Method and device for checking the accuracy of a map data conversion

Abstract

The application provides a precision checking method and device based on map data conversion, which comprises the following steps: obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map; performing sampling processing on the high-precision map and the simulation map respectively, and classifying each point obtained by the sampling processing to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; comparing the deviation with a preset threshold value, and determining a precision checking result according to a comparison result. The application can effectively calculate the deviation value of the map, and can simultaneously feed back a road number and a lane number to which a lane line not satisfying the deviation threshold value range belongs, so as to optimize the map in a targeted manner.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a precision checking method and device based on map data conversion. BACKGROUND

[0002] A map is a figure or image in two or more dimensions that represents, by certain rules and selectively, some phenomena on the plane or sphere of the earth (or other planets), which has strict mathematical basis, symbol system, and text notes, and can scientifically reflect the distribution characteristics and mutual relationship of natural and social economic phenomena by map generalization principles. A high-definition map (HD Map) is a special map relative to an ordinary navigation electronic map, and the data precision of the high-definition map can reach a centimeter level. Rich data dimensions contain information of surrounding static objects closely related to transportation in addition to road information. Due to these characteristics, the high-definition map plays a significant role in map positioning, auxiliary environment perception, and path planning in the field of autonomous driving.

[0003] In the virtual test link of autonomous driving, a high-precision simulation map is also the basis for scene testing. At present, the simulation map mostly adopts the OpenDRIVE format. The OpenDRIVE format is a high-definition map standard that can be used for simulation testing, which is developed by the German VIRES company and can realize the transmission of road data between different simulation engines. At present, the OpenDRIVE standard is maintained by the German Automation and Measurement System Standard Association (ASAM).

[0004] In the process of closed-loop simulation testing such as hardware-in-the-loop and vehicle-in-the-loop, the simulation engine internally uses the simulation map in the OpenDRIVE format, and the autonomous driving controller (ADU) internally uses the high-precision map for autonomous driving. Both maps record the same road logic and basic road network information.

[0005] However, in the actual simulation development process, the developer needs to develop a tool chain to reconstruct the high-precision map data into the simulation map in the OpenDRIVE format. Generally, in order to improve the efficiency of the simulation engine in loading and rendering the map and to improve the smoothness effect of the road connection, a cubic polynomial is often used to fit the road reference line in the process of map reconstruction. However, this reconstruction cannot guarantee that all the sampling points (Points) or Nodes coincide with the road reference line, and there may be local position deviations, which will directly affect the judgment of the lateral position of the target vehicle relative to the autonomous vehicle, and even affect the control logic of the ADU, leaving an uncontrollable influencing factor for the credibility of the simulation test. SUMMARY

[0006] The application provides a precision checking method and device based on map data conversion, which can effectively calculate the deviation value of the map, and can feed back the road number and lane number to which the lane line that does not meet the deviation threshold range belongs, so as to optimize the map in a targeted manner.

[0007] The application provides a precision checking method based on map data conversion, comprising: obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map; respectively sampling the high-precision map and the simulation map, and classifying each point obtained by sampling to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; obtaining the deviation corresponding to each lane line according to the first lane line point set and the second lane line point set and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; and comparing the deviation with a preset threshold and determining a precision checking result according to a comparison result.

[0008] According to the precision checking method based on map data conversion provided by the application, the high-precision map and the simulation map are respectively sampled and each point obtained by sampling is classified, comprising: sampling the high-precision map, and classifying each point in the high-precision map obtained by sampling based on a lane line connection relationship to obtain a first lane line point set corresponding to each lane line in the high-precision map; sampling each lane line in the simulation map based on a preset sampling step, and classifying each sampling point obtained by sampling to obtain a second lane line point set corresponding to each lane line.

[0009] According to the precision checking method based on map data conversion provided by the application, the high-precision map and the simulation map are respectively sampled and each point obtained by sampling is classified, comprising: sampling the high-precision map, and classifying each point in the high-precision map obtained by sampling based on a lane line connection relationship to obtain a first lane line point set corresponding to each lane line in the high-precision map; sampling each lane line in the simulation map based on a preset sampling step, and classifying each sampling point obtained by sampling to obtain a second lane line point set corresponding to each lane line.

