A high-precision map data representation method, related method and device

Abstract

The application discloses a high-precision map data representation method and related method and device. The method comprises the following steps: dividing a target road into multiple road segments, and assigning a road segment identifier to each road segment; determining an initial lane geometry according to a center line of each lane of each road segment; setting multiple nodes in the initial lane geometry according to a preset distance to obtain a section lane geometry; obtaining multiple rounds of driving collection data of the lane; determining an instantaneous speed and a direction corresponding to each node of the section lane geometry and an average speed between each two adjacent nodes based on the multiple rounds of driving collection data and the road segment identifier to obtain lane map data representation of the lane corresponding to the road segment identifier. The speed in the lane map data representation obtained by the method can reflect factors such as plane curvature, longitudinal section slope, transverse section super elevation in actual driving, so that the map data representation can reflect actual driving conditions, and high-precision automatic driving is realized.

Description

Technical Field

[0001] This invention relates to a high-precision map data representation method, related methods, and apparatus. Background Technology

[0002] The rapid development of autonomous driving technology has placed extremely high demands on map accuracy. Simultaneously, autonomous driving also relies on high-precision elevation data to accurately perceive road slopes and terrain changes, thereby optimizing vehicle driving strategies. In existing technologies, these requirements are met through the fusion of high-precision maps and real-time sensor data. High-precision maps (HD Maps) provide detailed road information, including lane-level accuracy, road signs, traffic light locations, and 3D elevation data. These maps are generated by fusing data from multiple sensors such as LiDAR, cameras, and GPS, and are continuously updated to stay current. High-precision maps enable autonomous vehicles to anticipate upcoming road conditions, allowing for better planning and execution of driving tasks. Furthermore, autonomous vehicles are equipped with LiDAR, radar, cameras, and ultrasonic sensors. The real-time data from these sensors is fused together to form a comprehensive environmental perception model, ensuring that the vehicle can accurately perceive its surroundings under various environmental and weather conditions, including detecting obstacles, recognizing pedestrians, and other vehicles. Summary of the Invention

[0003] To obtain high-precision map data representations for autonomous driving, this invention provides a high-precision map data representation method, related methods, and apparatus.

[0004] In a first aspect, embodiments of the present invention provide a high-precision map data representation method, which may include:

[0005] The target road is divided into multiple road segments, and each road segment is assigned a segment identifier;

[0006] For each lane of each road segment, the initial lane geometry is determined based on the centerline of the lane.

[0007] Multiple nodes are set at preset distances in the initial lane geometry to obtain the segment lane geometry;

[0008] Acquire multiple trips of traffic data for the lane;

[0009] Based on the data collected from multiple trips and the road segment identifiers, the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes are determined, thereby obtaining the lane map data representation of the lane corresponding to the road segment identifier.

[0010] In one or more optional embodiments of this application, the step of determining the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes based on the multi-trip driving data and the road segment identifier, to obtain the lane map data representation of the lane corresponding to the road segment identifier, includes:

[0011] The data collected from the multiple trips are clustered to obtain the instantaneous speed and direction of each node, the total travel time, and the time difference between each two adjacent nodes.

[0012] The total average speed is calculated based on the total driving time and the total length of the lane geometry of the section.

[0013] The average speed between each pair of adjacent nodes is calculated based on the total average speed, the time difference between each pair of adjacent nodes, and the preset distance.

[0014] Based on the instantaneous speed and direction of each node and the average speed between every two adjacent nodes, the lane map data representation of the lane corresponding to the road segment identifier is determined with the location of the road segment identifier as the origin.

[0015] In one or more optional embodiments of this application, calculating the average speed between two adjacent nodes based on the total average speed, the time difference between two adjacent nodes, and the preset distance includes:

[0016] For every two adjacent nodes, determine whether the time difference between the two adjacent nodes is valid data;

[0017] If so, the average speed between the two adjacent nodes is calculated based on the time difference between the two adjacent nodes and the preset distance;

[0018] If not, then the total average speed is taken as the average speed between the two adjacent nodes.

[0019] In one or more optional embodiments of this application, dividing the target road into multiple road segments and assigning a segment identifier to each road segment includes:

[0020] The target road is divided into multiple road segments, and a road segment marker is set at the beginning of each road segment.

