A method and system for road element conversion based on OpenDRIVE and knowledge graph

Abstract

This invention discloses a method and system for converting road elements based on OpenDRIVE and a knowledge graph. The method includes: constructing a UML diagram of roads in OpenDRIVE format to express the logical structure between road element types in OpenDRIVE format; extracting the attributes, data, and data types contained in each element according to the UML diagram, and constructing a knowledge graph schema layer; establishing a mapping relationship between OpenDRIVE data and the ontology, relations, and attributes of the knowledge graph schema layer, and constructing conversion rules between OpenDRIVE elements and high-precision map knowledge graphs; and converting the elements in OpenDRIVE into elements in the knowledge graph according to the conversion rules between OpenDRIVE elements and high-precision map knowledge graphs. This invention, by parsing OpenDRIVE's description of roads and road-related elements, constructing a knowledge graph schema layer, and establishing a conversion relationship between OpenDRIVE and knowledge graphs, achieves the complementary advantages of OpenDRIVE and knowledge graphs, and improves the efficiency of OpenDRIVE data in the real-time use and parsing of high-precision maps.

Description

Technical Field

[0001] This invention belongs to the field of high-precision map technology, specifically relating to a method and system for converting road elements based on OpenDRIVE and knowledge graph. Background Technology

[0002] With the continuous advancement of autonomous driving technology, the users of maps are gradually shifting from humans to vehicles, leading to increasingly higher requirements for the accuracy and real-time performance of road data. High-precision maps, as an essential condition and key support for autonomous driving, provide detailed and accurate road information, including lanes, traffic signs, and traffic lights, playing a crucial role in the perception, decision-making, planning, and control of autonomous vehicles. Compared to traditional navigation electronic maps, high-precision maps place greater emphasis on the accuracy of roads within the map and their compatibility with autonomous driving systems.

[0003] Currently, there is no unified international standard for high-precision maps. The more widely used and mainstream standards include Navigation Data Standard (NDS) and OpenDRIVE. NDS is a navigation data standard based on an embedded database. It uses a hierarchical and block-based storage method for map data, facilitating map updates. It supports various update methods, such as incremental updates based on regions or databases, or patch updates, improving the reliability, usability, and ease of updating of map data. The OpenDRIVE standard uses an extensible editing language to express road networks, recording the geometric attributes of roads, lanes, and other objects, as well as the characteristics along the road, providing a unified format for road descriptions and reducing the cost of file conversion during industry development and testing. As a standard specification for high-precision maps, OpenDRIVE can express detailed static road network information, providing strong guidance and specificity for autonomous driving. However, due to the massive amount of data in high-precision maps, OpenDRIVE standard data places a significant burden on machines during real-time use and parsing of high-precision maps, wasting time and resources. Furthermore, OpenDRIVE cannot conveniently represent the semantic relationships between various elements, and compared to knowledge graphs, it cannot be combined with rule-based reasoning.

[0004] Knowledge graphs, formally proposed by Google in 2012, are essentially semantic networks. Nodes in the network represent entities or concepts, and edges represent the semantic relationships between entities or concepts. Research areas for knowledge graphs include data storage, knowledge acquisition, and logical reasoning. Applying knowledge graphs to the construction of high-precision map data holds promise for efficient storage and processing of this data. However, current technologies only offer conversion methods between OpenDrive and Apollo data. For example, method 2020106726277 discloses a method for converting map formats based on different protocols, enabling the conversion from OpenDrive protocol format maps to Apollo protocol format maps. However, this method cannot incorporate rule-based knowledge reasoning, thus limiting its application scenarios. Currently, no conversion method between OpenDrive data and knowledge graphs has been discovered. Summary of the Invention

[0005] In view of this, the present invention proposes a road element conversion method and system based on OpenDRIVE and knowledge graph to solve the problem of low efficiency in the real-time use and parsing of OpenDRIVE data in high-precision maps.