[0010] According to the precision checking method based on map data conversion provided by the application, the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set, and the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set.

[0011] According to the precision checking method based on map data conversion provided by the application, the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set, and the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set.

[0012] According to the precision checking method based on map data conversion provided by the application, the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set, and the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set.

[0013] According to the precision checking method based on map data conversion provided by the application, the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set, and the deviation corresponding to each lane line is obtained according to the matched relationship between the first lane line point set and the second lane line point set.

[0014] The application further provides a precision checking device based on map data conversion, comprising: a data acquisition module, which acquires a high-precision map and a simulation map, wherein the simulation map is obtained by converting the high-precision map; a classification module, which respectively samples and processes the high-precision map and the simulation map, and classifies each point obtained by the sampling and processing to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; a deviation acquisition module, which obtains the deviation corresponding to each lane line according to the first lane line point set and the second lane line point set in combination with the matched relationship between the high-precision map and the simulation map; and a precision judgment module, which compares the deviation with a preset threshold and determines a precision checking result according to the comparison result.

[0015] The application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the precision checking method based on map data conversion according to any one of the above when executing the program.

[0016] The application further provides a non-transitory computer readable storage medium, which stores a computer program, wherein the computer program implements the steps of the precision checking method based on map data conversion according to any one of the above when executed by a processor.

[0017] The application further provides a computer program product, comprising a computer program, wherein the computer program implements the steps of the precision checking method based on map data conversion according to any one of the above when executed by a processor.

[0018] The application provides the precision checking method and device based on map data conversion, which classifies the points of the pre-acquired high-precision map and the simulation map, and combines the matching relationship of the pre-acquired high-precision map and the simulation map to acquire the deviation between the two maps, so as to filter out the corresponding points of which the deviation does not satisfy the threshold condition in post-processing, and to process these positions in a targeted manner to meet the deviation requirement, and meanwhile, the road number and the lane number to which the lane line not satisfying the deviation threshold range belongs can be fed back to optimize the map in a targeted manner. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort.

[0020] Figure 1 is one of the flowcharts of the precision checking method based on map data conversion provided by the present application;

[0021] Figure 2 is another flowchart of the precision checking method based on map data conversion provided by the present application;

[0022] Figure 3 is a schematic diagram of the lane line matching relationship of the high-precision map and the simulation map provided by the present application;

[0023] Figure 4 is a schematic diagram of the result of linear interpolation provided by the present application;

[0024] Figure 5 is a schematic diagram of the comparison result of the deviation and the preset threshold provided by the present application;

[0025] Figure 6 is a structural schematic diagram of the precision checking device based on map data conversion provided by the application;

[0026] Figure 7 is a structural schematic diagram of the electronic device provided by the application. DETAILED DESCRIPTION

[0027] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0028] Figure 1 A flowchart of a precision checking method based on map data conversion is shown, and the method comprises the following steps:

[0029] S11, obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map;

[0030] S12, respectively sampling the high-precision map and the simulation map, and classifying each point obtained by the sampling to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map;

[0031] S13, obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set, and combining a lane line matching relationship of the high-precision map and the simulation map obtained in advance;

[0032] S14, comparing the deviation with a preset threshold, and determining a precision checking result according to a comparison result.

[0033] It should be noted that S1N in the present specification does not represent the order of the precision checking method based on map data conversion, and the precision checking method based on map data conversion of the present application will be described in detail below. Figures 2-5

[0034] Step S11, obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map.

[0035] ​It should be noted that the high-definition map (High Definition Map, HD Map) is a special map relative to the ordinary navigation electronic map, and the data precision thereof can reach a centimeter level, and rich data dimensions contain information of surrounding static objects closely related to traffic in addition to road information; the simulation map is in the OpenDRIVE format.