[0021] In one or more optional embodiments of this application, dividing the target road into multiple road segments and setting a road segment marker at the beginning of each road segment includes:

[0022] The target road is divided into multiple road segments, and a ground barcode is placed at the beginning of each road segment;

[0023] Obtain the GPS location information corresponding to the starting position of the road segment;

[0024] The ground barcode and the GPS location information are associated as the road segment identifier.

[0025] In one or more optional embodiments of this application, the following further includes:

[0026] The width and speed limit information of the lanes are obtained and added to the lane map data representation.

[0027] In one or more optional embodiments of this application, after obtaining the lane map data representation, the method further includes:

[0028] Set up multiple driving scenarios and the corresponding speed reduction ratio for each driving scenario;

[0029] For each driving scenario, all speeds in the lane map data representation are multiplied by the speed reduction ratio corresponding to the driving scenario to obtain a new lane map data representation corresponding to the driving scenario.

[0030] In a second aspect, embodiments of the present invention provide a method for autonomous driving of vehicles, wherein lane map data representation obtained using the high-precision map data representation method described in the first aspect may include:

[0031] Acquire real-time vehicle image data from the vehicle to obtain road segment markings and the current lane;

[0032] Obtain the lane map data representation of the current lane corresponding to the road segment identifier;

[0033] Based on the lane map data, the real-time speed, real-time acceleration, and direction of real-time acceleration are determined.

[0034] Thirdly, embodiments of the present invention provide a high-precision map data representation device, which may include:

[0035] The first identification module is used to divide the target road into multiple road segments and assign a segment identification to each road segment;

[0036] The first processing module is used to determine the initial lane geometry for each lane of each road segment based on the centerline of the lane.

[0037] The first segmentation module is used to set multiple nodes at preset distances in the initial lane geometry to obtain the segment lane geometry.

[0038] The first acquisition module is used to acquire multiple traffic data collected from the lane.

[0039] The second processing module is used to determine the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes based on the data collected from the multiple trips and the road segment identification, so as to obtain the lane map data representation of the lane corresponding to the road segment identification.

[0040] Fourthly, embodiments of the present invention provide a vehicle autonomous driving device, which may include:

[0041] The identification module is used to acquire real-time vehicle image data from the vehicle to obtain road segment identification and the current lane.

[0042] The data acquisition module is used to acquire lane map data representation of the current lane corresponding to the road segment identifier;

[0043] An autonomous driving module is used to determine real-time speed, real-time acceleration, and the direction of real-time acceleration based on the lane map data representation.

[0044] Fifthly, embodiments of the present invention provide a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the high-precision map data representation method and / or the vehicle autonomous driving method described above.

[0045] Sixthly, embodiments of the present invention provide a computer program product, including a computer program / instruction that, when executed by a processor, implements the high-precision map data representation method as described above, and / or, an autonomous driving method for vehicles.

[0046] In a seventh aspect, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the high-precision map data representation method as described above, and / or, an autonomous driving method for vehicles.

[0047] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0048] This invention provides a high-precision map data representation method. The method divides a target road into multiple road segments and sets multiple nodes at preset distances for each lane of each road segment to obtain the segment's lane geometry. Based on acquired data from multiple driving trips, instantaneous velocity, direction, and average velocity are assigned to each node and every two adjacent nodes in the segment's lane geometry, resulting in a lane map data representation. The vector velocity data in the lane map data representation obtained by this method can reflect comprehensive factors such as planar curvature, longitudinal slope, and cross-sectional superelevation in actual driving, making the lane map data representation not just a simple path representation but also reflecting the actual driving conditions on the road, thereby achieving high-precision and timely guidance during autonomous driving. Furthermore, this lane map data representation only associates road segment identifiers and does not use latitude and longitude coordinate data or actual elevation data of road shape points to represent the road. Therefore, it is not easy to deduce specific geographical location and elevation information from this lane map data representation, effectively protecting geographical location privacy and complying with national mandatory standards.

