replace the corresponding lane line segment to update the lane line database.

[0061] In this embodiment, after obtaining the updated lane line segment, in the lane line database, the corresponding lane line segment is replaced to complete the update of the lane line database.

[0062] Optionally, if there is no corresponding lane line segment at the corresponding position of the current lane line segment in the lane line database, the current lane line segment is directly added to the lane line database.

[0063] In this optional embodiment, there may be no corresponding lane line segment for the newly obtained current lane line segment. For example, this section of the road is collected for the first time. Therefore, at this time, the previous lane line segment directly obtained is added to the lane line database to establish the lane line database, that is, the establishment of the lane line map.

[0064] Optionally, in the lane line database, multiple lane line segments with an end point distance less than the second preset threshold are spliced to obtain a complete lane line.

[0065] In this optional embodiment, the lane line is stored in the lane line database in the form of lane line segments. If a complete lane line is to be obtained, only multiple lane line segments with an end point distance less than the second preset threshold need to be spliced to obtain a complete lane line.

[0066] Specifically, when updating the lane line database, the number of fusions of each lane line segment in the lane line database is counted. The number of fusions indicates the number of times the lane line segment is fused when the lane line database is updated. For example, for a road, after ten collections, the number of fusions of a certain lane line segment is 10, indicating that the lane line segment has been seen ten times and has a high confidence level, so it can be retained. If there is a lane line segment with a fusion number of 1, it means that this lane line segment has only been fused once, and this area has not been seen again in all subsequent collections, then this lane line segment can be considered a false detection and will be deleted later. By recording the number of fusions of the lane line segment, the higher the number of fusions of the lane line segment, the more real the segment is. The lane line segments with lower fusion numbers are deleted to improve the accuracy of the lane line database. The second preset threshold can be set to 0.5 meters. It should be noted that specifically, what numerical conditions need to be met to delete the lane line segment can be reasonably set according to the actual accuracy requirements, and this application does not make specific restrictions.

[0067] The lane line map construction and update method of this application processes the road lane line data collected by a vehicle with high-precision positioning to establish and update the lane line database. It does not require a high-precision map as a base map, nor does it require a large amount of perception data. The processing process is more convenient and has a higher accuracy at the same time.

[0068] Figure 6 An embodiment of the lane line map construction and update system of this application is shown.

[0069] In Figure 6 In the shown embodiment, the lane line map construction and update system of this application includes: a module 602 for intercepting the current lane line obtained from the vehicle end in the cloud server according to a first preset length to obtain a plurality of current lane line segments; a module 603 for extracting the corresponding lane line segments at the corresponding positions of the current lane line segments in the lane line database; a module 604 for respectively extracting a plurality of lane line points on the current lane line segments and the corresponding lane line segments; a module 605 for performing quadratic curve fitting on the plurality of lane line points and intercepting the fitted lane line curve according to the first preset length to obtain updated lane line segments; a module 606 for using the updated lane line segments to replace the corresponding lane line segments to update the lane line database.

[0070] Optionally, in module 603, the distances between the current lane line segments and each original lane line segment in the lane line database are calculated respectively; if the distance is less than the first preset threshold, the corresponding original lane line segment is determined as the corresponding lane line segment.

[0071] Optionally, in module 603, on the current lane line segment, multiple judgment points are extracted according to a second preset length; perpendicular lines to the current lane line segment are drawn at each judgment point; if there is a perpendicular line intersecting with the original lane line segment, the average length of each intersecting line segment is used as the distance between the current lane line and the corresponding original lane line segment; if there is no perpendicular line intersecting with the original lane line segment, the distance between the current lane line segment and the original lane line segment is infinite.

[0072] Optionally, in module 604, multiple lane line points are extracted on the corresponding lane line segment according to a third preset length; on the current lane line segment, multiple lane line points are extracted according to the update weight, where the lower the update weight, the greater the spacing of the lane line points extracted on the current lane line segment.

[0073] Optionally, in module 604, the ratio of the third preset length to the update weight is used as the spacing of the lane line points extracted on the current lane line segment; multiple lane line points are extracted according to the spacing.

[0074] Optionally, in module 603, if there is no corresponding lane line segment at the corresponding position of the current lane line segment in the lane line database, the current lane line segment is directly added to the lane line database.

[0075] Optionally, it further includes a lane line splicing module, which splices multiple lane line segments with an end point distance less than a second preset threshold in the lane line database to obtain a complete lane line.

[0076] The lane line map construction and update system of the present application processes the road lane line data collected by a high-precision positioning vehicle to establish and update the lane line database, does not require a high-precision map as a base map, nor a large amount of perception data, the processing process is more convenient, and at the same time has a high precision.

[0077] In a specific embodiment of the present application, a computer-readable storage medium stores computer instructions, where the computer instructions are operated to execute the lane line map construction and update method described in any embodiment. Among them, the storage medium can be directly in hardware, in a software module executed by a processor, or in a combination of both.

