High-precision map compilation method, electronic device and machine-readable storage medium

Abstract

The present invention provides a method for compiling a high-precision map, an electronic device, and a machine-readable storage medium. The method includes obtaining high-precision map data in a first format; wherein the high-precision map data in the first format includes a road facility model, a lane model, and a road model; receiving a compilation operation for the high-precision map data in the first format; and outputting high-precision map data in a second format based on the high-precision map data in the first format; wherein the high-precision map data in the second format is OpenDrive map data, and the OpenDrive map data includes an Objects layer module, a Signals layer module, a Lanes layer module, a Connection layer module, a Junction layer module, and a Road layer module; the road model and lane model correspond to the Road layer module; the road model and lane model correspond to the Lanes layer module; the road facility model, lane model, and road model correspond to the Signals layer module; the road facility model, lane model, and road model correspond to the Objects layer module; the road model corresponds to the Junction layer module; and the road model corresponds to the Connection layer module.

Description

Technical Field

[0001] The present invention relates to the field of high-precision map technology, and in particular to a high-precision map compilation method, electronic device, and machine-readable storage medium. Background Art

[0002] Currently, the field of autonomous driving often requires the use of high-precision maps in a specific format, such as OpenDrive. However, the format of the collected raw HD map data or HD map data in other formats often does not meet the standards of the specified format or is not sufficiently compatible with the HD map in the specified format, making it impossible to use directly and having poor versatility. Summary of the Invention

[0003] An object of the present invention is to solve any of the technical problems existing in the above-mentioned prior art.

[0004] A further object of the present invention is to provide OpenDrive high-precision map data with high versatility.

[0005] In particular, the present invention provides a method for compiling a high-precision map, comprising:

[0006] Acquire high-precision map data in a first format; wherein the high-precision map data in the first format includes at least one of a road facility model, a lane model, and a road model;

[0007] Receiving a compilation operation for high-precision map data in the first format;

[0008] Outputting high-precision map data in a second format according to the high-precision map data in the first format; wherein

[0009] The high-precision map data in the second format is OpenDrive map data, and the OpenDrive map data includes an Objects layer module, a Signals layer module, a Lanes layer module, a Connection layer module, a Junction layer module, and a Road layer module;

[0010] The road model and the lane model correspond to the Road layer module;

[0011] The road model and the lane model correspond to the Lanes layer module;

[0012] The road facility model, the lane model and the road model correspond to the Signals layer module;

[0013] The road facility model, the lane model, and the road model correspond to the Objects layer module;

[0014] The road model corresponds to the Junction layer module;

[0015] The road model corresponds to the Connection layer module.

[0016] Optionally, the road model includes a road reference line and road attributes, and the lane model includes a road connectivity relationship;

[0017] The road model and the lane model correspond to the Road layer module in the following manner:

[0018] Reading the road reference line, the road attributes, and the road connectivity relationship;

[0019] Calculating the road reference line attributes, the road connectivity relationship, and the road attribute geometry according to the read road reference line, the road attributes, and the road connectivity relationship;

[0020] The calculated results are mapped to the Road layer module.

[0021] Optionally, the lane model includes a lane centerline, lane lines, lane attributes, and road connectivity, and the road model includes a road reference line;

[0022] The road model and the lane model correspond to the Lanes layer module in the following manner:

[0023] Reading the lane centerline, the road reference line, the lane line, the lane attributes, and the lane connectivity relationship;

[0024] Calculating lane attributes, lane connectivity, lane width, and lane geometry based on the read lane centerline, the road reference line, the lane line, the lane attributes, and the lane connectivity relationship;

[0025] Map the calculation results to the Lanes layer module.

[0026] Optionally, the road facility model includes signal lights and traffic signs, the lane model includes lane center lines, and the road model includes road reference lines;

[0027] The road facility model, the lane model, and the road model correspond to the Signals layer module in the following manner:

[0028] Reading the lane centerline, the road reference line, the signal light, and the traffic sign;

[0029] Matching the signal lights and traffic signs with the road according to the read lane center line, the road reference line, the signal lights, and the traffic signs;

[0030] Map the corresponding results to the Signals layer module.

[0031] Optionally, the road facility model includes a median strip, the lane model includes a lane centerline, a crosswalk, a stop line, and a guide arrow, and the road model includes a road reference line;

[0032] The road facility model, the lane model, and the road model correspond to the Objects layer module in the following manner:

[0033] Reading the lane center line, the road reference line, the isolation strip, the crosswalk, the stop line, and the guide arrow;

[0034] Matching the road printing surface with the road according to the read lane center line, the road reference line, the isolation strip, the crosswalk, the stop line, and the guide arrow;

[0035] Map the corresponding results to the Objects layer module.

