A high-precision map data generation method, device and server

Abstract

The present disclosure relates to a high-precision map data generation method, device and server. The method comprises: obtaining current crowd-sourced collected map element data, and determining the position of an intersection in the map element data; segmenting the map element data according to the intersection position; aggregating the map element data belonging to adjacent two intersections in the segmented map element data respectively to obtain the map element data of each road; and taking the road as a basic unit, processing the map element data of each road respectively to obtain high-precision map data. The road is taken as a basic unit to process the map element data of the road for high-precision map, so as to obtain the entire high-precision map data, reduce the influence of vehicle inertial navigation on data accuracy as much as possible, and ensure the accuracy of high-precision map data.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of geographic information, in particular to a high-precision map data generation method, device and server. BACKGROUND

[0002] The high-precision map refers to a map with high precision and fine definition, which is different from the electronic map widely known at present. The precision thereof needs to reach the decimeter level and can accurately distinguish various lanes.

[0003] Unlike the traditional electronic map, the main service target of the high-precision map is automatic driving and advanced intelligent auxiliary driving, such as unmanned vehicles. Unlike human drivers, machine drivers lack innate visual recognition and logical analysis capabilities. For example, people can easily and accurately locate themselves using images and GPS, and identify obstacles, people and traffic lights, but this is a very difficult task for current robots.

[0004] Therefore, the high-precision map is an essential part of current automatic driving and unmanned vehicle technology. The high-precision map contains a large amount of driving assistance information, of which the most important is the accurate three-dimensional representation (centimeter-level precision) of the road network. In addition, the high-precision map also needs to have higher real-time performance than the traditional map to ensure the "new" and "fresh" of the map elements. Since the road network changes every day, such as repair, road marking line wear and repainting, and traffic sign changes. These changes need to be reflected on the high-precision map in a timely manner to ensure the safety of unmanned vehicle driving.

[0005] How to realize the production and update of large-scale high-precision maps has become a problem in the map industry. The current generation method of high-precision map data is usually through crowdsourcing collection, that is, relying on vehicles equipped with various sensors (inertial navigation, cameras, laser radars, millimeter radars, etc.) to perform automatic driving tasks. During the crowdsourcing collection process, the inertial navigation of the vehicle provides the absolute position of the vehicle, and the camera or laser radar can measure the relative position relationship between the map elements around the vehicle and the vehicle. The combination of the two can produce the map element data required for high-precision maps.

[0006] Since the error of the vehicle inertial navigation can be considered constant within a relatively short period of time or distance, the error will change with the extension of time and distance. If all the data collected by crowdsourcing is used as a whole to produce and update high-precision maps, the change of the inertial navigation error will inevitably affect the accuracy of the high-precision map data. SUMMARY

[0007] In view of the above problems, the present disclosure is proposed to provide a high-precision map data generation method, device and server which can overcome the above problems or at least partially solve the above problems.

[0008] In a first aspect, the embodiments of the present disclosure provide a method for generating high-precision map data, comprising:

[0009] obtaining map element data collected by crowdsourcing, and determining positions of intersections in the map element data;

[0010] segmenting the map element data according to the positions of the intersections;

[0011] aggregating map element data belonging to between two adjacent intersections in the segmented map element data, to obtain map element data of each road;

[0012] processing the map element data of each road as a basic unit to obtain high-precision map data.

[0013] In an embodiment, the positions of the intersections in the road map data are determined, specifically comprising:

[0014] identifying lane line data in the map element data;

[0015] determining the positions of the intersections in the map element data according to distances between lane lines in the lane line data along the extension direction of the road.

[0016] In an embodiment, the lane line data in the map element data is identified, comprising:

[0017] identifying center points and positions of lane lines in each road map photo collected by crowdsourcing;

[0018] smoothly connecting each center point with a distance less than a preset first distance threshold to obtain corresponding lane line data.

[0019] In an embodiment, the intersections in the road map data are determined according to distances between lane lines in the lane line data, comprising:

[0020] judging whether the distance between two adjacent lane lines disconnected along the extension direction of the road in the lane line data is greater than a preset second distance threshold, and if so, determining that the lane lines are intersections.

