Marking method, driving path planning and decision method, device and vehicle

Abstract

The application relates to the technical field of map data, in particular to a marking method, a driving path planning and decision method, a device and a vehicle. The marking method comprises the following steps: acquiring a global planning path of a vehicle on a standard map; mapping the global planning path to a high-precision map, and identifying first lane data of a local planning path in a target range of the high-precision map and second lane data related to the local planning path; and marking a driving space type on the high-precision map according to the first lane data and / or the second lane data, wherein the driving space type comprises at least one of drivable space, recommended driving space, non-recommended driving space and alternative driving space. Thus, the problem that map data in the prior art cannot meet the local planning and decision requirements of complex scenes is solved.

Description

Technical Field

[0001] This application relates to the field of map data technology, and in particular to a marking method, a driving route planning and decision-making method, an apparatus, and a vehicle. Background Technology

[0002] In local planning for autonomous driving, the conventional approach to map data search is as follows: Obtain path data from the starting point to the destination based on a standard map plan. This data consists of a set of nodes arranged in a specific order. Node data includes the index for connecting to the next node, the sequence of road segment shape points, the length of the road segment, and the attribute level of the road segment. Based on the path data, index the node sequence from the starting point to the destination and accumulate the distance to a preset distance. Then, based on the node information within this distance, sequentially search for lane information and the connections between lanes from high-precision map data using node segment IDs, shape points, road attributes, etc.

[0003] Because highways have simple topological relationships, the planned paths are simply a sequence of nodes connected one-to-one, eliminating the need to construct path association information. However, as the collection scope of high-precision map data expands and its application scenarios extend beyond highways, there is an urgent need for more complex scenarios within cities. Therefore, based on standard maps... Figure 1 Linking paths with a single path cannot meet the needs of local planning and decision-making in complex scenarios. Summary of the Invention

[0004] This application provides a marking method, a driving path planning and decision-making method, an apparatus, and a vehicle to solve the problems in related technologies where map data cannot meet the local planning and decision-making needs of complex scenarios.

[0005] The first aspect of this application provides a method for marking driving space types on a map, comprising the following steps: obtaining a global planned path of a vehicle on a standard map; mapping the global planned path to a high-precision map, and identifying first lane data of a local planned path and second lane data related to the local planned path within a target range of the high-precision map; marking driving space types on the high-precision map based on the first lane data and / or the second lane data, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space, and alternative driving space.

[0006] Optionally, the driving space type is marked on the high-precision map based on the first lane data and / or the second lane data, including: marking drivable space, suggested driving space and non-suggested driving space on the high-precision map based on the first lane data; and marking alternative driving spaces on the high-precision map data based on the first lane data and the second data.

[0007] Optionally, based on the first lane data, drivable space, suggested drivable space, and non-suggested drivable space are marked on the high-precision map, including: identifying target road segments with entrances and exits on the local planned path in the first lane data, wherein the target road segment includes at least one lane; marking lanes with entrances and exits in the target road segment as non-suggested drivable space; marking lanes without entrances and exits in the target road segment as suggested drivable space; and marking road segments other than the target road segments on the planned path as drivable space.

[0008] Optionally, marking alternative driving spaces on a high-precision map based on the first lane data and the second lane data includes: creating a spatial network within a target area of ​​the high-precision map, wherein the spatial network includes multiple grids; determining the grid parameters of a target grid among the multiple grids based on the first lane data and the second lane data; and marking alternative driving spaces on the high-precision map based on the grid parameters of the target grid.

[0009] Optionally, the paths in the first lane data and the second lane data are divided into multiple road segments, and the grid parameters include at least one road segment, with each road segment having multiple road segment attributes; alternative driving spaces are marked on the high-precision map according to the grid parameters of the target grid, including: calculating alternative values ​​for each target grid based on the relevant data of each road segment attribute, where the alternative value represents a reference value for the road segment within the target grid to meet vehicle passage conditions; constructing a first matrix based on the alternative values ​​of all target grids for each road segment attribute; obtaining a second matrix by weighting the first matrix for each road segment attribute and the weights of each road segment attribute; and using alternative paths planned based on the second matrix as alternative driving spaces.