[0006] In a first aspect, this invention discloses a method for converting road elements based on OpenDRIVE and knowledge graphs, the method comprising:

[0007] Construct a UML diagram of the OpenDRIVE format road to express the logical structure between OpenDRIVE feature types;

[0008] Extract the attributes, data, and data types contained in each element from the UML diagram to construct the knowledge graph schema layer;

[0009] Establish mapping relationships between OpenDRIVE data and knowledge graph schema layer ontology, relations, and attributes, and construct conversion rules between OpenDRIVE elements and high-precision map knowledge graphs;

[0010] Based on the conversion rules between OpenDRIVE features and high-precision map knowledge graphs, the features in OpenDRIVE are converted into features in the knowledge graph.

[0011] Based on the above technical solutions, preferably, the construction of the UML diagram of the OpenDRIVE format road specifically includes:

[0012] Based on OpenDRIVE's data organization structure, extract the relevant definitions and classifications of roads and related elements from OpenDRIVE;

[0013] Establish semantic relationships between different element types to form a hierarchical UML diagram representing the elements in OpenDRIVE.

[0014] Based on the above technical solutions, preferably, in the UML diagram, road elements include lanes, roads, junctions, objects, railways, and traffic signals.

[0015] Based on the above technical solutions, preferably, the construction of the knowledge graph pattern layer specifically includes:

[0016] The autonomous vehicle ontology is constructed as the core of the entire knowledge graph schema layer, and ontology concepts for road network, road facilities, and background information are constructed respectively.

[0017] The ontology relationships of the schema layer of knowledge graphs are constructed from three perspectives: spatial relationships, temporal relationships, and semantic relationships.

[0018] To construct the ontological attributes of the pattern layer of the knowledge graph for the element attributes that affect vehicle cognition or the use of high-precision maps during autonomous driving.

[0019] Based on the above technical solutions, the preferred ontological concept of the road network includes Road, Intersection, Junction, RoadElement, LaneSection, and Lane.

[0020] The ontological concept of road facilities includes road structure components, road signs, road markings, road safety fixtures, and road equipment.

[0021] Road structure components are part of the road structure, located in the road area or between the road area and the non-road area. They are components that do not belong to the road network, but are closely related to the road network.

[0022] Road safety facilities refer to the facilities that ensure driving safety;

[0023] Road equipment includes traffic lights, streetlights, and measurement devices;

[0024] The ontological concepts for background information include Services and Administrative Areas.

[0025] Based on the above technical solutions, preferably, the establishment of the mapping relationship between OpenDRIVE data and the ontology, relations, and attributes of the knowledge graph pattern layer, and the construction of the conversion rules between OpenDRIVE elements and high-precision map knowledge graphs specifically include:

[0026] Establish a mapping relationship between knowledge graphs and OpenDRIVE data, based on common paths and related elements between them;

[0027] Based on the mapping relationship between OpenDRIVE and knowledge graph, corresponding data structure conversion rules, data type conversion rules, and semantic information conversion rules are established.

[0028] Based on the above technical solutions, preferably, the establishment of a mapping relationship between the knowledge graph and OpenDRIVE data, based on common roads and related elements, specifically includes:

[0029] Based on the roads, lanes, and road-related element subsets mapped to OpenDRIVE in the knowledge graph, establish the correspondence between each element and OpenDRIVE; for road network elements, the correspondence includes one-to-one, one-to-many, and many-to-one relationships.

[0030] A second aspect of the present invention discloses a road element pose conversion system based on OpenDRIVE and knowledge graphs, the system comprising:

[0031] Construct a UML diagram of the OpenDRIVE format road to express the logical structure between OpenDRIVE feature types;

[0032] Based on the road element UML diagram, extract the attributes, data and data types contained in each element, and construct the knowledge graph pattern layer;

[0033] Establish mapping relationships between OpenDRIVE data and knowledge graph schema layer ontology, relations, and attributes, and construct conversion rules between OpenDRIVE elements and high-precision map knowledge graphs;

[0034] Based on the conversion rules between OpenDRIVE features and high-precision map knowledge graphs, the features in OpenDRIVE are converted into features in the knowledge graph.