[0021] In an embodiment, the map element belonging to between two adjacent intersections in the segmented road map data is aggregated to obtain map element data of each road, comprising:

[0022] determining the start point and the end point of the lane line in the lane line data belonging to between two adjacent intersections along the extension direction of the road;

[0023] According to the collection time of the lane line head point and the tail point, the map element data belonging to the second aspect, the embodiment of the disclosure provides a high-precision map data generation device, comprising:

[0024] An acquisition module is configured to acquire road map data obtained through crowdsourcing;

[0025] An intersection determination module is configured to determine the position of an intersection in the road map data;

[0026] A segmentation module is configured to segment the road map data according to the position of the intersection;

[0027] An aggregation module is configured to aggregate map elements belonging to adjacent two intersections in the segmented road map data, to obtain map element data of each road;

[0028] A generation module is configured to process the map element data of each road as a basic unit, to obtain high-precision map data.

[0029] In one embodiment, the intersection determination module is configured to identify lane line data in the map element data; and determine the position of the intersection in the map element data according to the distance between each lane line in the lane line data along the extension direction of the road.

[0030] In a third aspect, the embodiment of the disclosure provides a high-precision map, and data of the high-precision map is obtained through the foregoing high-precision map generation method.

[0031] In a fourth aspect, the embodiment of the disclosure provides a high-precision map server, comprising a memory, a processor, and computer instructions stored in the memory and executable on the processor, and the instructions can implement the foregoing high-precision map data generation method when executed by the processor.

[0032] In a fifth aspect, the embodiment of the disclosure provides a computer readable storage medium, and the medium stores computer instructions, and the instructions can implement the foregoing high-precision map data generation method when executed by a processor.

[0033] The technical solution provided by the embodiment of the disclosure has at least the following beneficial effects:

[0034] The high-precision map data generation method, device and server provided by the embodiments of the present disclosure can determine the positions of the intersections in the map element data collected by crowdsourcing, and segment the map element data according to the intersections, aggregate the segmented map element data between two adjacent intersections, and obtain the map element data of each road, so as to realize the processing required for high-precision map of the map element data of the road as a basic unit, and obtain the entire high-precision map data. The present disclosure divides the collected map element data into smaller ranges, processes the data of each smaller range, and thus guarantees the accuracy of the entire high-precision map data, reduces the influence of vehicle inertial navigation on the accuracy of the data as much as possible, and processes the data of the road as a basic unit, so that the amount of data of a single road is small and the data can be processed in parallel to improve the efficiency of production and updating of the high-precision map data.

[0035] Other features and advantages of the present disclosure will be set forth in the following description, and in part will become apparent from the description, or can be learned by practice of the present disclosure. The objects and other advantages of the present disclosure can be achieved and obtained by the structure particularly pointed out in the written description, claims, and drawings.

[0036] The technical solutions of the present disclosure will be described in detail below with the help of the drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0037] The accompanying drawings are included to provide a further understanding of the present disclosure, and constitute a part of the specification, and are used to explain the present disclosure together with the embodiments of the present disclosure, and do not constitute a limitation on the present disclosure. In the drawings:

[0038] Figure 1 The flowchart of the high-precision map data generation method provided by the embodiments of the present disclosure is shown in the figure.

[0039] Figure 2 The schematic diagram of the map element contained in the high-precision map provided by the embodiments of the present disclosure is shown in the figure.

[0040] Figure 3 The flowchart of determining the position of the intersection in the road map data in the embodiments of the present disclosure is shown in the figure.

[0041] Figure 4 The schematic diagram of the map element data of a certain area collected by the crowdsourcing vehicle in the embodiments of the present disclosure is shown in the figure.

[0042] Figure 5 The structure block diagram of the high-precision map data generation device provided by the embodiments of the present disclosure is shown in the figure. DETAILED DESCRIPTION

[0043] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0044] To solve the problem that high-precision map data generated finally is inaccurate due to the uncertainty of vehicle inertial navigation error in the production process, the inventors of the present disclosure found that the error of vehicle inertial navigation is usually unchanged within a relatively short distance, and therefore it is necessary to divide the large-scale urban map data collected by crowdsourcing vehicles into smaller ranges, process the data in each smaller range, and further ensure the accuracy of the high-precision map data as a whole.