[0036] In addition, in order to better illustrate the simulation map precision checking method and process proposed in the present scheme, a section of high-precision map in an actual project is intercepted, denoted as HDM. And the tool is used to convert it into a simulation map in the OpenDRIVE format, denoted as SM. In the actual simulation process, the road network is the core of the simulation map, and the non-road network information in the high-precision map is automatically ignored. The map can be understood as essentially describing the real road according to certain rules and data structures. Therefore, whether it is a high-precision map HDM or a simulation map SM, it maps the same section of real road.

[0037] The high-precision map can be used for positioning, perception, and planning modules of an automatic driving controller (ADU), and there is a significant difference between the road network definition and positioning of the simulation OpenDRIVE format map.

[0038] For example, Lanelet2 is a set of map frameworks for automatic driving, and the basic elements of the map include points (Points), polyline sequences (Linestrings) composed of points in front and back connection, polygons (Polygons), atomic units of lanes (Lanelets), areas (Arears), and control elements (Regulatory elements). Among them, Points are the only basic elements that store actual position information, and other basic elements are directly or indirectly composed of Points. Linestrings are ordered arrays generated by at least two Points through linear interpolation, used for shape description. Lanelet is the smallest unit to define a lane, and the traffic rules on a Lanelet are constant. The left and right lanestrings in the Lanelet represent the left and right lane lines of the lane. OSM (OpenStreetMap) is the standard format of the Lanelet2 software output map. The Nodes, Ways, and Relations three elements in OSM constitute the picture of the map. In the.osm file, Points are mapped to Nodes, and Lanestrings are mapped to Ways.

[0039] Different from Lanelet2 which uses points to describe and construct high-precision map features, OpenDRIVE uses lane lines describing road shape and driving direction as the basic elements of map description. In the OpenDRIVE standard, the road reference line is the most basic and necessary element of road description. All features on the road can be constructed along the reference line. The reference line can be a straight line, a circular arc line, a spiral line or a cubic polynomial fitting curve. In OpenDRIVE, the road is divided into driving directions by the center lane, and the number of the center lane is 0 without width. Other lanes are numbered in descending order to the right and ascending order to the left with the center lane as the starting point. If no lane deviation is used, the center lane is equivalent to the road reference line.

[0040] In step S12, the high-precision map and the simulation map are respectively sampled, and each point obtained by the sampling is classified to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map.

[0041] In this embodiment, referring to Figure 2 , the high-precision map and the simulation map are respectively sampled, and each point obtained by the sampling is classified, including: the high-precision map is sampled, and based on the lane line connection relationship, each point in the high-precision map obtained by the sampling is classified to obtain a first lane line point set corresponding to each lane line of the high-precision map; and based on a preset sampling step, each lane line of the simulation map is respectively sampled, and the sampling points obtained by the sampling are classified to obtain a second lane line point set corresponding to each lane line.

[0042] Specifically, based on the lane line connection relationship, each point in the high-precision map obtained by the sampling is classified, including: obtaining the starting point of each lane line; based on the lane line connection relationship, searching along the starting point of each lane line respectively, and according to all points obtained by the searching and the starting point of the corresponding lane line, obtaining a first lane line point set corresponding to each lane line.

[0043] It should be noted that for the high-precision map HDM, the basic element is a point point, and a plurality of points are connected to form a polyline sequence Linestrings composed of points connected in sequence. All Linestrings and points representing the same continuous lane line are determined by the connection relationship, so all points in the high-precision map HMD can be classified according to the continuous lane line. The representation method of the point is P_i_j_HMD, wherein i represents the sequence number of the continuous lane line, and j represents the sequence number of the point constituting the lane line. When j=0, it means that the point is the starting point of the lane line.

[0044] In the actual classification process, a starting point P_i_0_HMD is determined for each lane line using the visualization tool, and all lane line starting points form a set P0_HMD. Then, the first point P_1_0_HMD in P0_HMD is taken as the starting point, and the next point P_1_1_HMD is searched based on the connection relationship, until the point P_1_m_HDM has no next search object. This means that P_1_m_HMD is the end point of the first continuous lane line, and these points are stored in L_1_HDM. Through this classification method, each point in P0_HMD is searched one by one until the classification of all points in HMD is completed, and the classification results are stored in the dictionary L_HDM. Wherein:

[0045] P0_HMD = [P_1_0_HMD, P_2_0_HMD, ···, P_n_0_HMD], where n is the total number of lane lines;

[0046] L_i_HDM = [P_i_1_HMD, P_i_2_HMD, ···, P_i_m_HMD], i represents the ith continuous lane line, and m is the sequence number of the last point;

[0047] L_HDM = {1: L_i_HDM, 2: L_2_HMD, ···, n: L_n_HDM}, n is the total number of lane lines.