[0035] A third aspect of the present invention discloses an electronic device comprising: at least one processor, at least one memory, a communication interface, and a bus;

[0036] The processor, memory, and communication interface communicate with each other through the bus.

[0037] The memory stores program instructions that can be executed by the processor, which invokes the program instructions to implement the method as described in the first aspect of the present invention.

[0038] In a fourth aspect, the present invention discloses a computer-readable storage medium storing computer instructions that cause a computer to perform the method described in the first aspect of the present invention.

[0039] The present invention has the following advantages over the prior art:

[0040] 1) Based on the analysis of OpenDRIVE's description of roads and road-related elements, this invention constructs a knowledge graph pattern layer. From spatial and semantic perspectives, it analyzes the mapping patterns and conversion rules between OpenDRIVE and the knowledge graph regarding roads and related elements, and establishes a conversion relationship between OpenDRIVE and the knowledge graph. By leveraging OpenDRIVE's ability to express detailed lane information and the knowledge graph's advantage of acquiring new knowledge through reasoning, this invention achieves complementary advantages between OpenDRIVE and the knowledge graph, improving the efficiency of OpenDRIVE data in the real-time use and parsing of high-precision maps.

[0041] 2) This invention constructs ontology concepts for road network, road facilities, and background information respectively, and constructs the schema layer ontology relations of the knowledge graph from three perspectives: spatial relations, temporal relations, and semantic relations. It constructs the schema layer ontology attributes of the knowledge graph for the element attributes that affect vehicle cognition or the use of high-precision maps during autonomous driving, which can realize a comprehensive and accurate expression of road element knowledge graph and provide data support for format conversion.

[0042] 3) This invention establishes a mapping relationship between roads and related elements common to knowledge graphs and OpenDRIVE data. Based on this mapping relationship, corresponding data structure conversion rules, data type conversion rules, and semantic information conversion rules are established. This enables accurate conversion between OpenDRIVE and knowledge graphs. High-precision maps can achieve the fusion and reuse of data in different formats through the conversion between the two, and can perform reasoning through knowledge graphs, making them more adaptable. This provides technical support for the use of high-precision map data, global and local planning for autonomous driving, and decision-making for autonomous vehicles. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a schematic diagram of a UML diagram showing the hierarchical representation of various elements in OpenDRIVE;

[0045] Figure 2 This is a schematic diagram of the ontology concept of a road network;

[0046] Figure 3 This is a schematic diagram illustrating the basic concept of road infrastructure.

[0047] Figure 4 This is a schematic diagram of the concept of the road structure components.

[0048] Figure 5 This is a schematic diagram of the concept of road safety facilities.

[0049] Figure 6 This is a schematic diagram of the road equipment itself.

[0050] Figure 7 This is a schematic diagram of the ontology concept for background information;

[0051] Figure 8 A schematic diagram of the administrative division concept;

[0052] Figure 9 A schematic diagram of the mapping of road network elements in the OpenDRIVE and knowledge graph pattern layers;

[0053] Figure 10 This is a schematic diagram of an example road;

[0054] Figure 11 This is a schematic diagram of an example road in OpenDRIVE format;

[0055] Figure 12 for Figure 11 The UML diagram corresponding to the example road;

[0056] Figure 13 An example diagram illustrating the mapping relationship between OpenDRIVE and the knowledge graph schema layer;

[0057] Figure 14 This is a partial example diagram of the converted knowledge graph;

[0058] Figure 15 This is an example diagram of the knowledge graph showing the relationship between lanes before and after the transformation. Detailed Implementation

[0059] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0060] In the process of autonomous driving, it is necessary not only to perform global road planning based on high-precision map data and lane-level local road planning based on dynamic road data, but also to perform rule-based reasoning on the safety, legality, and rationality of autonomous vehicles based on prior rules such as laws and regulations, thereby assisting autonomous vehicles in dynamic real-time decision-making. Because OpenDRIVE and knowledge graphs have different data storage and organizational structures, they also differ in data representation and construction. OpenDRIVE can express detailed lane information, while knowledge graphs can acquire new knowledge through reasoning to assist autonomous vehicles in making driving decisions. Therefore, to achieve the complementary advantages of OpenDRIVE and knowledge graphs, it is necessary to construct conversion rules between them regarding roads and road-related elements.