[0045] The high-precision map data generation method provided by the embodiments of the present disclosure adopts a processing mode taking the road formed between intersections as a basic unit. Since the length of such road is usually between 500-1000 meters for urban roads, the condition that the vehicle inertial navigation error remains unchanged can be met. Based on this, the high-precision map data generation method provided by the embodiments of the present disclosure, as shown in Figure 1 includes the following steps:

[0046] S11, acquiring current map element data collected by crowdsourcing, and determining the position of the intersection in the map element data;

[0047] S12, dividing the map element data according to the intersection position;

[0048] S13, respectively aggregating the map element data belonging to the map element data between adjacent two intersections in the divided map element data, to obtain the map element data of each road;

[0049] S14, taking the road as a basic unit, respectively processing the map element data of each road to obtain high-precision map data.

[0050] Before describing the above-mentioned high-precision map data generation method provided by the embodiments of the present disclosure, the map elements in the high-precision map are first described briefly.

[0051] Referring to Figure 2 , the map elements usually contained in the high-precision map include, for example, the following:

[0052] Lane line: ground lane dividing line, which can be represented by a series of continuous points.

[0053] Ground arrow: the ground is used to indicate the vehicle lane marking. Four corner points of the circumscribed rectangle are used to represent, and the type value is used to distinguish different types such as straight, left turn, right turn, etc. Figure 2 The number in the rectangular frame is the type value.

[0054] Pole: the pole on both sides of the road. It can be any artificial pole on both sides of the road, which is represented by a point projected on the ground.

[0055] Sign: road sign. Four corner points of the circumscribed rectangle are used to represent Figure 2 The sign adopts the way of top view, and a line of the sign projected on the ground is used to represent.

[0056] Of course, in the embodiments of the present disclosure, the map elements include but are not limited to the above four kinds, and more map elements such as road edges and zebra crossings can be extended.

[0057] In the above map elements, because the discontinuity of the lane line is closely related to the discontinuity of the road, the lane line is usually disconnected at the road intersection. In order to realize the generation and update of the high-precision map data according to the road as the basic unit, in the above step S11, the position of the intersection in the road map data is determined by referring to Figure 3 As shown in the following formula, for example, the position of the intersection in the road map data can be determined in the following way:

[0058] S31, identifying the lane line data in the map element data;

[0059] In the collection process, the crowd-sourcing vehicle photographs the road passed, obtains a series of photos of the road, and the photos contain lane lines. The center points of the lane lines are identified, and the position data of the center points is determined. After the center points of the lane lines in each photo are identified, a series of continuous points can be obtained. By respectively smoothing the connection between the center points with a distance less than a preset first distance threshold, the corresponding lane line data can be obtained.

[0060] The reason for selecting the center points with a distance less than the preset first distance threshold to connect is that although the interval of the center points collected in some photos is not necessarily equal due to the factors of the shooting environment, excluding the factors of the vehicle speed, there may be no or no lane line data in some photos during the shooting process. Therefore, the distance between the identified center points of the lane lines is not necessarily fixed, but the interval between them will not be too large. Therefore, according to the preset first distance threshold, the continuous points can be connected.

[0061] Here, other set points of the lane line can also be used to obtain the lane line data, such as the first point and the tail point, which are not limited in the embodiments of the present disclosure.

[0062] S32, determining intersection data in the map element data according to distances between each lane line in the lane line data.

[0063] Since the lane lines are disconnected at the intersection, in the step S32, it can be determined whether the intersection is encountered by judging whether the distance between two adjacent lane lines disconnected in the lane line data along the road extension direction is greater than a preset second distance threshold, and if so, it is determined that the two lane lines are at the intersection.

[0064] The second distance threshold is usually much larger than the first distance threshold, and the value range of the second distance threshold is usually above tens of meters.

[0065] Referring to Figure 4 As shown in FIG. 1, the distance between the lane line 1 and the lane line 1' disconnected in the road extension direction is greater than the preset second distance threshold, so it is determined that the intersection is at this position.

[0066] In an embodiment, the map elements between adjacent two intersections are aggregated in the step S13 to obtain the map element data of each road, which can be realized by the following steps:

[0067] First, the starting point and the ending point of the lane line belonging to the lane line data between the adjacent two intersections are determined, and then the map element data belonging to the map elements collected within the collection time is aggregated with the data of the lane line according to the collection time of the starting point and the ending point of the lane line.