[0048] In addition, based on a preset sampling step, each lane line in the simulation map is sampled, and the sampled points are classified to obtain a second lane line point set corresponding to each lane line. It should be noted that the reference line is the most basic and essential element for describing a road in the OpenDRIVE standard, and all other features of the road can be described based on the reference line. Therefore, the simulation map cannot directly obtain the coordinates of all points from the map file (.xodr file) like the high-precision map HMD, but calculates the coordinates through the description of the reference line, the logical relationship between the road predecessors and successors, the expression of the lane width, and other information. In the OpenDRIVE format, the description of the reference line and the lane width is described by a cubic polynomial. Therefore, the above coordinate calculation effect can be achieved by calling a library file that supports parsing the OpenDRIVE format to obtain the corresponding point coordinates.

[0049] In the actual classification process, the API tool of OpenDRIVE provides a method to parse the.xodr file, and the coordinate values of the specified points on the specified lane line can be obtained through SetLanePosition and GetPositionData. Since the points whose coordinates are to be obtained need to be specified by oneself, the following rules can be used to realize the sampling of points: along the forward direction of the lane line, take a point every preset sampling step to ensure that the number of sampling points of the simulation map is greater than the number of Points of the simulation map for the same lane line; if the length of the unsampled part of the lane line being sampled is less than the preset sampling step, the sampling point is the end point of the lane line; to ensure that the sampled points are on the lane line, set the offset value to half of the current lane width when calling SetLanePosition. It should be noted that the preset sampling step can be set according to the actual design requirements and the driving path, such as 0.5 meters, i.e. sample_step = 0.5 meters, which is not further limited here.

[0050] After the above-mentioned manner is used to realize the sampling of points, the set L_i_HDM of all the sampling points P_i_k_SM on the ith continuous lane line can represent the lane line. The starting points P_i_0_SM of all the lane lines are combined into a set P0_SM. Finally, a dictionary L_SM storing all the sampling points of the complete map can also be obtained.

[0051] L_SM = {1 : L_i_SM, 2 : L_2_SM, ···, n : L_n_SM}, n is the total number of lane lines;

[0052] P0_SM = [P_1_0_SM, P_2_0_SM, ···, P_n_0_SM], n is the total number of lane lines;

[0053] L_i_SM = [P_i_1_SM, P_i_2_SM, ···, P_i_k_SM], i represents the ith continuous lane line, and k is the serial number of the last point.

[0054] In an optional embodiment, with continued reference to Figure 2Because the surface of the earth is uneven, it cannot be accurately described by a mathematical formula, so it is impossible to obtain a reference surface that can be used for map surveying. There are many coordinate systems that can be used when describing the positioning of the earth, such as the WGS84 coordinate system used by high-precision maps and the Universal Transverse Mercator Grid System (UTM) coordinate system used by the simulation map of the OpenDRIVE standard. Therefore, it is necessary to convert the latitude and longitude coordinates in the HDM to the UTM coordinates used in the SM, that is, after obtaining the first lane line point set, further comprising: converting the first lane line point set to the UTM coordinate system based on the Universal Transverse Mercator Grid System (UTM) coordinate system used by the simulation map.

[0055] It should be noted that the UTM coordinate is a plane rectangular coordinate, and such a coordinate grid system and the projection it is based on have been widely used in topographic maps. The representation format of the UTM coordinate is: longitude zone latitude zone east north, where east represents the projection distance from the central meridian of the longitude zone, and north represents the projection distance from the equator. The units of these two values are meters. UTM has 60 longitude zones and 20 latitude zones. When positioning UTM, it is necessary to determine the sub-area code in which it is located.