[0061] This invention proposes a road element conversion method based on OpenDRIVE and knowledge graph, the method comprising:

[0062] S1. Construct a UML diagram of OpenDRIVE format roads to express the logical structure between OpenDRIVE format road feature types.

[0063] This invention uses ASAM OpenDRIVE version 1.7.0 as an example to construct a UML diagram of roads and road-related element types, thereby expressing the logical structure between element types.

[0064] First, based on OpenDRIVE's data organization structure, extract the relevant definitions and classifications of roads and related elements in OpenDRIVE. In OpenDRIVE, road elements generally include lanes, roads, junctions, objects, railroads, and signals.

[0065] Then, semantic relationships between the various element types are established, forming a UML diagram representing the hierarchical relationships between the elements in OpenDRIVE.

[0066] like Figure 1The diagram shown is a schematic representation of the hierarchical structure of elements in OpenDRIVE. The hierarchical representation of elements in OpenDRIVE allows for a direct observation of the elements and their attribute data at different levels.

[0067] S2. Extract the attributes, data, and data types contained in each element from the UML diagram to construct the knowledge graph pattern layer.

[0068] Step S2 specifically includes the following sub-steps:

[0069] S21. Extract the attributes, data, and data types contained in each feature from the UML diagram.

[0070] S22. Construct the ontology of autonomous vehicles as the core of the entire knowledge graph pattern layer, and construct ontology concepts for road network, road facilities, and background information respectively.

[0071] Based on the relevant analysis of road networks, an ontology concept for road networks is constructed, such as... Figure 2 The diagram shows the ontological concept of a road network, including roads, intersections, junctions, road elements, lane sections, and lanes. Figure 2 In this context, Road represents a road between two intersections; Intersection represents an intersection, used to connect roads and lanes; Junction represents a point of intersection within a road, which is part of an intersection. When an intersection consists of multiple road junctions, Junction represents an intersection attached to an Intersection; RoadElement represents a road element, which is a collection of lanes in the same driving direction; LaneSection represents a lane segment, which contains lanes in the same driving direction on a road. New lane segments are created when the number of lanes changes; Lane represents a lane, which is the most basic unit in a road.

[0072] The concept of road infrastructure is mainly divided into five parts, such as... Figure 3 The diagram shows the main concept of road facilities, which include road structure components, road signs, road markings, road safety facilities, and road equipment.

[0073] Road structural components are part of the road structure. These elements are located within road areas or between road and non-road areas. They are not part of the road network but are closely related to it, such as curbs and manhole covers. Figure 4 The diagram shows a conceptual representation of the road structure components, including tram tracks, curbs, bollards, gutters, traffic islands, railway level crossings, and manholes.

[0074] Road safety facilities refer to facilities that ensure driving safety, such as... Figure 5 The diagram shows a conceptual illustration of road safety facilities, including guardrails, delineators, piers, speed bumps, and guardrails.

[0075] Road infrastructure primarily serves to assist and guide vehicle driving, such as... Figure 6 The diagram shown is a conceptual illustration of road equipment, which includes traffic lights, streetlights, and measurement devices.

[0076] In autonomous driving, roads and related information alone are insufficient for navigation tasks; background information is needed to assist in vehicle navigation and global path planning. Therefore, this invention also constructs an ontology concept for background information. Figure 7 This is a schematic diagram of the background information ontology concept, including services and administrative areas. Services include Points of Interest (POI) information on the map, such as city centers and tourist attractions, which assist vehicles in determining destinations and routes during autonomous driving and global planning. Since local laws and regulations may differ between cities, such as vehicle restriction rules, the traffic rules of the administrative area in which the vehicle is located must be considered during driving. In this embodiment, administrative areas are divided into four levels. Figure 8 This is a schematic diagram of the administrative division concept, including the country, provincial-level administrative regions (Order1 Area), prefecture-level administrative regions (Order2 Area), county-level administrative regions (Order3 Area), and township-level administrative regions (Order8 Area).