[0049] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0050] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0051] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0052] Figure 1 This is a schematic diagram illustrating the steps of the high-precision map data representation method provided in an embodiment of the present invention;

[0053] Figure 2 A schematic diagram of the initial lane geometry provided in an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of the section lane geometry provided in an embodiment of the present invention;

[0055] Figure 4 This is a schematic diagram of lane map data representation provided in an embodiment of the present invention;

[0056] Figure 5 This is a schematic diagram of the steps of the vehicle autonomous driving method provided in an embodiment of the present invention;

[0057] Figure 6A schematic diagram of the structure of the high-precision map data representation device provided in the embodiments of this application;

[0058] Figure 7 This is a schematic diagram of the structure of the vehicle autonomous driving device provided in the embodiments of this application. Detailed Implementation

[0059] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0060] The inventors discovered that in existing technologies, autonomous driving technology heavily relies on high-precision maps; however, existing nationally mandated standard maps differ significantly from the requirements of autonomous driving in terms of accuracy and data content. Based on this, the inventors conducted further research and development, resulting in this invention, which provides a high-precision map data representation method, related methods, and apparatus.

[0061] Example 1

[0062] Embodiment 1 of the present invention provides a high-precision map data representation method, referring to... Figure 1 As shown, the method may include the following steps S101-S105:

[0063] S101: Divide the target road into multiple road segments and assign a segment identifier to each road segment.

[0064] S102: For each lane of each road segment, determine the initial lane geometry based on the lane's centerline.

[0065] S103: Set multiple nodes in the initial lane geometry according to a preset distance to obtain the segment lane geometry.

[0066] S104: Acquire multiple traffic data for each lane.

[0067] S105: Based on data collected from multiple trips and road segment markings, determine the instantaneous speed and direction of each node of the lane geometry in the section and the average speed between every two adjacent nodes to obtain the lane map data representation of the lane corresponding to the road segment marking.

[0068] This invention provides a high-precision map data representation method. This method divides a target road into multiple road segments and sets multiple nodes at preset distances for each lane in each road segment to obtain the segment's lane geometry. Based on acquired data from multiple driving trips, instantaneous velocity, direction, and average velocity are assigned to each node and every two adjacent nodes in the segment's lane geometry to obtain the map data representation. The vector velocity data in the lane map data representation obtained by this method can reflect comprehensive factors such as planar curvature, longitudinal slope, and cross-sectional superelevation in actual driving, making the lane map data representation not just a simple path representation but also reflecting the actual driving conditions on the road, thereby achieving high-precision and timely guidance during autonomous driving. Furthermore, this lane map data representation only associates road segment identifiers and does not use latitude and longitude coordinate data or actual elevation data of road shape points to represent the road. Therefore, it is not easy to deduce specific geographical location and elevation information from this lane map data representation, effectively protecting geographical location privacy and complying with national mandatory standards.

[0069] In step S101 above, the target road is divided into multiple road segments, and each road segment is assigned a segment identifier. Specifically, the target road can be divided into multiple road segments, a ground barcode is set at the starting position of each road segment, the GPS location information corresponding to the starting position of the road segment is obtained, and the ground barcode and GPS location information are associated as the segment identifier.

[0070] In one specific embodiment, a certain highway is taken as the target road and divided into multiple road segments. The first road segment is the first 100 meters of the highway. A ground barcode is set at the starting point of the highway. At the same time, the GPS location signal corresponding to the starting point is obtained. The GPS location signal may have errors, but it can roughly locate the vicinity of the starting point of the highway. The ground barcode is associated with the GPS location information to obtain the road segment identifier.

[0071] One method of dividing the target road into multiple road segments is to use objects such as signs, markings, 100-meter markers, and milestones along the road to divide it into segments. For example, if the target road is a highway, it can be divided into segments every 100 meters based on the 100-meter markers along the highway.

[0072] In this embodiment, the ground barcode can be a barcode or a QR code; the form of the ground barcode is not limited here. However, the ground barcode needs to be unique within a certain range. For example, within a city, a ground barcode can only point to a specific road segment. Since the ground barcode is associated with GPS location information, when identifying the ground barcode, the specific road segment can be located by combining it with real-time GPS signals.

[0073] In step S102 above, for each lane of each road segment, the lane centerline can be used to determine the initial lane geometry. (Refer to...) Figure 2 The road segment shown is an example of a road segment with three lanes, namely the first lane, the second lane and the third lane from top to bottom. In the latter half of the road segment, the third lane merges into the second lane. Taking the third lane as an example, the geometry of the center line of the third lane in the figure is determined, that is, the red line in the figure, to represent the initial lane geometry of the third lane.