[0056] In addition, the conversion of latitude and longitude to UTM coordinates can be performed in the existing manner, and the present application does not make further limitations. It should be noted that after the conversion of latitude and longitude to UTM, the coordinate value is generally 6 or 7 bits. In order to simply express the coordinate value, the simulation map SM usually sets a reference point O (Offset_x, Offset_y), and the description of other features in the map is based on the reference point O. Therefore, after the conversion of the coordinates (latitude and longitude format) of the points in the HDM to UTM coordinates, the coordinates of the reference point need to be subtracted, so that the coordinate values of the same point in the HDM and the SM can be corresponded.

[0057] In an optional embodiment, continuing to refer to Figure 2 In addition, according to the first lane line point set and the second lane line point set, and in combination with the lane line matching relationship of the high-precision map and the simulation map that is obtained in advance, before obtaining the deviation of each lane line, further comprising: obtaining the lane line matching relationship of the high-precision map and the simulation map.

[0058] It should be noted that in step S11, all points in the high-precision map HDM are classified, and the classification results are stored in L HDM. In step S12, the simulation map is sampled, and the classification is realized at the same time, and the classification results are stored in L SM. Both L HDM and L SM store the numbers of n lane lines and the point sets discretized by each lane line. Before formally calculating the coordinate difference, it is necessary to ensure that each lane line in L HDM is accurately mapped to a lane line in L SM, and there is no one-to-many or many-to-one situation, refer to Figure 3 .

[0059] Step S13, according to the first lane line point set and the second lane line point set, and combining the lane line matching relationship of the high-precision map and the simulation map obtained in advance, the deviation corresponding to each lane line is obtained.

[0060] In this embodiment, according to the first lane line point set and the second lane line point set, and combining the lane line matching relationship of the high-precision map and the simulation map obtained in advance, the deviation corresponding to each lane line is obtained, including: using linear interpolation, mapping each second lane line point set to the high-precision map to obtain a third lane line point set; based on the matching relationship of the lane lines in the high-precision map and the simulation map obtained in advance, obtaining the paired first lane line point set and the third lane line point set; according to the paired first lane line point set and the third lane line point set, obtaining the deviation corresponding to each lane line. Specifically:

[0061] First, using linear interpolation, each second lane line point set is mapped to the high-precision map to obtain a third lane line point set. It should be noted that based on the above method, the point sets L HMD and L SM classified according to lane lines are obtained for HDM and SM maps, although both sets of point sets are discretized representations of the same lane line of the same road, but their discretization methods are different. For the SM map, it is sampled at a fixed sampling step Sample_step along the lane line forward direction; and for the HDM map, the discretization degree is determined by the map making process, which is not a fixed step, so the two sets of point sets cannot be directly compared.

[0062] For example, refer to Figure 4 , the first curve represents the lane line recorded in the high-precision map HDM, in which the black dot p i represents all points in the range of the lane line. The distance between two adjacent points is not equal, which is generally inversely proportional to the road curvature change rate. The arrow represents the logical relationship between the points. The third curve is the lane line described by the simulation map SM in the OpenDRIVE format through the cubic polynomial method. The black square p' iThis represents the points obtained through equal-distance sampling in step two. The distance between points is equal to the sampling step size, sample_step, which is 0.5 meters in this case.

[0063] Sampling point p' on the lane line of the SM map i The coordinates are (x' i ,y' i The mapping point p” of this point on the HDM lane line of the simulation map is determined by interpolation. i Its coordinates are (x' i ,y”i).p' i and p” i Having the same x-coordinate, the y-coordinate represents the deviation between two points. Since high-resolution (HDM) maps connect two points with straight lines, the simplest linear interpolation method can be used to determine all mapping points p". i Finally, update the point set in Li_i_HDM using p”. i Replace p i .

[0064] It should be noted that in this embodiment, the high-precision map (HDM) uses linear interpolation to map each set of second lane line points onto the high-precision map, resulting in a third set of lane line points, i.e., the new L_HDM dataset. Each point p” in the L_i_HDM dataset for any consecutive lane line… i (x' i ,y”i) are all related to point p' in L_i_SM in the simulation map. i (x' i ,y' i One-to-one correspondence.