[0077] S23. Construct the ontology relations of the schema layer of the knowledge graph from three perspectives: spatial relations, temporal relations, and semantic relations.

[0078] Among them, spatial relationships represent topological relationships and directional relationships. Table 1 shows the spatial relationship table, Table 2 shows the temporal relationship table, and Table 3 shows the semantic relationship table.

[0079] Table 1 Spatial Relationship Table

[0080]

[0081] Table 2 Time Relationship Table

[0082] Relationship Name Relationship Chinese name isBeforeOf Before... isAfterOf After... Occurs When it happens Start start End Finish During During... Overlapstime Intersect with...

[0083] Table 3 Semantic Relationship Table

[0084] Relationship Name Relationship Chinese name isPartof Whole-part relationship IsKindof Father-son relationship isOn On ... isActOn Effect on hasPredecessorOf Previous lane hasSuccessorOf Next lane isIntentTo intention isStraightOf straight isTurnRightOf Turn right isTurnLeftOf Turn left isLeftOf On the left side isRightOf on the right

[0085] Based on the spatial relationship table, temporal relationship table, and semantic relationship table above, analyze the relationships between the ontological concepts of road network, road facilities, and background information, and construct the schema layer ontological relationships of the knowledge graph.

[0086] S24. Construct the ontological attributes of the pattern layer of the knowledge graph for the element attributes that affect vehicle cognition or the use of high-precision maps during autonomous driving.

[0087] Based on the ontological relationships of the schema layer of the knowledge graph, the element attributes that affect vehicle cognition or the use of high-precision maps during autonomous driving are obtained, and the schema layer ontological attributes of the knowledge graph are constructed.

[0088] S3. Establish the mapping relationship between OpenDRIVE data and the ontology, relations, and attributes of the knowledge graph pattern layer, and construct the conversion rules between OpenDRIVE elements and high-precision map knowledge graph.

[0089] Step S3 specifically includes the following sub-steps:

[0090] S31. Establish a mapping relationship between the common roads and related elements in the knowledge graph and OpenDRIVE data.

[0091] Specifically, based on the roads, lanes, and road-related element subsets mapped to OpenDRIVE in the knowledge graph, a correspondence between each element and OpenDRIVE is established. For road network elements, the correspondence includes one-to-one, one-to-many, and many-to-one relationships. For example... Figure 9 The diagram shows the mapping of road network elements in the OpenDRIVE and knowledge graph pattern layers.

[0092] A one-to-one relationship includes Junction-{Intersection, Junction}. In OpenDRIVE, the Junction element represents the connection between roads connected to an intersection and virtual roads within the intersection, as well as the connection between vehicles on two roads. In the knowledge graph, the Junction element represents an intersection within a road, and is part of an intersection. When an intersection consists of multiple road junctions, Junction represents an intersection attached to an Intersection, and Intersection represents an intersection connecting roads and lanes.

[0093] One-to-many relationships include Lane Section-LaneSection and Lane-Lane. In OpenDRIVE, a LaneSection element represents a lane segment, including two roads traveling in opposite directions and the lanes within those roads. In a knowledge graph, a LaneSection represents a lane segment traveling in the same direction, which is half the content represented by a Lane Section in OpenDRIVE. When the lane is a one-way street, both represent the same content. In this case, the same Lane Section may correspond to one or more LaneSections.

[0094] Many-to-one relationships include Road-Road and Road-RoadElement. In OpenDRIVE, a Road element represents a road segment. If road attributes change, such as the number of lanes, at least two Road elements are needed to represent the road. In a knowledge graph, a Road represents a road between two intersections, and a RoadElement represents a set of lanes traveling in the same direction. Therefore, multiple Road elements in OpenDRIVE may correspond to one Road element or two RoadElement elements in the knowledge graph.

[0095] S32. Based on the above mapping relationship between OpenDRIVE and knowledge graph, the conversion rules between OpenDRIVE elements and knowledge graph.