[0074] In step S103 above, starting from the initial point of the initial lane geometry, nodes are set at preset distances within the initial lane geometry to obtain the segment lane geometry. For example, according to... Figure 2 The schematic diagram of the segment lane geometry obtained from the initial lane geometry is shown below. Figure 3 As shown, the red line on the third lane represents the geometry of the lane section, and the blue dots above it are the nodes.

[0075] The nodes are set according to a preset distance, meaning the distance between any two adjacent nodes is equal to the preset distance. It's important to note that this preset distance can be configured appropriately when setting up the nodes to ensure that nodes are placed at all inflection points on the lane geometry of the section, facilitating the subsequent recording of the instantaneous speed and direction at these inflection points.

[0076] In step S104 above, the data collection vehicle can travel multiple times in one lane to obtain multiple trips of driving data. This driving data is mainly lane-level navigation data, containing the data required by this method, such as the timestamp, trajectory points, instantaneous speed, and direction of instantaneous speed at each time point.

[0077] This process also requires preprocessing, which mainly includes data cleaning and timestamp conversion. Timestamp conversion refers to changing each timestamp data to the difference between that timestamp and the start timestamp of that trip; that is, converting the timestamp data into the travel duration at each time point.

[0078] In step S105 above, based on data collected from multiple trips and road segment markings, the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes are determined, thus obtaining the lane map data representation of the lane corresponding to the road segment marking. Specifically, this includes the following steps S1051-S1054:

[0079] S1051: Cluster the data collected from multiple trips to obtain the instantaneous speed and direction of each node, the total travel time, and the time difference between each pair of adjacent nodes.

[0080] In this embodiment of the application, there is no single way to cluster the data collected from multiple trips to obtain the instantaneous speed and direction of each node, the total travel time, and the time difference between each two adjacent nodes. Two methods are provided below.

[0081] The first method involves clustering multiple driving data points, aggregating similar driving trajectories and data such as instantaneous speed at each trajectory point to form an aggregated driving path, so that each point on the aggregated driving path has a corresponding timestamp, instantaneous speed, and direction.

[0082] Next, by comparing the aggregated driving path and the segment lane geometry, find the point on the aggregated driving path that is closest to each node on the segment lane geometry, assign all the data of that point to the node, and determine the timestamp, instantaneous speed and direction of each node.

[0083] At this point, the timestamp of the last node of the lane geometry in the section is the total travel time.

[0084] Finally, by subtracting the timestamps between any two adjacent nodes, the time difference can be determined.

[0085] The second approach involves clustering multiple traffic data points for each node in the lane geometry of a section, using the designated node as the center point, to obtain the timestamp, instantaneous speed, and direction corresponding to that node.

[0086] The method for obtaining the total travel time and the time difference between every two adjacent nodes is the same as the first method, and will not be repeated here.

[0087] Those skilled in the art can select appropriate clustering algorithms based on the existing situation and technology, such as commonly used clustering algorithms: K-means, DBSCAN (Density-Based Spatial Clustering of Applications with Noise), etc., and no specific limitation is required in the embodiments of this application.

[0088] S1052: The total average speed is calculated based on the total travel time and the total length of the lane geometry in the section.

[0089] Specifically, the total average speed can be obtained by dividing the total length of the lane geometry of the section by the total travel time.

[0090] S1053: Calculate the average speed between two adjacent nodes based on the total average speed, the time difference between two adjacent nodes, and the preset distance.

[0091] In one specific embodiment, step S1053 specifically includes the following steps S10531-S10533:

[0092] S10531: For every two adjacent nodes, determine whether the time difference between the two adjacent nodes is valid data: if yes, proceed to step S10532; if no, proceed to step S10533.

[0093] S10532: Calculate the average speed between two adjacent nodes based on the time difference and the preset distance between them.

[0094] Specifically, the average speed can be obtained by dividing the preset distance between two adjacent nodes by the time difference.

[0095] S10533: Use the total average speed as the average speed between two adjacent nodes.