[0036] Optionally, the road model includes a road reference line and road attributes;

[0037] The road model corresponds to the Junction layer module in the following way:

[0038] Reading the road reference line and the road attributes, and creating an intersection according to the road reference line and the road attributes;

[0039] Merge the created intersection with the intersection in the high-precision map data in the first format;

[0040] The merged result is mapped to the Junction layer module.

[0041] Optionally, the road model includes connectivity relations and road attributes;

[0042] The road model corresponds to the Connection layer module in the following way:

[0043] Reading the connectivity relationship and the road attributes, and establishing a correspondence relationship between a predecessor and a successor road according to a correspondence relationship between a road ID in the connectivity relationship and a predecessor and a successor road ID in the corresponding road attributes;

[0044] Establish the connectivity relationship between the predecessor and the successor in the intersection according to the corresponding relationship between the predecessor and the successor;

[0045] The established connectivity relationship is mapped to the Connection layer module.

[0046] Optionally, before outputting the high-precision map data in the second format according to the high-precision map data in the first format, the method further includes:

[0047] Optimizing the high-precision map data in the first format;

[0048] The optimizing process of the high-precision map data in the first format includes:

[0049] Creating intersections for the high-precision map data in the first format; and / or

[0050] Merging non-intersection roads in the high-precision map data in the first format; and / or

[0051] Merging the roads within the intersection of the high-precision map data in the first format; and / or

[0052] Optimize the road reference lines in the high-precision map data in the first format.

[0053] Optionally, creating an intersection for the high-precision map data in the first format includes:

[0054] Reading a road reference line list in the high-precision map data in the first format, and filtering out non-intersection data in the road reference line list;

[0055] Obtaining a successor road list and / or a predecessor road list of the roads in the non-intersection data;

[0056] Determine the number of successors and / or predecessors in the non-intersection data, and obtain a successor road list and / or a predecessor road list for each successor and / or predecessor;

[0057] Determining whether the successor road list and / or predecessor road list of each successor and / or predecessor road list contains road data within the intersection;

[0058] If not included, the successor road list and / or predecessor road list of each successor and / or predecessor is added to the road list in the intersection;

[0059] The merging of non-intersection roads of the high-precision map data in the first format includes:

[0060] Reading a road list in the high-precision map data in the first format and filtering out non-intersection roads in the road list;

[0061] If the successor of a road not in an intersection is a road in an intersection, obtain the road entering the intersection;

[0062] Obtain the list of predecessor roads at the intersection and merge it into the road list;

[0063] If the front road in the non-intersection road is the intersection road, obtain the road at the exit of the intersection;

[0064] Obtain a list of subsequent roads at the exit intersection and merge it into the road list;

[0065] The merging of roads within the intersection of the high-precision map data in the first format includes:

[0066] Reading a road list from the high-precision map data in the first format and filtering out non-intersection roads in the road list;

[0067] If the subsequent part of the road list that has not been filtered out is a non-intersection road, obtain a list of roads in the intersection that will exit the intersection and merge it into the road list;

[0068] Optimizing the road reference lines in the high-precision map data in the first format includes:

[0069] Obtaining a lane centerline and a leftmost lane line group in the high-precision map data in the first format;

[0070] Obtain the predecessor and successor of the lane centerline and the leftmost lane line group;

[0071] Determining whether the distance and angle difference at the junction exceed a first preset threshold and a second preset threshold, respectively; wherein the junction represents the topological connection between a geometric line of the current lane and a geometric line of a preceding or succeeding lane, the distance refers to the distance between the starting point of the current line and the ending point of the preceding line, or the distance between the ending point of the current line and the starting point of the succeeding line, and the angle refers to the angle between the first line segment of the current line and the last line segment of the preceding line, or the angle between the last line segment of the current line and the first line segment of the succeeding line;

[0072] If so, perform Bezier curve optimization on the road reference line and update the lane line.

[0073] Optionally, obtaining the lane center line and the leftmost lane line group in the high-precision map data in the first format may include:

[0074] Reading a list of non-intersection lane centerlines in the high-precision map data in the first format, and obtaining a leftmost lane centerline of the same road in the list of non-intersection lane centerlines;

[0075] If the leftmost lane centerline and the left lane line are not in the same direction, flip the left lane line to obtain the lane centerline and the leftmost lane line group; and / or

[0076] Reading a list of intersection lane centerlines in the high-precision map data in the first format, and obtaining lane centerlines belonging to the same road in the intersection lane centerline list to form a lane centerline group;

[0077] Determine whether the lane centerlines in the lane centerline group intersect;

[0078] If there is an intersection, the lane centerline with the intersection is simplified to obtain the lane centerline and the leftmost lane line group.

[0079] According to another aspect of the present invention, there is provided an electronic device, comprising:

[0080] A memory and a processor, wherein a control program is stored in the memory, and when the control program is executed by the processor, it is used to implement the high-precision map compilation method according to any one of the above-mentioned methods.