[0068] Of course, the map elements whose position relationship is between the starting point and the ending point of the lane line can also be aggregated.

[0069] However, compared with the way of selecting which map element data to aggregate according to the position relationship, the way of selecting which map element data to aggregate according to the collection time of the starting point and the ending point of the lane line is better, because the crowd-sourcing vehicles collect data along the set route, so all the map element data between the collection time of the starting point and the ending point of the lane line must belong to the road according to the time. However, according to the geographical position, it is difficult to accurately determine which map elements in the photo belong to the road due to the influence of the clarity of the photo, the environment of the road, the edge distortion of the lens and other factors in the process of crowd-sourcing collection and shooting, so the accuracy is poor.

[0070] In an embodiment, the step S14 is implemented by taking the road as the basic unit, respectively matching the map element data of each road with the corresponding map element of the road in the existing high-precision map data to improve the accuracy of the position of the currently collected road map element data.

[0071] Then the current collected map element data of each road is fused with the map element data of each road in the existing high-precision map data, to obtain fused high-precision map data, and then the relevant map element data is extracted from the fused high-precision map data to obtain the final high-precision map data.

[0072] In order to illustrate the above-mentioned high-precision map data generation method provided by the embodiments of the present disclosure, a simple example is used for illustration as follows:

[0073] Still referring to Figure 4 , the left half of the figure is a schematic diagram of the map element data of a certain area collected by the crowd-sourcing vehicle. The long lines in the figure are the recognized lane lines, and the dots represent the rod-shaped objects (such as street lamp poles, power poles, sign poles, etc.) on both sides of the road. Whether a road passes through an intersection can be determined according to the intermittent relationship of the lane lines in the extension direction of the road. The collected map element data is segmented according to the intersections, that is, the map element data of the road formed between two adjacent intersections is aggregated. For example Figure 4 , the right half of the figure is an enlarged view of the road circled by the left circle. As can be seen from the enlarged view, the circle contains two roads numbered 7 and 8.

[0074] After the entire area is divided into roads similar to the roads numbered 7 and 8, the map element data of the roads can be further matched, fused, updated, etc. based on the road as the basic unit, and the entire high-precision map data is finally obtained.

[0075] Based on the same inventive concept, the embodiments of the present disclosure also provide a high-precision map data generation device, a high-precision map and a high-precision map server. Since the principles of the problems solved by these devices and servers are similar to the above-mentioned high-precision map data generation method, the implementation of the devices and servers can be referred to the implementation of the above-mentioned method, and the repeated parts will not be described again.

[0076] The high-precision map data generation device provided by the embodiments of the present disclosure, referring to Figure 5 , includes:

[0077] The acquisition module 51 is configured to acquire the road map data collected by the crowd-sourcing.

[0078] The intersection determination module 52 is configured to determine the position of the intersection in the road map data.

[0079] The segmentation module 53 is configured to segment the road map data according to the position of the intersection.

[0080] The aggregation module 54 is configured to aggregate the map elements belonging to the map elements between two adjacent intersections in the segmented road map data, respectively, to obtain the map element data of each road.

[0081] The generation module 55 is configured to process the map element data of each road to obtain the high-precision map data.

[0082] In an embodiment, the intersection determination module 52 is further configured to identify lane line data in the map element data, and determine the position of the intersection in the map element data according to the distance between the lane lines in the lane line data along the extension direction of the road.

[0083] In an embodiment, the intersection determination module 52 is further configured to identify the center points of the lane lines and the positions of the center points in each road map photo collected by crowdsourcing, and connect the center points with a distance less than a preset first distance threshold to obtain the lane line data.

[0084] In an embodiment, the intersection determination module 52 is further configured to determine whether the distance between two adjacent lane lines disconnected along the extension direction of the road in the lane line data is greater than a preset second distance threshold, and determine that the lane lines are an intersection if the distance is greater than the preset second distance threshold.

[0085] In an embodiment, the aggregation module 54 is configured to determine the start point and the end point of the lane line in the lane line data belonging to the map elements between two adjacent intersections along the extension direction of the road, and aggregate the map element data collected within a collection time and the lane line data according to the collection time of the start point and the end point of the lane line.

[0086] The disclosure further provides a high-precision map, and data of the high-precision map is obtained by the method for generating a high-precision map.