[0096] In this embodiment, when acquiring multiple driving data for a lane in step S104, data cleaning is required to remove abnormal or erroneous data. This may cause errors or even null values ​​in the time difference between two adjacent nodes, rendering the time difference invalid. In such cases, step S10533 needs to be executed for special processing to maintain data consistency and continuity, preventing individual abnormal data from affecting the overall driving analysis results. This improves the reliability and accuracy of the navigation system, providing more stable and accurate driving guidance.

[0097] S1054: Based on the instantaneous speed and direction of each node and the average speed between every two adjacent nodes, determine the lane map data representation of the lane corresponding to the road segment identifier, with the location of the road segment identifier as the origin.

[0098] Specifically, a Cartesian coordinate system can be established with the location of the road segment marker as the origin. Data such as the instantaneous speed and direction of each node and the average speed between each two adjacent nodes can be transformed into this coordinate system to obtain the lane map data representation and establish the correlation between the road segment marker and the lane map data representation.

[0099] In one specific embodiment, according to Figure 3 The lane map data obtained from the lane geometry of the shown section is represented as follows: Figure 4As shown, with the road segment marker as the origin, each node has a corresponding instantaneous speed and direction, and there is an average speed between every two adjacent nodes. The instantaneous speed is exemplarily shown at the first inflection point of the third lane in the figure. This value is a vector, represented by the magenta arrow in the diagram. The diagram also exemplifies the average speed before the first inflection point in the third lane. .

[0100] In this embodiment of the application, the driving data obtained in step S104 can also include the width information and speed limit information of each lane. This information can be added to the lane map data representation to better constrain the vehicle in autonomous driving, so that the vehicle does not exceed the lane range or exceed the speed limit.

[0101] In this embodiment, after determining the lane map data representation, adjustments can be made to the lane map data representation for different driving scenarios. Specifically, this includes setting multiple driving scenarios and a corresponding speed reduction ratio for each scenario; for each driving scenario, multiplying all instantaneous and average speeds in the lane map data representation by the corresponding speed reduction ratio; and simultaneously applying kinematic formulas to change the direction of the instantaneous speeds to obtain a new lane map data representation corresponding to the driving scenario. These driving scenarios include, but are not limited to, adverse weather conditions such as rain and fog, further improving the performance of autonomous driving.

[0102] In this embodiment, after determining the lane map data representation, adjustments can be made to the lane map data representation for different driving scenarios. For example, for severe weather such as rain or fog, the lane map data representation can be adjusted by combining multiple factors such as the deceleration ratio, changes in speed direction, and vehicle dynamics models to more accurately simulate actual conditions and ensure driving safety. The method for adjusting the lane map data representation according to the driving scenario specifically includes the following steps:

[0103] The first step is to identify the driving scenario by using sensor data (such as cameras, radar, lidar, etc.) to identify the current driving scenario, including weather conditions (rain, fog, etc.), lighting conditions, road conditions (slippery, icy, etc.) and traffic flow.

[0104] The second step is to preset the speed reduction ratio. Based on historical data or expert experience, a speed reduction ratio (such as 0.8) is set for different driving scenarios to adjust the instantaneous speed and average speed in the lane map data representation.

[0105] The third step is to establish a vehicle dynamics model that can describe the vehicle's acceleration, deceleration, and braking performance under different road conditions. For example, in rainy or snowy weather, the vehicle's acceleration and deceleration capabilities decrease, and its braking distance increases. The model can also include a stability assessment to consider the vehicle's stability on slippery surfaces, such as the risks of sideslip and loss of control. Appropriate stability assessment mechanisms need to be incorporated into the model to predict and prevent these risks.

[0106] The fourth step involves combining the deceleration ratio and the vehicle dynamics model, and applying kinematic formulas (such as acceleration formulas, speed update formulas, etc.) to determine the change in instantaneous velocity direction in the lane map data representation.

[0107] Example 2

[0108] Based on the same inventive concept, this invention also provides a vehicle autonomous driving method, which uses lane map data representation obtained by the high-precision map data representation method described in Embodiment 1 above, and refers to... Figure 5 As shown, the method includes:

[0109] S201: Obtain real-time vehicle image data from the vehicle to obtain road segment identification and the current lane.

[0110] S202: Obtain the lane map data representation of the current lane corresponding to the road segment identifier.