[0081] According to another aspect of the present invention, a machine-readable storage medium is provided, on which a machine executable program is stored, wherein the machine executable program is used to implement the high-precision map compilation method according to any one of the above-mentioned methods when executed by a processor.

[0082] In the high-precision map compilation method of the present invention, high-precision map data in a first format is obtained, a compilation operation for the high-precision map data in the first format is received, and high-precision map data in a second format is output based on the high-precision map data in the first format. This can obtain OpenDrive map data with high versatility, which can be well applied to fields such as autonomous driving simulation and autonomous driving algorithm testing.

[0083] Based on the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will become more aware of the above and other objects, advantages and features of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Hereinafter, some specific embodiments of the present invention will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0085] Figure 1 is a flowchart of a method for compiling a high-precision map according to an embodiment of the present invention;

[0086] Figure 2 is a schematic diagram of high-precision map data in a first format according to another embodiment of the present invention;

[0087] Figure 3 is a flowchart of a method for compiling a high-precision map according to an embodiment of the present invention;

[0088] Figure 4 is a schematic block diagram of an electronic device according to an embodiment of the present invention;

[0089] Figure 5 is a schematic block diagram of a machine-readable storage medium according to one embodiment of the present invention;

[0090] Figure 6 is a schematic diagram of OpenDrive map data according to another embodiment of the present invention;

[0091] Figure 7 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0092] Figure 8 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0093] Figure 9 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0094] Figure 10 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0095] Figure 11 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0096] Figure 12 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0097] Figure 13 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0098] Figure 14 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0099] Figure 15 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0100] Figure 16 is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention;

[0101] Figure 17 It is a flowchart of a method for compiling a high-precision map according to another embodiment of the present invention. DETAILED DESCRIPTION

[0102] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0103] Figure 1 FIG. 1 is a flow chart of a method for compiling a high-precision map according to an embodiment of the present invention. The method may include steps S102 to S106.

[0104] Step S102: Obtain high-precision map data in a first format.

[0105] In this step, it can be the storage path of the high-precision map data in the first format, or it can be a file of the high-precision map data in the first format. The acquired high-precision map data can be the original high-precision map data collected in advance. Those skilled in the art can understand the meaning of high-precision map in this solution, and high-precision map is a clear concept in this field. The first format of high-precision map data is not strictly limited, and can generally be GeoJson format. In some embodiments of the present application, the first format of the original high-precision map can be a high-precision map in a non-OpenDrive format, or it can be a file that only includes high-precision map data, etc. The high-precision map data may include multiple layer data, such as Figure 2As shown, high-precision map data in the first format can generally include at least one of a road facility model, a lane model, and a road model. The road facility model layer can generally include layer elements such as signals, traffic signs, medians, roadsides, feature attributes, and association relationships. Feature attributes generally refer to the type of road facility model. Roadside generally refers to roadside facilities, such as streetlights, obstacles, and green belts. Association relationships generally refer to the correspondence between roadside and other road facility models and road or lane identification (ID). The lane model layer can generally include layer elements such as lane lines, lane centerlines, stop lines, guide arrows, crosswalks, road connectivity, lane attributes, association relationships, and topological relationships. Road connectivity refers to the connectivity between a road and its corresponding predecessor or successor road, generally associated using a road ID. Lane attributes generally refer to certain lane type fields, such as lane width, type, driving direction, and length. Association relationships generally refer to the relationship between the predecessor and successor of a lane centerline. A road model layer typically includes elements such as road reference lines, connectivity relationships, and road attributes. Connectivity relationships in a road model generally refer to the topological connectivity between a road and its subsequent roads, as well as the connectivity between a road and the roads within its intersection. Road attributes are additional information about the road, such as driving direction, speed limit, and road type. In some embodiments of this application, the relative positions or data of various layers can be interchangeable.

[0106] Step S104: Receive a compilation operation on high-precision map data in the first format.

[0107] In this step, a visual operation interface may be provided, and a compile button may be provided on the visual operation interface. The technician's click of the compile button may be regarded as a compilation operation. Of course, the visual operation interface may also be provided with a selection button for high-precision map data in the first format. The technician may click the selection button to find the high-precision map data in the first format to be compiled, thereby facilitating the execution of step S102.

[0108] Step S106: Outputting high-precision map data in a second format based on the high-precision map data in the first format.

[0109] In this step, the high-precision map data in the second format can be OpenDrive map data, which includes the Objects layer module, Signals layer module, Lanes layer module, Connection layer module, Junction layer module, and Road layer module. The road model and lane model correspond to the Road layer module; the road model and lane model correspond to the Lanes layer module; the road facility model, lane model, and road model correspond to the Signals layer module; the road facility model, lane model, and road model correspond to the Objects layer module; the road model corresponds to the Junction layer module; and the road model corresponds to the Connection layer module. The high-precision map data in the second format can be output to a specified path.