[0078] The software module can reside in a RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, register, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. The exemplary storage medium is coupled to the processor such that the processor can read information from and write information to the storage medium.

[0079] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The general-purpose processor may be a microprocessor, but in an alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in combination with a DSP core, or any other such configuration. In an alternative, the storage medium may be integral with the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and the storage medium may reside in the user terminal as discrete components.

[0080] In a specific embodiment of the present application, a computer device includes a processor and a memory, and the memory stores computer instructions, wherein: the processor operates the computer instructions to execute the lane line map construction and update method described in any of the embodiments.

[0081] In the embodiments provided in the present application, the cloud server may be replaced by a local server, and the embodiments of the present application do not make any limitations thereto.

[0082] In the embodiments provided in the present application, it should be understood that the disclosed device may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other may be through some interfaces, and the indirect couplings or communication connections of the devices or units may be in electrical, mechanical, or other forms.

[0083] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] The above are only embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for constructing and updating a lane line map, characterized in that, it includes: The cloud server intercepts the current lane line obtained by the vehicle terminal with high-precision positioning according to a first preset length to obtain a plurality of current lane line segments; In the lane line database, the corresponding lane line segments at the corresponding positions of the current lane line segments are extracted; A plurality of lane line points are respectively extracted on the current lane line segment and the corresponding lane line segment, including, On the corresponding lane line segment, a plurality of the lane line points are extracted according to a third preset length, On the current lane line segment, according to the update weight, a plurality of the lane line points are extracted, wherein the lower the update weight, the larger the spacing of the lane line points extracted on the current lane line segment, wherein the spacing between the lane line points on the current lane line segment is greater than the spacing between the lane line points on the corresponding lane line segment; the plurality of lane line points are subjected to quadratic curve fitting, and the fitted lane line curve is intercepted according to the first preset length to obtain updated lane line segments; The corresponding lane line segments are replaced with the updated lane line segments to update the lane line database.

2. The lane line map construction and update method according to claim 1, characterized in that, The step of extracting the original lane line segments at the corresponding positions of the current lane line segments in the lane line database includes: Calculating the distances between the current lane line segment and each original lane line segment in the lane line database respectively; If the distance is less than a first preset threshold, the corresponding original lane line segment is determined as the corresponding lane line segment.

3. The lane line map construction and update method according to claim 2, characterized in that, The step of calculating the distances between the current lane line segment and each original lane line segment in the lane line database respectively includes: On the current lane line segment, a plurality of judgment points are extracted according to a second preset length; Perpendicular lines to the current lane line segment are made at each of the judgment points; If there is an intersection between the perpendicular line and the original lane line segment, the average length of each intersecting segment is used as the distance between the current lane line and the corresponding original lane line segment; If there is no intersection between the perpendicular line and the original lane line segment, the distance between the current lane line segment and the original lane line segment is infinite.

4. The lane line map construction and update method according to claim 1, characterized in that, The step of extracting a plurality of the lane line points according to the update weight includes: Taking the ratio of the third preset length to the update weight as the spacing of the lane line points extracted on the current lane line segment; Extracting a plurality of the lane line points according to the spacing.

5. The lane line map construction and update method according to claim 1, characterized in that, it further includes: If there is no corresponding lane line segment at the corresponding position of the current lane line segment in the lane line database, the current lane line segment is directly added to the lane line database.

6. The lane line map construction and update method according to claim 1, characterized in that, further comprising: In the lane line database, splicing multiple lane line segments with an end point distance less than a second preset threshold to obtain a complete lane line.

7. A lane line map construction and update system, characterized in that, comprising: A module for intercepting the current lane line obtained by the vehicle end with high-precision positioning in the cloud server according to a first preset length to obtain a plurality of current lane line segments; A module for extracting corresponding lane line segments at corresponding positions of the current lane line segments in the lane line database; A module for respectively extracting a plurality of lane line points on the current lane line segment and the corresponding lane line segment, including, extracting a plurality of the lane line points on the corresponding lane line segment according to a third preset length, on the current lane line segment, extracting a plurality of the lane line points according to an update weight, wherein the lower the update weight, the greater the distance between the lane line points extracted on the current lane line segment, wherein the distance between the lane line points on the current lane line segment is greater than the distance between the lane line points on the corresponding lane line segment; a module for performing quadratic curve fitting on the plurality of lane line points and intercepting the fitted lane line curve according to the first preset length to obtain updated lane line segments; A module for using the updated lane line segments to replace the corresponding lane line segments to update the lane line database.

8. A computer-readable storage medium storing computer instructions, wherein the computer instructions are operated to execute the lane line map construction and update method according to any one of claims 1-6.

9. A computer device comprising a processor and a memory, the memory storing computer instructions, wherein: The processor operates the computer instructions to execute the lane line map construction and update method according to any one of claims 1-6.

Images