[0111] S203: Determine the real-time speed, real-time acceleration, and direction of real-time acceleration based on lane map data.

[0112] In this embodiment, based on the acquired real-time vehicle image data, the current lane is located by identifying the number of lane lines and the position of the edge lines on the road. Then, road segment markings on the side of the road are identified, and the corresponding lane map data representation is obtained based on the current lane. The lane map data representation includes the instantaneous speed and direction of multiple nodes, as well as the average speed between every two nodes.

[0113] In the process of autonomous driving, the instantaneous speed, direction and average speed at each time point in the lane map data are used as the real-time speed. The real-time acceleration is calculated from the speed before and after the current moment. The direction of the real-time acceleration is calculated by combining the centripetal force calculation formula and the real-time speed. The real-time power output of the vehicle is determined based on Newton's second law to achieve autonomous driving.

[0114] Example 3

[0115] Based on the same inventive concept, embodiments of the present invention also provide a high-precision map data representation device, referring to... Figure 6 As shown, the device includes:

[0116] The first identification module 101 is used to divide the target road into multiple road segments and assign a segment identification to each road segment;

[0117] The first processing module 102 is used to determine the initial lane geometry for each lane of each road segment based on the centerline of the lane.

[0118] The first segmentation module 103 is used to set multiple nodes at preset distances in the initial lane geometry to obtain the segment lane geometry.

[0119] The first acquisition module 104 is used to acquire multiple traffic data collected from the lane.

[0120] The second processing module 105 is used to determine the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes based on the multi-trip driving data and the road segment identification, so as to obtain the lane map data representation of the lane corresponding to the road segment identification.

[0121] Example 4

[0122] Based on the same inventive concept, embodiments of the present invention also provide a vehicle autonomous driving device, referring to... Figure 7 As shown, the device includes:

[0123] The identification determination module 201 is used to acquire real-time vehicle image data from the vehicle to obtain road segment identification and the current lane;

[0124] Data acquisition module 202 is used to acquire lane map data representation of the current lane corresponding to the road segment identifier;

[0125] The autonomous driving module 203 is used to determine the real-time speed, real-time acceleration, and direction of real-time acceleration based on the lane map data representation.

[0126] In this embodiment of the invention, the lane map data representation of the current lane corresponding to the road segment identifier acquired by the data acquisition module 202 is obtained through the high-precision map data representation device in Embodiment 3. The high-precision map data representation device specifically includes:

[0127] The first identification module 101 is used to divide the target road into multiple road segments and assign a segment identification to each road segment;

[0128] The first processing module 102 is used to determine the initial lane geometry for each lane of each road segment based on the centerline of the lane.

[0129] The first segmentation module 103 is used to set multiple nodes at preset distances in the initial lane geometry to obtain the segment lane geometry.

[0130] The first acquisition module 104 is used to acquire multiple traffic data collected from the lane.

[0131] The second processing module 105 is used to determine the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes based on the multi-trip driving data and the road segment identification, so as to obtain the lane map data representation of the lane corresponding to the road segment identification.

[0132] Example 5

[0133] Based on the same inventive concept, embodiments of the present invention also provide a computer-readable storage medium storing a computer program / instruction thereon, which, when executed by a processor, implements the high-precision map data representation method as described in Embodiment 1 above, and / or the vehicle autonomous driving method as described in Embodiment 2 above.

[0134] Example 6

[0135] Based on the same inventive concept, embodiments of the present invention also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the high-precision map data representation method as described in Embodiment 1 above, and / or the vehicle autonomous driving method as described in Embodiment 2 above.

[0136] Example 7

[0137] Based on the same inventive concept, embodiments of the present invention also provide a computer device, including a memory, a processor, and a computer program stored in the memory. When the processor executes the computer program, it implements the high-precision map data representation method as described in Embodiment 1 above, and / or the vehicle autonomous driving method as described in Embodiment 2 above.

[0138] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0139] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A high-precision map data representation method characterized by, include: The target road is divided into multiple road segments, and each road segment is assigned a segment identifier; For each lane of each road segment, the initial lane geometry is determined based on the centerline of the lane. Multiple nodes are set at preset distances in the initial lane geometry to obtain the segment lane geometry; Acquire multiple trips of traffic data for the lane; Based on the data collected from multiple trips and the road segment identifiers, the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes are determined, thereby obtaining the lane map data representation of the lane corresponding to the road segment identifier.