[0110] In this embodiment, high-precision map data in a first format is obtained, a compilation operation for the high-precision map data in the first format is received, and high-precision map data in a second format is output based on the high-precision map data in the first format. This can result in OpenDrive map data with high versatility, which can be well applied to fields such as autonomous driving simulation and autonomous driving algorithm testing.

[0111] Specifically, OpenDrive is a high-precision map format that is primarily used for the production of high-precision autonomous driving reference maps, serving as a basis for vehicle-mounted references and driving simulator testing and applications. OpenDrive map data is primarily used to describe road network structures, scenes, roads, and road conditions, and is primarily used for simulation. Figure 6 , OpenDrive map data can include Objects layer module, Signals layer module, Lanes layer module, Connection layer module, Junction layer module and Road layer module. Among them, Connection layer module and Junction layer module belong to intersection module; Objects layer module, Signals layer module, Lanes layer module and Road layer module belong to road module. Objects layer module can mainly contain object information in the map; Signals layer module can mainly contain sign information in the map; Lanes layer module can mainly contain lane information in the map; Connection layer module can mainly contain intersection connection relationship information in the map; Junction layer module can mainly contain intersection information in the map; Road layer module can mainly contain road information in the map.

[0112] See also Figure 7,In one embodiment of the present invention, the road model includes a road reference line and road attributes, and the lane model includes a road connectivity relationship;

[0113] The road model and lane model correspond to the Road layer module in the following ways:

[0114] Step S11: Read road reference lines, road attributes, and road connectivity relationships;

[0115] Step S12: Calculating the road reference line attributes, road connectivity relationships, and road attribute geometry based on the read road reference line, road attributes, and road connectivity relationships;

[0116] Step S13: Map the calculated results to the Road layer module.

[0117] Specifically, calculating road reference line attributes can refer to converting or reassigning field types in the road attribute table. For example, mapping the original road type 1, original road type 2, or original road type 3 to a new type value a. Another example is replacing the geometry of a road reference line with the geometry of a lane line, or replacing a certain type of value with a default value. Calculating road connectivity relationships can refer to establishing connectivity relationships between road IDs in the connectivity relationship table and their corresponding predecessor or successor road IDs. Calculating road attribute geometry can refer to simplifying, abstracting, or reassigning spatial attribute fields in the road attribute layer.

[0118] See also Figure 8 ,In one embodiment of the present invention, the lane model includes a lane center line, lane lines, lane attributes, and road connectivity relationships, and the road model includes a road reference line;

[0119] The road model and lane model correspond to the Lanes layer module in the following way:

[0120] Step S21: Read the lane centerline, road reference line, lane line, lane attributes, and lane connectivity;

[0121] Step S22: Calculating lane attributes, lane connectivity, lane width, and lane geometry based on the read lane centerline, road reference line, lane line, lane attributes, and lane connectivity;

[0122] Step S23: Map the calculation results to the Lanes layer module.

[0123] Specifically, the calculation of lane attributes can be the conversion and unification of lane attributes. The calculation of lane connectivity can refer to grouping by road ID or lane centerline ID for easy query. The calculation of lane width can generally refer to calculating the distance between the left lane line and the right lane line on the lane. In the process of calculating the distance between the left lane line and the right lane line on the lane, a polynomial expression can be fitted piecewise according to the change in lane width. The calculation of lane geometry can be performed by simplifying, extracting, and reassigning the spatial attribute fields of the lane layer. Reassignment can be understood as simplifying, extracting, and reassigning the values ​​of the spatial fields to the original fields after operations such as simplifying and extracting them.

[0124] See also Figure 9 ,In one embodiment of the present invention, the road facility model includes signal lights and traffic signs, the lane model includes lane center lines, and the road model includes road reference lines;

[0125] The road facility model, lane model, and road model correspond to the Signals layer module in the following ways:

[0126] Step S31: Reading lane center lines, road reference lines, signal lights, and traffic signs;

[0127] Step S32: Matching the signal lights and traffic signs with the road based on the read lane center lines, road reference lines, signal lights, and traffic signs;

[0128] Step S33: Map the corresponding results to the Signals layer module.

[0129] Specifically, using traffic lights as an example, lane centerline data contains the corresponding road ID, while traffic light data contains the road ID corresponding to the lane centerline. Based on the shared road IDs between the two, a mapping relationship between traffic light IDs and road IDs is established, thereby achieving the correspondence between traffic lights and roads. Traffic signs and roads can also be mapped using the same method for mapping traffic lights and roads, which will not be further elaborated here. In addition to establishing the corresponding relationship using the above method, the above-mentioned relationship can also be established using an association table.