[0096] Specifically, to address issues such as different data structures, inconsistent data types, and semantic mismatches that may arise during conversion, corresponding conversion rules are established, including data structure conversion rules, data type conversion rules, and semantic information conversion rules.

[0097] S4. Based on the conversion rules between OpenDRIVE features and high-precision map knowledge graph, convert the features in OpenDRIVE into features in the knowledge graph.

[0098] Addressing the differences in high-precision map data storage standards between the current OpenDRIVE standard and knowledge graphs, as well as the limitations of storing and utilizing high-precision map data in a single manner, this paper establishes conversion rules between OpenDRIVE elements and high-precision map knowledge graphs by analyzing OpenDRIVE roads and road-related elements. This enables the conversion of high-precision map data between the OpenDRIVE standard and knowledge graph formats. High-precision maps can achieve the fusion and reuse of data from different formats through this conversion, and can also perform reasoning through knowledge graphs to assist autonomous vehicles in making driving decisions.

[0099] This invention can provide technical support for the use of high-precision map data, global and local planning for autonomous driving, and decision-making for autonomous vehicles.

[0100] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0101] Figure 10 This demonstrates an example road that can be described using the OpenDRIVE standard. Figure 11 To and Figure 10 The corresponding example road in OpenDRIVE format. In OpenDRIVE, road data is primarily represented by the `Road` keyword, which includes road information, lane information, road objects, and road sign information. For example... Figure 11 As shown, a Road is composed of Lane Sections, which represent a segment of the road. New Lane Sections are created when the number of lanes in the road changes. Lanes are represented using Lanes. Within each Lane Section, a center lane is defined as a position reference, with an ID of 0. The IDs of other lanes decrease sequentially along the road reference line. The relationships between lanes are represented using Predecessor and Successor.

[0102] Step 1: Parse the OpenDRIVE format road data and construct a UML diagram of the road elements.

[0103] like Figure 12 As shown Figure 11 The UML diagram corresponding to the instance road in OpenDRIVE format.

[0104] Step 2: Construction of the knowledge graph pattern layer.

[0105] (1) Construct the ontology concept according to steps S21 to S22.

[0106] (2) Construct ontology relationships.

[0107] The ontology relationships are constructed based on the UML diagram of the example road. A semantic relationship `isPartOf` exists between the Road and Lane Section within this example road, representing the overall part relationship between the Road and Lane Section. Different Lanes within the same Lane Section have left-right adjacency relationships, as shown in Tables 1-3, using `isLeftOf` and `isRightOf` to represent left and right adjacency. Within different Lane Sections, lanes have semantic sequential connections, using `haspredecessorof` and `hassuccessorof` to represent the connection between the preceding and following lanes.

[0108] (3) Construct the ontology attributes.

[0109] The ontology attributes of roads, lane segments, and lanes in the instance road are constructed. Table 4 shows an example of the ontology attributes of the instance road.

[0110] Table 4 Example of Road Body Attributes

[0111]

[0112]

[0113] Step 3: Establish the mapping relationship between OpenDRIVE and the road elements in the knowledge graph pattern layer.

[0114] Based on the UML diagram of the example roads and the constructed knowledge graph schema layer, establish the mapping relationship between OpenDRIVE and the knowledge graph schema layer, such as... Figure 13 The diagram shows an example of the mapping relationship between OpenDRIVE and the knowledge graph schema layer. Transformation rules for feature instances are constructed based on the established mapping relationship.