2. The method of claim 1, wherein, Based on the data collected from multiple trips and the road segment identifiers, the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes are determined to obtain the lane map data representation of the lane corresponding to the road segment identifier, including: The data collected from the multiple trips are clustered to obtain the instantaneous speed and direction of each node, the total travel time, and the time difference between each two adjacent nodes. The total average speed is calculated based on the total driving time and the total length of the lane geometry of the section. The average speed between each pair of adjacent nodes is calculated based on the total average speed, the time difference between each pair of adjacent nodes, and the preset distance. Based on the instantaneous speed and direction of each node and the average speed between every two adjacent nodes, the lane map data representation of the lane corresponding to the road segment identifier is determined with the location of the road segment identifier as the origin.

3. The method of claim 2, wherein, The step of calculating the average speed between two adjacent nodes based on the total average speed, the time difference between two adjacent nodes, and the preset distance includes: For every two adjacent nodes, determine whether the time difference between the two adjacent nodes is valid data; If so, the average speed between the two adjacent nodes is calculated based on the time difference between the two adjacent nodes and the preset distance; If not, then the total average speed is taken as the average speed between the two adjacent nodes.

4. The method of claim 1, wherein, The process of dividing the target road into multiple road segments and assigning a segment identifier to each road segment includes: The target road is divided into multiple road segments, and a road segment marker is set at the beginning of each road segment.

5. The method according to claim 4, characterized in that, The process of dividing the target road into multiple road segments and setting a road segment marker at the beginning of each road segment includes: The target road is divided into multiple road segments, and a ground barcode is placed at the beginning of each road segment; Obtain the GPS location information corresponding to the starting position of the road segment; The ground barcode and the GPS location information are associated as the road segment identifier.

6. The method according to claim 1, characterized in that, Also includes: The width and speed limit information of the lanes are obtained and added to the lane map data representation.

7. The method according to claim 1, characterized in that, After obtaining the lane map data representation, the following is also included: Set up multiple driving scenarios and the corresponding speed reduction ratio for each driving scenario; For each driving scenario, all speeds in the lane map data representation are multiplied by the speed reduction ratio corresponding to the driving scenario to obtain a new lane map data representation corresponding to the driving scenario.

8. A method for autonomous driving of a vehicle, characterized in that, Lane map data representation obtained using the high-precision map data representation method according to any one of claims 1-7, the method comprising: Acquire real-time vehicle image data from the vehicle to obtain road segment markings and the current lane; Obtain the lane map data representation of the current lane corresponding to the road segment identifier; Based on the lane map data, the real-time speed, real-time acceleration, and direction of real-time acceleration are determined.

9. A high-precision map data representation device, characterized in that, include: The first identification module is used to divide the target road into multiple road segments and assign a segment identification to each road segment; The first processing module is used to determine the initial lane geometry for each lane of each road segment based on the centerline of the lane. The first segmentation module is used to set multiple nodes at preset distances in the initial lane geometry to obtain the segment lane geometry. The first acquisition module is used to acquire multiple traffic data collected from the lane. The second processing module is used to determine the instantaneous speed and direction corresponding to each node of the lane geometry of the section and the average speed between every two adjacent nodes based on the data collected from the multiple trips and the road segment identification, so as to obtain the lane map data representation of the lane corresponding to the road segment identification.

10. An automatic driving device for vehicles, characterized in that, include: The identification module is used to acquire real-time vehicle image data from the vehicle to obtain road segment identification and the current lane. The data acquisition module is used to acquire lane map data representation of the current lane corresponding to the road segment identifier; An autonomous driving module is used to determine real-time speed, real-time acceleration, and the direction of real-time acceleration based on the lane map data representation.

11. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the high-precision map data representation method according to any one of claims 1-7, and / or the vehicle autonomous driving method according to claim 8.

12. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the high-precision map data representation method according to any one of claims 1-7, and / or the vehicle autonomous driving method according to claim 8.

13. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the high-precision map data representation method according to any one of claims 1-7, and / or the vehicle autonomous driving method according to claim 8.

Images