[0130] See also Figure 10 In one embodiment of the present invention, the road facility model includes a median strip, the lane model includes a lane centerline, a crosswalk, a stop line, and a guide arrow, and the road model includes a road reference line;

[0131] Road facility models, lane models, and road models correspond to the Objects layer module in the following ways:

[0132] Step S41: Read the lane centerline, road reference line, isolation strip, crosswalk, stop line, and guide arrows;

[0133] Step S42: Matching the road printing surface with the road based on the read lane center line, road reference line, isolation strip, crosswalk, stop line, and guide arrow;

[0134] Step S43: Map the corresponding results to the Objects layer module.

[0135] Specifically, lane centerlines, road reference lines, medians, crosswalks, stop signs, and guide arrows can all be used as elements of a road printed surface, thereby determining the road printed surface. Road printed surface data typically includes the associated lane ID, and lane data includes the road reference line ID. Based on the correspondence between lane IDs and road reference line IDs, a mapping relationship between the road printed surface and the road reference line can be established, thereby aligning the road printed surface with the road. In addition to the aforementioned mapping scheme, the mapping between road printed surfaces and roads can also be achieved through an association table.

[0136] See also Figure 11 ,In one embodiment of the present invention, the road model includes a road reference line and road attributes;

[0137] The road model corresponds to the Junction layer module in the following ways:

[0138] Step S51: Read the road reference line and road attributes, and create an intersection according to the road reference line and road attributes;

[0139] Step S52: merging the created intersection with the intersection in the high-precision map data in the first format;

[0140] Step S53: Map the merged result to the Junction layer module.

[0141] See also Figure 12 ,In one embodiment of the present invention, the road model includes connectivity,relationships and road attributes;

[0142] The road model corresponds to the Connection layer module in the following ways:

[0143] Step S61: Reading the connectivity relationship and road attributes, and establishing a correspondence between the predecessor and successor roads based on the correspondence between the road ID in the connectivity relationship and the predecessor and successor road IDs in the corresponding road attributes;

[0144] Step S62: establishing a connectivity relationship between the predecessor and the successor in the intersection according to the correspondence between the predecessor and the successor;

[0145] Step S63: Map the established connectivity relationship to the Connection layer module.

[0146] In one embodiment of the present invention, before outputting high-precision map data in a second format according to the high-precision map data in a first format, the method further includes:

[0147] Optimize the high-precision map data in the first format;

[0148] See also Figure 17 , the optimization processing of the high-precision map data in the first format includes:

[0149] Step S702: Creating intersections for high-precision map data in the first format; and / or

[0150] Step S703: merging non-intersection roads in the high-precision map data of the first format; and / or

[0151] Step S704: merging the roads within the intersection of the high-precision map data in the first format; and / or

[0152] Step S705: Optimizing the road reference lines in the high-precision map data in the first format; and / or

[0153] Step S706: Processing the connectivity of the high-precision map data in the first format; and / or

[0154] Step S707: Convert the coordinates of the high-precision map data in the first format.

[0155] In this embodiment, optimizing the high-precision map data in the first format can reduce anomalies in the high-precision map data, resulting in more accurate high-precision map data and facilitating the output of more standardized OpenDrive map data. Furthermore, steps S702 through S707 may be executed independently of one another, without any order restrictions or dependencies. That is, steps S702 through S707 may be executed in an "and / or" relationship.

[0156] In one embodiment of the present invention, see Figure 13 , creating intersections for high-precision map data in the first format includes:

[0157] Step S71: Reading a road reference line list in the high-precision map data in the first format, and filtering out non-intersection data in the road reference line list;

[0158] Step S72: Obtain a list of successor roads and / or a list of predecessor roads of the roads in the non-intersection data. This step may traverse the non-intersection data and obtain a list of successor roads and / or a list of predecessor roads of the roads in the non-intersection data according to a relationship table. The relationship table here may be a connectivity relationship table.

[0159] Step S73: determining the number of successors and / or predecessors in the non-intersection data, and obtaining a successor road list and / or a predecessor road list of each successor and / or predecessor;

[0160] Step S74: determining whether the successor road list and / or predecessor road list of each successor and / or predecessor road list contains road data within the intersection;

[0161] If not included, step S75: adding the successor road list and / or predecessor road list of each successor and / or predecessor to the road list within the intersection.

[0162] In this embodiment, creating intersections for HD map data can make the HD map data more compatible with the OpenDrive intersection data representation specifications. If the successor road list and / or predecessor road list of each successor and / or predecessor contains road data within the intersection, no processing is required.

[0163] In one embodiment of the present invention, see Figure 14 Merging non-intersection roads of the high-precision map data in the first format may include:

[0164] Step S81: Reading a road list in the high-precision map data in the first format, and filtering out non-intersection roads in the road list;

[0165] Step S82: If the successor of the non-intersection road is a road within the intersection, obtain the road at the entrance to the intersection. In this step, the non-intersection roads may be traversed to obtain the successor therein.