[0065] Secondly, based on the matching relationship between the lane lines in the pre-acquired high-precision map and the simulation map, the first lane line point set and the third lane line point set are obtained.

[0066] Finally, based on the paired first lane line point set and the third lane line point set, the deviation of each lane line is obtained, specifically including: based on the paired first lane line point set and the third lane line point set, the deviation of all points of each lane line is obtained; based on the deviation of all points of each lane line, the deviation of each lane line is obtained.

[0067] p” i and p' i If two points have the same x-coordinate, the difference between their y-coordinates is the deviation.

[0068] It should be noted that, please continue to refer to the above. Figure 4 ,p” i and p' iThe two points have the same abscissa, and the difference between the ordinates is the deviation of a single point, that is, f(x i )=y’ i -y” i ;

[0069] Correspondingly, the deviation of each lane line is represented as:

[0070] f(n)=[f(x’0),f(x’1),f(x’2),···,f(x’ k )],

[0071] where k is the number of sampling points, and n is the lane line serial number.

[0072] In step S14, the deviation is compared with a preset threshold, and the accuracy check result is determined according to the comparison result.

[0073] In this embodiment, the deviation is compared with a preset threshold, and the accuracy check result is determined according to the comparison result, including: if the deviation meets the preset threshold condition, it is determined that the accuracy check result is that the accuracy is met; otherwise, it is determined that the accuracy check result is that the accuracy is not met, and the corresponding lane line needs to be adjusted. It should be noted that in the actual development process, a maximum preset threshold delta_max is set. Generally, it is considered that the lane whose deviation is less than the preset threshold delta_max has an acceptable influence on the automatic driving simulation test result, and the corresponding accuracy check result is that the accuracy is met, while the deviation greater than the threshold delta_max is unacceptable and needs to be modified.

[0074] For example, the deviation calculation result of a lane line is shown in Figure 5 , assuming that the threshold delta_max is 10 cm, then the deviation value in the range of -10-10 cm is acceptable, and the deviation outside this range is unacceptable. Therefore, the lane line corresponding to the black origin point on the simulation map SM in the figure needs to be modified.

[0075] Further, in order to facilitate the modification of the simulation map, based on the coordinate values of these points, the road number and lane number where they are located are calculated in reverse by using the analysis tool of OpenDRIVE. Finally, the obtained road number and lane number are de-duplicated, and these positions are processed in a targeted manner to meet the map deviation requirements.

[0076] In summary, the embodiment of the present application classifies the points of the pre-acquired high-precision map and the simulation map, and combines the matching relationship of the pre-acquired high-precision map and the simulation map to acquire the deviation between the two maps, so that in post-processing, the corresponding points of which the deviation does not satisfy the threshold condition are screened out, thereby processing these positions in a targeted manner to meet the deviation requirement, and meanwhile, the road number and the lane number to which the lane line not satisfying the deviation threshold range belongs can be fed back, so as to optimize the map in a targeted manner.

[0077] The map data conversion-based precision checking device provided by the present application is described below, and the map data conversion-based precision checking device described below can be correspondingly referred to the map data conversion-based precision checking method described above.

[0078] Figure 6 A structure diagram of a map data conversion-based precision checking device is shown, and the device comprises:

[0079] The data acquisition module 61 acquires a high-precision map and a simulation map, and the simulation map is converted from the high-precision map;

[0080] The classification module 62 respectively samples the high-precision map and the simulation map, and classifies each point obtained by the sampling to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map;

[0081] The deviation acquisition module 63 obtains the deviation corresponding to each lane line according to the first lane line point set and the second lane line point set, and combines the lane line matching relationship of the pre-acquired high-precision map and the simulation map;

[0082] The precision judgment module 64 compares the deviation with a preset threshold, and determines the precision checking result according to the comparison result.