[0010] Optionally, the data for multiple road segment attributes includes at least one of road segment direction, road segment width, road segment class, and number of lanes.

[0011] A second aspect of this application provides a driving route planning and decision-making method, comprising the following steps: obtaining a global planned route for a vehicle on a standard map; mapping the global planned route to a high-precision map, and identifying first lane data and second lane data related to the local planned route within a target area of ​​the high-precision map; marking driving space types on the high-precision map based on the first lane data and / or the second lane data, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space, and alternative driving space; and performing local planning and decision-making based on the driving space type and the high-precision map.

[0012] A third aspect of this application provides a map driving space type marking device, comprising: a first acquisition module for acquiring a global planned path of a vehicle on a standard map; a first mapping module for mapping the global planned path to a high-precision map and identifying first lane data and second lane data related to the local planned path within a target range of the high-precision map; and a first marking module for marking driving space types on the high-precision map data based on the first lane data and / or the second lane data of the planned path, and performing path planning on the high-precision map based on the driving space types, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space, and alternative driving space.

[0013] Optionally, the first marking module is further configured to: mark driving space types on a high-precision map based on the first lane data and / or the second lane data, including: marking drivable space, suggested driving space and non-suggested driving space on the high-precision map based on the first lane data; and marking alternative driving spaces on the high-precision map data based on the first lane data and the second data.

[0014] Optionally, the first marking module is further configured to: identify target road segments with entrances and exits on the local planned path in the first lane data, wherein the target road segment includes at least one lane; mark lanes with entrances and exits in the target road segment as non-recommended driving spaces; mark lanes without entrances and exits in the target road segment as recommended driving spaces; and mark road segments other than the target road segment on the planned path as driving spaces.

[0015] Optionally, the first marking module is further configured to: create a spatial network within the target area of ​​the high-precision map, wherein the spatial network includes multiple grids; determine the grid parameters of the target grid among the multiple grids based on the first lane data and the second lane data; and mark alternative driving spaces on the high-precision map based on the grid parameters of the target grid.

[0016] Optionally, the lane data of the road segments within the target grid includes actual data of at least one road segment attribute, the grid parameters include lane data of the road segments within the target grid, and each road segment has multiple road segment attributes; the first marking module is further used to: calculate the candidate value of each target grid based on the relevant data of each road segment attribute, wherein the candidate value represents the reference value of the road segment within the target grid that meets the vehicle passage conditions; construct a first matrix based on the candidate values ​​of all target grids for each road segment attribute; obtain a second matrix by weighting the first matrix of each road segment attribute and the weights of each road segment attribute; and use the candidate path planned based on the second matrix as the candidate driving space.

[0017] Optionally, the data for multiple road segment attributes includes at least one of road segment direction, road segment width, road segment class, and number of lanes.

[0018] A fourth aspect of this application provides a driving route planning and decision-making device, comprising: a second acquisition module for acquiring a global planned route of a vehicle on a standard map; a second mapping module for mapping the global planned route to a high-precision map and identifying first lane data and second lane data related to the local planned route within a target range of the high-precision map; a second marking module for marking driving space types on the high-precision map data based on the first lane data and / or the second lane data of the planned route, and performing route planning on the high-precision map based on the driving space types, wherein the driving space types include at least one of drivable space, suggested driving space, non-suggested driving space, and alternative driving space; and a planning module for performing local planning and decision-making based on the driving space types and the high-precision map.

[0019] A fifth aspect of this application provides a vehicle, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement the map driving space type marking method or the driving path planning and decision method as described in the above embodiments.