[0166] Step S83: Obtain a list of predecessor roads at the road entering the intersection and merge it into the road list;

[0167] Step S84: If the preceding vehicle in the non-intersection road is an intersection road, obtain the exit road of the intersection;

[0168] Step S85: Obtain a list of subsequent roads in the road exiting the intersection and merge it into the road list.

[0169] In this embodiment, the preceding road list can be obtained recursively. The recursion terminates when the preceding road is not within the intersection or the preceding road length exceeds a first threshold. The subsequent road list can also be obtained recursively. The recursion terminates when the subsequent road is not within the intersection or the subsequent road length exceeds a second threshold. Those skilled in the art can configure the values ​​of the first and second thresholds, which are not specifically defined herein. This embodiment can simplify and optimize road and lane data.

[0170] In one embodiment of the present invention, see Figure 15 Merging the roads within the intersection of the high-precision map data in the first format may include:

[0171] Step S91: Reading a road list in the high-precision map data in the first format, and filtering out non-intersection roads in the road list;

[0172] Step S92: If the subsequent roads in the unfiltered portion of the road list are non-intersection roads, a list of roads in the intersection that will exit the intersection is obtained and merged into the road list.

[0173] In one embodiment of the present invention, see Figure 16 , optimizing the road reference lines in the first format of high-precision map data includes:

[0174] Step S601: Obtaining a lane centerline and a leftmost lane line group in high-precision map data in a first format;

[0175] Step S602: Obtain the predecessor and successor of the lane centerline and the leftmost lane line group. In this step, the predecessor and successor of the lane centerline and the leftmost lane line group can be obtained using the connectivity table.

[0176] Step S603: determining whether the distance and angle difference at the attachment point exceed a first preset threshold and a second preset threshold respectively;

[0177] If yes, step S604: perform Bezier curve optimization on the road reference line and update the lane line.

[0178] If not, step S605: no processing.

[0179] In this embodiment, the connection point represents the topological connection point between the geometric line of the current lane and the geometric line of the predecessor or successor lane. The distance refers to the distance between the starting point of the current line and the end point of the predecessor line, or the distance between the end point of the current line and the starting point of the successor line. The angle refers to the angle between the first line segment of the current line and the last line segment of the predecessor line, or the angle between the last line segment of the current line and the first line segment of the successor line. Those skilled in the art can set the size of the first preset threshold and the second preset threshold, which is not specifically limited here. The meaning of the current lane can refer to the lane to be requested, such as requiring the predecessor or successor of lane a, and the current lane mentioned later refers to lane a.

[0180] In one embodiment of the present invention, obtaining a lane centerline and a leftmost lane line group in high-precision map data in a first format may include:

[0181] Read the non-intersection lane centerline list in the high-precision map data of the first format, and obtain the leftmost lane centerline of the same road in the non-intersection lane centerline list;

[0182] If the leftmost lane centerline and the left lane line are not in the same direction, flip the left lane line to obtain the lane centerline and the leftmost lane line group; and / or

[0183] Reading a list of intersection lane centerlines in the high-precision map data of the first format, obtaining lane centerlines belonging to the same road in the intersection lane centerline list to form a lane centerline group;

[0184] Determine whether the lane centerlines in the lane centerline group intersect;

[0185] If there is an intersection, the lane centerline with the intersection is simplified to obtain the lane centerline and the leftmost lane line group.

[0186] In this embodiment, the general principle of de-intersection simplification is to start from the leftmost side of the road, connect the front and rear roads in sequence without crossing, and put the excess part at the end to ensure that the front and rear roads are connected without crossing connections.

[0187] In one embodiment of the present invention, optimizing high-precision map data may include:

[0188] Convert the coordinates of high-precision map data.

[0189] Specifically, for example, the WGS84 standard coordinate system can be used, then encrypted into the GCJ02 coordinate system, and then geometric operations can be performed using the UTM projected coordinate system. Converting the coordinates of high-precision map data is conducive to achieving map data encryption and standardization.

[0190] In one embodiment of the present invention, before outputting the OpenDrive map data in step S106 , a quality check may be performed on the OpenDrive map data.

[0191] Specifically, for example, road connectivity and elevation anomalies can be checked, the association between intersections and lanes can be checked, and the connectivity of roads within intersections can be checked, etc., thereby further improving the quality of OpenDrive map data.

[0192] In one embodiment of the present invention, optimizing high-precision map data may include:

[0193] Process the connectivity of high-precision map data.

[0194] Specifically, processing the connectivity of high-precision map data may include:

[0195] Based on the associative relationship data of the high-precision map data, the roads, lanes and features of the geometric layer are topologically processed, and the relationship data is updated.

[0196] Figure 31 is a flow chart of a method for compiling a high-precision map according to one embodiment of the present invention. The compilation method may include the following steps:

[0197] Step S502: Obtain high-precision map data in the first format.

[0198] Step S504: receiving a compilation operation for high-precision map data in a first format;

[0199] Step S506: Optimize the high-precision map data in the first format.