[0115] In OpenDRIVE, roads use absolute IDs, while lanes use relative IDs. A road has a center lane with a relative ID of 0 and no width; this lane marks the division between two opposing driving directions. Centered on this lane and referencing the road direction, lane IDs increment to the left and decrement to the right (e.g., ...). Figure 11 (Lane 0, -1, -2 in the example). LaneSection has an attribute indicating the starting position, used to distinguish different Lane Sections within the same road, such as... Figure 11In the example road, Lane Section 02 is located at the beginning of Road 02, so its starting position attribute is 0.0. Lane Section 03's starting position is 50 meters from the starting point of Road 02, so its starting position attribute is 50.0. During the conversion, all relative IDs are converted to absolute IDs by combining the ID of the element's parent element with the element's ID to represent the absolute ID. For example, in a knowledge graph, the relationship between Road and RoadElement is 1:2. Since the ID of a Road is an absolute ID, the IDs of two RoadElements are 0_-1 and 0_1. Suppose a Lane Section within a RoadElement with ID 0_-1 has a starting point of 0.0, and to distinguish it, "lanesection" is added to the beginning of the ID, so the ID of this Lane Section is lanesection_0_1_0.0. The absolute ID of a lane is composed of the lane's relative ID, the road ID, and the starting point of the lane segment. To make it easier to distinguish, "lane" is added to the beginning of the ID, such as the ID of the first road on the right in the LaneSection above, which is lane_0_0.0_-1.

[0116] In OpenDRIVE, the correlation between lanes is used... <predecessor>and <successor>During the transformation, knowledge graph edges are used to connect lanes and lane nodes. The edges are named `haspredecessorof` and `hassuccessorof`. Semantic connections are also built between different lanes within the same lane segment. For example, in Lane Section 02, Lane-2 and Lane-1 have an `isRightOf` relationship. In the knowledge graph, the nodes representing Lane-2 and Lane-1 need to be connected using the name `isRightOf`, and the connection should point to Lane-1.

[0117] Step 4: Based on the conversion rules, convert the elements in OpenDRIVE into elements in the knowledge graph.

[0118] Figure 14 This is a partial example diagram of the converted knowledge graph. Figure 15 This is an example diagram of the knowledge graph showing the relationship between lanes before and after the transformation.

[0119] This invention performs semantic mapping and data structure transformation on roads and road-related elements using OpenDRIVE and knowledge graphs, enabling the fusion of multi-source data for high-precision maps and cross-standard sharing.

[0120] Corresponding to the above method embodiments, the present invention also proposes a road element conversion system based on OpenDRIVE and knowledge graph, the system comprising:

[0121] Construct a UML diagram of the OpenDRIVE format road to express the logical structure between OpenDRIVE feature types;

[0122] Based on the road element UML diagram, extract the attributes, data and data types contained in each element, and construct the knowledge graph pattern layer;

[0123] Establish mapping relationships between OpenDRIVE data and knowledge graph schema layer ontology, relations, and attributes, and construct conversion rules between OpenDRIVE elements and high-precision map knowledge graphs;

[0124] Based on the conversion rules between OpenDRIVE features and high-precision map knowledge graphs, the features in OpenDRIVE are converted into features in the knowledge graph.

[0125] The above system embodiments and method embodiments are one-to-one correspondences. For a brief description of the system embodiments, please refer to the method embodiments.

[0126] The present invention also discloses an electronic device, comprising: at least one processor, at least one memory, a communication interface, and a bus; wherein the processor, memory, and communication interface communicate with each other through the bus; the memory stores program instructions executable by the processor, and the processor calls the program instructions to implement the aforementioned method of the present invention.

[0127] The present invention also discloses a computer-readable storage medium that stores computer instructions, which cause the computer to implement all or part of the steps of the method described in the embodiments of the present invention. The storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0128] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, meaning they can be distributed across multiple network units. Those skilled in the art can select some or all of the modules to achieve the purpose of this embodiment without any inventive effort, based on actual needs.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.< / successor> < / predecessor>