[0083] In the embodiment, the classification module 62 comprises: a first classification unit which samples the high-precision map, and classifies each point in the high-precision map obtained by the sampling based on the lane line connection relationship to obtain the first lane line point set corresponding to each lane line in the high-precision map; and a second classification unit which samples each lane line in the simulation map based on a preset sampling step, and classifies the sampling points obtained by the sampling to obtain the second lane line point set corresponding to each lane line.

[0084] Specifically, the first classification unit comprises: a starting point acquisition subunit acquiring starting points of each lane line; a searching subunit searching along the starting points of each lane line based on a lane line connection relationship; and a first classification subunit obtaining a first lane line point set corresponding to each lane line according to all points obtained by searching and the starting points of the corresponding lane line.

[0085] In an optional embodiment, the device further comprises a coordinate conversion module converting the first lane line point set to a Universal Transverse Mercator (UTM) coordinate system based on a UTM coordinate system used by the simulation map.

[0086] In an optional embodiment, the device further comprises a matching relationship acquisition module acquiring a lane line matching relationship between the high-precision map and the simulation map before obtaining the deviations corresponding to each lane line.

[0087] The deviation acquisition module 63 comprises: an interpolation unit mapping each second lane line point set to the high-precision map to obtain a third lane line point set by using linear interpolation; a paired point set acquisition unit acquiring paired first lane line point sets and third lane line point sets based on the matching relationship between the lane lines in the high-precision map and the simulation map obtained in advance; and a deviation acquisition unit obtaining deviations corresponding to each lane line according to the paired first lane line point sets and the third lane line point sets.

[0088] Further, the deviation acquisition unit comprises: a point deviation acquisition subunit obtaining deviations of all points corresponding to each lane line according to the paired first lane line point sets and the third lane line point sets; and a lane line deviation acquisition subunit obtaining deviations corresponding to each lane line according to the deviations of all points corresponding to each lane line.

[0089] The precision judgment module 64 comprises: a comparison unit comparing the deviations with a preset threshold; and a result determination unit determining that the precision check result is precision compliance if the deviations meet the preset threshold condition, and otherwise determining that the precision check result is precision non-compliance, and the corresponding lane line needs to be readjusted.

[0090] In summary, the embodiments of the present application classify the points of the high-precision map and the simulation map obtained by the data acquisition module through the classification module, and obtain the deviations between the two maps by combining the matching relationship between the high-precision map and the simulation map obtained in advance through the deviation acquisition module, so as to filter out the corresponding points that do not meet the threshold condition by using the precision judgment module when post-processing, thereby processing these positions in a targeted manner to meet the deviation requirement, and at the same time, feeding back the road number and lane number to which the lane line that does not meet the deviation threshold range belongs, so as to optimize the map in a targeted manner.

[0091] Figure 7 An example of an entity structure diagram of an electronic device is shown inFigure 7 As shown, the electronic device can include a processor 71, a communications interface 72, a memory 73, and a communications bus 74, wherein the processor 71, the communications interface 72, and the memory 73 complete communications with each other through the communications bus 74. The processor 71 can invoke a logical instruction in the memory 73 to execute a precision checking method based on map data conversion, which includes: obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map; respectively performing sampling processing on the high-precision map and the simulation map, and classifying each point obtained by the sampling processing to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set, and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; and comparing the deviation with a preset threshold value, and determining a precision checking result according to a comparison result.

[0092] In addition, the logical instruction in the memory 73 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or partly or parts of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0093] In another aspect, the present application also provides a computer program product comprising a computer program, which can be stored on a non-transitory computer readable storage medium, and the computer program is executable by a processor to enable the computer to perform the precision checking method based on map data conversion provided by the above-mentioned methods, which comprises: obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map; performing sampling processing on the high-precision map and the simulation map respectively, and classifying each point obtained by the sampling processing to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set, and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; and comparing the deviation with a preset threshold value, and determining a precision checking result according to a comparison result.

[0094] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having a computer program stored thereon, and the computer program is executable by a processor to implement the precision checking method based on map data conversion provided by the above-mentioned methods, which comprises: obtaining a high-precision map and a simulation map, the simulation map being converted from the high-precision map; performing sampling processing on the high-precision map and the simulation map respectively, and classifying each point obtained by the sampling processing to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set, and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; and comparing the deviation with a preset threshold value, and determining a precision checking result according to a comparison result.