[0200] Step S508: Output high-precision map data in a second format based on the high-precision map data in the first format.

[0201] Step S510: Perform a quality check on the high-precision map data in the second format.

[0202] Step S512: Output high-precision map data in the second format.

[0203] Figure 4 is a schematic block diagram of an electronic device according to an embodiment of the present invention. Based on the same concept, the present invention also provides an electronic device 400. Electronic device 400 includes a memory 402 and a processor 401. Memory 402 stores a control program that, when executed by processor 401, implements any of the above-described methods for compiling a high-precision map.

[0204] Figure 5 The figure is a schematic block diagram of a machine-readable storage medium according to an embodiment of the present invention. Based on the same concept, the present invention also provides a machine-readable storage medium having a machine-executable program stored thereon, wherein the machine-executable program, when executed by a processor, is used to implement the high-precision map compilation method of the above embodiment.

[0205] The machine-readable storage medium 300 may be a tangible device that can hold and store instructions for use by an instruction execution device. The machine-readable storage medium 300 may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of the machine-readable storage medium 300 include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanical encoding device such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof.

[0206] The machine executable program 310 or code described herein can be downloaded from the machine-readable storage medium 300 to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer program from the network and forwards the computer program to be stored in the machine-readable storage medium 300 in each computing / processing device.

[0207] The machine executable program 310 for performing the operations of the present application can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer program can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, by utilizing the state information of a computer program to personalize an electronic circuit, such as a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit can execute the computer program, thereby implementing various aspects of the present application.

[0208] Various aspects of the present application are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by a computer program.

[0209] These machine-executable programs 310 can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, such that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These machine-executable programs 310 can also be stored in a machine-readable storage medium 300, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the machine-readable storage medium 300 storing the machine-executable programs 310 comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0210] At this point, those skilled in the art will recognize that, although a number of exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications consistent with the principles of the present invention may be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be understood and deemed to cover all such other variations or modifications.

Claims

1. A method for compiling a high-precision map, comprising: Acquiring high-precision map data in a first format; wherein the high-precision map data in the first format includes at least one of a road facility model, a lane model, and a road model; Receiving a compilation operation for high-precision map data in the first format; Outputting high-precision map data in a second format according to the high-precision map data in the first format; wherein The high-precision map data in the second format is OpenDrive map data, and the OpenDrive map data includes an Objects layer module, a Signals layer module, a Lanes layer module, a Connection layer module, a Junction layer module, and a Road layer module; The road model and the lane model correspond to the Road layer module; The road model and the lane model correspond to the Lanes layer module; The road facility model, the lane model and the road model correspond to the Signals layer module; The road facility model, the lane model, and the road model correspond to the Objects layer module; The road model corresponds to the Junction layer module; The road model corresponds to the Connection layer module.

2. The method for compiling a high-precision map according to claim 1, wherein: The road model includes road reference lines and road attributes, and the lane model includes road connectivity relationships; The road model and the lane model correspond to the Road layer module in the following manner: Reading the road reference line, the road attributes, and the road connectivity relationship; Calculating the road reference line attributes, the road connectivity relationship, and the road attribute geometry according to the read road reference line, the road attributes, and the road connectivity relationship; The calculated results are mapped to the Road layer module.

3. The method for compiling a high-precision map according to claim 1, wherein: The lane model includes a lane centerline, lane lines, lane attributes, and road connectivity, and the road model includes a road reference line; The road model and the lane model correspond to the Lanes layer module in the following manner: Reading the lane centerline, the road reference line, the lane line, the lane attributes, and the lane connectivity relationship; Calculating lane attributes, lane connectivity, lane width, and lane geometry based on the read lane centerline, the road reference line, the lane line, the lane attributes, and the lane connectivity relationship; Map the calculation results to the Lanes layer module.

4. The method for compiling a high-precision map according to claim 1, wherein: The road facility model includes signal lights and traffic signs, the lane model includes lane center lines, and the road model includes road reference lines; The road facility model, the lane model, and the road model correspond to the Signals layer module in the following manner: Reading the lane centerline, the road reference line, the signal light, and the traffic sign; Matching the signal lights and traffic signs with the road according to the read lane center line, the road reference line, the signal lights, and the traffic signs; Map the corresponding results to the Signals layer module.

5. The method for compiling a high-precision map according to claim 1, wherein: The road facility model includes a median strip, the lane model includes a lane centerline, a crosswalk, a stop line, and a guide arrow, and the road model includes a road reference line; The road facility model, the lane model, and the road model correspond to the Objects layer module in the following manner: Reading the lane center line, the road reference line, the isolation strip, the crosswalk, the stop line, and the guide arrow; Matching the road printing surface with the road according to the read lane center line, the road reference line, the isolation strip, the crosswalk, the stop line, and the guide arrow; Map the corresponding results to the Objects layer module.