Claims

1. A road element conversion method based on OpenDRIVE and knowledge graph, characterized in that, The method includes: Constructing a UML diagram of OpenDRIVE format roads to represent the logical structure between OpenDRIVE format road feature types; specifically, constructing the UML diagram of OpenDRIVE format roads includes: Based on OpenDRIVE's data organization structure, extract the relevant definitions and classifications of roads and related elements from OpenDRIVE; Establish semantic relationships between different element types to form a hierarchical UML diagram representing the elements in OpenDRIVE; Extract the attributes, data, and data types contained in each element from the UML diagram to construct the knowledge graph schema layer; Establish mapping relationships between OpenDRIVE data and the ontology, relations, and attributes of the knowledge graph schema layer; construct conversion rules between OpenDRIVE features and high-precision map knowledge graphs, specifically including: To establish a mapping relationship between knowledge graphs and OpenDRIVE data, based on the common paths and related elements in knowledge graphs and OpenDRIVE data; Based on the mapping relationship between OpenDRIVE and knowledge graph, establish corresponding data structure conversion rules, data type conversion rules, and semantic information conversion rules; Based on the conversion rules between OpenDRIVE features and high-precision map knowledge graphs, the features in OpenDRIVE are converted into features in the knowledge graph.

2. The road element conversion method based on OpenDRIVE and knowledge graph according to claim 1, characterized in that, In the UML diagram, road elements include lanes, roads, junctions, objects, railroads, and signals.

3. The road element conversion method based on OpenDRIVE and knowledge graph according to claim 2, characterized in that, The construction of the knowledge graph pattern layer specifically includes: The autonomous vehicle ontology is constructed as the core of the entire knowledge graph schema layer, and ontology concepts for road network, road facilities, and background information are constructed respectively. The ontology relationships of the schema layer of knowledge graphs are constructed from three perspectives: spatial relationships, temporal relationships, and semantic relationships. To construct the ontological attributes of the pattern layer of the knowledge graph for the element attributes that affect vehicle cognition or the use of high-precision maps during autonomous driving.

4. The road element conversion method based on OpenDRIVE and knowledge graph according to claim 3, characterized in that, The ontological concepts of a road network include Road, Intersection, Junction, RoadElement, LaneSection, and Lane; The ontological concept of road facilities includes road structure components, road signs, road markings, road safety fixtures, and road equipment. Road structure components are part of the road structure, located in the road area or between the road area and the non-road area. They are components that do not belong to the road network, but are closely related to the road network. Road safety facilities refer to the facilities that ensure driving safety; Road equipment includes traffic lights, streetlights, and measurement devices; The ontological concepts for background information include Services and Administrative Areas.

5. The road element conversion method based on OpenDRIVE and knowledge graph according to claim 1, characterized in that, The specific steps for establishing a mapping relationship between knowledge graphs and OpenDRIVE data, based on common paths and related elements, include: Based on roads, lanes, and road-related element subsets in the knowledge graph, establish the correspondence between each element and OpenDRIVE; for road network elements, the correspondence includes one-to-one, one-to-many, and many-to-one relationships.

6. A road element conversion system based on OpenDRIVE and knowledge graph, characterized in that, The system includes: Constructing a UML diagram of an OpenDRIVE format road to represent the logical structure between OpenDRIVE feature types; specifically, constructing the UML diagram of an OpenDRIVE format road includes: Based on OpenDRIVE's data organization structure, extract the relevant definitions and classifications of roads and related elements from OpenDRIVE; Establish semantic relationships between different element types to form a hierarchical UML diagram representing the elements in OpenDRIVE; Based on the road element UML diagram, extract the attributes, data and data types contained in each element, and construct the knowledge graph pattern layer; Establish mapping relationships between OpenDRIVE data and the ontology, relations, and attributes of the knowledge graph schema layer; construct conversion rules between OpenDRIVE features and high-precision map knowledge graphs, specifically including: To establish a mapping relationship between knowledge graphs and OpenDRIVE data, based on the common paths and related elements in knowledge graphs and OpenDRIVE data; Based on the mapping relationship between OpenDRIVE and knowledge graph, establish corresponding data structure conversion rules, data type conversion rules, and semantic information conversion rules; Based on the conversion rules between OpenDRIVE features and high-precision map knowledge graphs, the features in OpenDRIVE are converted into features in the knowledge graph.

7. An electronic device, characterized in that, include: At least one processor, at least one memory, a communication interface, and a bus; The processor, memory, and communication interface communicate with each other through the bus. The memory stores program instructions that can be executed by the processor, which invokes the program instructions to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause the computer to perform the method as described in any one of claims 1 to 5.

Images