[0087] The disclosure further provides a high-precision map server, which comprises a memory, a processor, and computer instructions stored in the memory and executable on the processor, and the instructions are executable on the processor to implement the method for generating a high-precision map.

[0088] The disclosure further provides a computer readable storage medium, which stores computer instructions executable on a processor to implement the method for generating a high-precision map.

[0089] The method, device and server for generating high-precision map data provided by the embodiments of the present disclosure determine the positions of the intersections in the map element data collected by crowdsourcing, and segment the map element data according to the intersections, aggregate the map element data between adjacent two intersections after segmentation, and obtain the map element data of each road, so as to realize the processing required by the high-precision map of the road as a basic unit, and obtain the entire high-precision map data. The embodiments of the present disclosure segment the collected map element data in a smaller range, process the data in each smaller range, and further ensure the accuracy of the entire high-precision map data, and reduce the influence of the vehicle inertial navigation on the data accuracy as much as possible. In addition, the data processing mode of taking the road as a basic unit has a smaller amount of data of a single road, and can be processed in parallel to improve the efficiency of the production and updating of the high-precision map data.

[0090] Those skilled in the art will understand that the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage, etc.) containing computer-usable program code.

[0091] The present disclosure is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0092] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0093] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable devices provide processes for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks Figure 1 one or more blocks or steps of functions specified in the flowchart

[0094] Obviously, those skilled in the art can make various modifications and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure belong to the scope of the claims of the present disclosure and their equivalent technologies, the present disclosure also intends to include these modifications and variations.

Claims

1. A method for generating high-precision map data, comprising: obtaining current crowd-sourced collected map element data, and determining positions of intersections in the map element data; segmenting the map element data according to the intersection positions to obtain roads formed between intersections as basic units, which satisfy the condition that vehicle inertial navigation error remains consistent; aggregating map element data belonging to adjacent two intersections in the segmented map element data to obtain map element data of each road; processing the map element data of each road as basic units to obtain high-precision map data. 2.The method of claim 1, wherein the positions of intersections in the road map data are determined in particular by: identifying lane line data in the map element data; determining the positions of intersections in the map element data according to distances between lane lines in the road extension direction in the lane line data. 3.The method of claim 2, wherein the lane line data in the map element data is identified by: identifying center points of lane lines and their positions in each road map photo collected by crowd-sourcing; smoothly connecting each center point with a distance less than a preset first distance threshold to obtain corresponding lane line data. 4.The method of claim 2, wherein the intersections in the road map data are determined according to distances between lane lines in the lane line data by: judging whether the distance between two adjacent lane lines disconnected in the road extension direction in the lane line data is greater than a preset second distance threshold, and if so, determining that the lane lines are intersections. 5.The method of any one of claims 2-4, wherein the map element data belonging to adjacent two intersections in the segmented road map data is aggregated to obtain map element data of each road by: determining the start point and end point of lane lines in lane line data belonging to adjacent two intersections in the road extension direction; aggregating map element data belonging to the lane line data in the road extension direction in the collection time in the map element data according to the collection time of the start point and end point of the lane lines. 6.An apparatus for generating high-precision map data, comprising: an obtaining module configured to obtain current crowd-sourced collected road map data; an intersection determining module configured to determine positions of intersections in the road map data; a segmentation module configured to segment the road map data according to the intersection positions to obtain roads formed between intersections as basic units, which satisfy the condition that vehicle inertial navigation error remains consistent; an aggregation module configured to aggregate map element data belonging to adjacent two intersections in the segmented road map data to obtain map element data of each road; a generation module configured to process the map element data of each road as basic units to obtain high-precision map data. 7.The apparatus of claim 6, wherein the intersection determining module is configured to identify lane line data in the map element data, and determine a position of an intersection in the map element data according to distances between lane lines in the lane line data along a road extension direction. 8.A high-definition map, data of which is generated by the method of any one of claims 1-5.

9. A high-definition map server comprising: A memory, a processor, and computer instructions stored on the memory and executable on the processor, the instructions being executable by the processor to implement the method of claim 1-5 for generating high-definition map data. 10.A computer readable storage medium having stored thereon computer instructions, the instructions being executable by a processor to implement the method of claim 1-5 for generating high-definition map data.

Images