[0095] The device embodiments described above are only schematic, wherein the units shown as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment scheme. Those skilled in the art can understand and implement without creative labor.

[0096] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0097] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A map data conversion-based precision checking method, characterized by, The method comprises the following steps: obtaining a high-precision map and a simulation map converted from the high-precision map; sampling the high-precision map and the simulation map respectively, and classifying each point obtained by sampling to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; comparing the deviation with a preset threshold, and determining a precision check result according to a comparison result; the step of obtaining a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance comprises the following steps: mapping each second lane line point set to the high-precision map by linear interpolation to obtain a third lane line point set; obtaining a matched second lane line point set and a third lane line point set based on the matching relationship of the lane lines in the high-precision map and the simulation map obtained in advance; obtaining a deviation corresponding to each lane line according to the matched second lane line point set and the third lane line point set.

2. The map data conversion-based precision checking method according to claim 1, characterized by, the steps of sampling the high-precision map and the simulation map respectively and classifying each point obtained by sampling comprise the following steps: sampling the high-precision map, and classifying each point in the high-precision map obtained by sampling based on a lane line connection relationship to obtain a first lane line point set corresponding to each lane line of the high-precision map; sampling each lane line in the simulation map based on a preset sampling step, and classifying a sampling point obtained by sampling to obtain a second lane line point set corresponding to each lane line.

3. The map data conversion-based precision checking method according to claim 2, characterized in that, the step of classifying each point in the high-precision map obtained by sampling based on the lane line connection relationship comprises the following steps: obtaining a starting point of each lane line; searching along the starting point of each lane line based on a lane line connection relationship, and obtaining a first lane line point set corresponding to each lane line according to all points obtained by searching and the starting point of the corresponding lane line.

4. The map data conversion-based precision checking method according to claim 3, characterized by, the step of obtaining a deviation corresponding to each lane line according to the matched first lane line point set and the third lane line point set comprises the following steps: obtaining a deviation of all points corresponding to each lane line according to the matched first lane line point set and the third lane line point set; obtaining a deviation corresponding to each lane line according to the deviation of all points corresponding to each lane line.

5. The map data conversion-based precision checking method according to claim 1, characterized by, after obtaining the first lane line point set, the method further comprises the following step: converting the first lane line point set to a Universal Transverse Mercator (UTM) coordinate system used by the simulation map.

6. The map data conversion-based precision checking method according to claim 1, characterized by, the step of comparing the deviation with a preset threshold and determining a precision check result according to a comparison result comprises the following steps: if the deviation meets a preset threshold condition, determining that the precision check result is that the precision meets the requirement; otherwise, determining that the precision check result is that the precision does not meet the requirement, and each lane line needs to be adjusted again.

7. A map data conversion-based precision checking device, characterized by comprising: ​ The data acquisition module acquires a high-precision map and a simulation map, the simulation map being converted from the high-precision map; The classification module respectively performs sampling processing on the high-precision map and the simulation map, and classifies each point obtained by the sampling processing to obtain a first lane line point set corresponding to each lane line of the high-precision map and a second lane line point set corresponding to each lane line of the simulation map; The deviation acquisition module obtains a deviation corresponding to each lane line according to the first lane line point set and the second lane line point set, and in combination with a lane line matching relationship of the high-precision map and the simulation map obtained in advance; The precision judgment module compares the deviation with a preset threshold, and determines a precision check result according to a comparison result; The deviation acquisition module comprises: The interpolation unit maps each second lane line point set to the high-precision map by linear interpolation to obtain a third lane line point set; The paired point set acquisition unit acquires paired second lane line point sets and third lane line point sets based on a matching relationship of lane lines in the high-precision map and the simulation map obtained in advance; The deviation acquisition unit obtains a deviation corresponding to each lane line according to the paired second lane line point sets and the third lane line point sets.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps of the precision check method based on map data conversion according to any one of claims 1 to 6. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the precision check method based on map data conversion according to any one of claims 1 to 6.

Images