6. The method for compiling a high-precision map according to claim 1, wherein: The road model includes road reference lines and road attributes; The road model corresponds to the Junction layer module in the following way: Reading the road reference line and the road attributes, and creating an intersection according to the road reference line and the road attributes; Merge the created intersection with the intersection in the high-precision map data in the first format; The merged result is mapped to the Junction layer module.

7. The method for compiling a high-precision map according to claim 1, wherein: The road model includes connectivity relations and road attributes; The road model corresponds to the Connection layer module in the following way: Reading the connectivity relationship and the road attributes, and establishing a correspondence relationship between a predecessor and a successor road according to a correspondence relationship between a road ID in the connectivity relationship and a predecessor and a successor road ID in the corresponding road attributes; Establish the connectivity relationship between the predecessor and the successor in the intersection according to the corresponding relationship between the predecessor and the successor; The established connectivity relationship is mapped to the Connection layer module.

8. The method for compiling a high-precision map according to claim 1, wherein: Before outputting high-precision map data in a second format according to the high-precision map data in the first format, the method further includes: Optimizing the high-precision map data in the first format; The optimizing process of the high-precision map data in the first format includes: Creating intersections for the high-precision map data in the first format; and / or Merging non-intersection roads in the high-precision map data in the first format; and / or Merging the roads within the intersection of the high-precision map data in the first format; and / or Optimize the road reference lines in the high-precision map data in the first format.

9. The method for compiling a high-precision map according to claim 8, wherein: The step of creating an intersection for the high-precision map data in the first format includes: Reading a road reference line list in the high-precision map data in the first format, and filtering out non-intersection data in the road reference line list; Obtaining a successor road list and / or a predecessor road list of the roads in the non-intersection data; Determine the number of successors and / or predecessors in the non-intersection data, and obtain a successor road list and / or a predecessor road list for each successor and / or predecessor; Determining whether the successor road list and / or predecessor road list of each successor and / or predecessor road list contains road data within the intersection; If not included, the successor road list and / or predecessor road list of each successor and / or predecessor is added to the road list in the intersection; The merging of non-intersection roads of the high-precision map data in the first format includes: Reading a road list in the high-precision map data in the first format and filtering out non-intersection roads in the road list; If the successor of a road not in an intersection is a road in an intersection, obtain the road entering the intersection; Obtain the list of predecessor roads at the intersection and merge it into the road list; If the front road in the non-intersection road is the intersection road, obtain the road at the exit of the intersection; Obtain a list of subsequent roads at the exit intersection and merge it into the road list; The merging of roads within the intersection of the high-precision map data in the first format includes: Reading a road list from the high-precision map data in the first format and filtering out non-intersection roads in the road list; If the subsequent part of the road list that has not been filtered out is a non-intersection road, obtain a list of roads in the intersection that will exit the intersection and merge it into the road list; Optimizing the road reference lines in the high-precision map data in the first format includes: Obtaining a lane centerline and a leftmost lane line group in the high-precision map data in the first format; Obtain the predecessor and successor of the lane centerline and the leftmost lane line group; Determining whether the distance and angle difference at the junction exceed a first preset threshold and a second preset threshold, respectively; wherein the junction represents the topological connection between the geometric line of the current lane and the geometric line of the preceding or succeeding lane, the distance refers to the distance between the starting point of the current line and the ending point of the preceding line, or the distance between the ending point of the current line and the starting point of the succeeding line, and the angle refers to the angle between the first line segment of the current line and the last line segment of the preceding line, or the angle between the last line segment of the current line and the first line segment of the succeeding line; If so, perform Bezier curve optimization on the road reference line and update the lane line.

10. The method for compiling a high-precision map according to claim 9, wherein: The step of obtaining the lane centerline and the leftmost lane line group in the high-precision map data in the first format may include: Reading a list of non-intersection lane centerlines in the high-precision map data in the first format, and obtaining a leftmost lane centerline of the same road in the list of non-intersection lane centerlines; If the leftmost lane centerline and the left lane line are not in the same direction, flip the left lane line to obtain the lane centerline and the leftmost lane line group; and / or Reading a list of intersection lane centerlines in the high-precision map data in the first format, and obtaining lane centerlines belonging to the same road in the intersection lane centerline list to form a lane centerline group; Determine whether the lane centerlines in the lane centerline group intersect; If there is an intersection, the lane centerline with the intersection is simplified to obtain the lane centerline and the leftmost lane line group.

11. An electronic device comprising: A memory and a processor, wherein a control program is stored in the memory, and when the control program is executed by the processor, it is used to implement the high-precision map compilation method according to any one of claims 1 to 10.

12. A machine-readable storage medium having a machine-executable program stored thereon, wherein When the machine executable program is executed by a processor, it is used to implement the high-precision map compilation method according to any one of claims 1 to 10.

Images