[0020] Therefore, this application has at least the following beneficial effects:

[0021] This application's embodiments can map globally planned paths from a standard map to a high-precision map. Based on planned and unplanned paths, driving space types are marked on the high-precision map. Based on these driving space types, path planning can be directly re-performed on the high-precision map, eliminating the need to re-plan paths using the standard map and re-map them to the high-precision map. This allows for faster path re-planning, facilitating better, faster, and smarter path planning during local planning and decision-making processes, meeting the needs of local planning and decision-making in various complex scenarios. Therefore, it solves the technical problem in related technologies where map data cannot meet the needs of local planning and decision-making in complex scenarios.

[0022] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0023] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0024] Figure 1 This is a flowchart of a map driving space type marking method provided according to an embodiment of this application;

[0025] Figure 2 This is a schematic diagram illustrating the classification of driving space types according to embodiments of this application;

[0026] Figure 3 This is a schematic diagram of road segment mapping to a spatial network according to an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of a first matrix provided according to an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of a roundabout scene provided according to an embodiment of this application;

[0029] Figure 6 This is a schematic diagram of an alternative driving space provided according to one embodiment of this application;

[0030] Figure 7 This is a flowchart of the driving path planning and decision-making method provided according to the embodiments of this application;

[0031] Figure 8 This is a flowchart of a driving path planning and decision-making method according to an embodiment of this application;

[0032] Figure 9 A block diagram illustrating a map driving space type marking device according to an embodiment of this application;

[0033] Figure 10 This is a block diagram of a driving path planning and decision-making device provided according to an embodiment of this application;

[0034] Figure 11 This is a structural schematic diagram of a vehicle provided according to an embodiment of this application. Detailed Implementation

[0035] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0036] The marking method, driving path planning and decision-making method, apparatus, and vehicle of embodiments of this application are described below with reference to the accompanying drawings. Addressing the issues mentioned in the background art regarding autonomous driving scenarios with simple road topology (high-speed or urban expressways), and the inability to meet the needs of local planning and decision-making applications in complex urban scenarios where high-precision map data for a single planned path cannot, the significant impact of roads associated with the planned path on local planning and decision-making, and the large volume of high-precision map data that cannot be loaded over a wide area, this application provides a method for marking driving space types on a map. This method can search and load non-planned paths associated with planned paths, expand the field of view of high-precision map data based on global planning, and mark the type of driving space to reduce local planning and decision-making calculations. Thus, it solves the problems in related technologies where map data cannot meet the local planning and decision-making needs of complex scenarios.

[0037] Specifically, Figure 1 This is a schematic diagram illustrating the process of marking driving space types on a map, as provided in an embodiment of this application.

[0038] like Figure 1 As shown, the method for marking driving space types on this map includes the following steps:

[0039] In step S101, the global planned path of the vehicle on the standard map is obtained.

[0040] The standard map is a regular, ordinary map. It can be a service or data. Regardless of the form, as long as the global planning path can be obtained, it is acceptable. In other words, the function of the standard map is to obtain the global planning path.

[0041] In step S102, the global planning path is mapped to the high-precision map, and the first lane data of the local planning path and the second lane data related to the local planning path are identified within the target range of the high-precision map.

[0042] The target range can be set according to specific circumstances. For example, with the vehicle as the center, it can consider high-precision map data within 50m to the left, right, and front of the vehicle. The target range moves with the vehicle and is derived from the GPS update frequency. Lane data includes lane attribute information and shape points. The planned path is the driving path planned by the user on the standard map. The path related to the local planned path can be called the non-planned path, which is the driving path not considered when the standard map planned the driving path. The target path is the non-planned path related to the local planned path.

[0043] Because high-precision maps have higher accuracy, richer content, and a larger data volume than standard maps, the data volume of high-precision maps is far greater than that of standard maps. Therefore, in this embodiment, after obtaining the global planning path from the standard map, the global planning path can be mapped onto the high-precision map. This further identifies the first lane data and the second lane data related to the local planning path within the target area of ​​the high-precision map. This expands the path planning range of the map, allowing non-planned paths related to the planned path to be considered in subsequent vehicle local planning and driving decisions.

[0044] In step S103, driving space types are marked on a high-precision map based on the first lane data and / or the second lane data, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space, and alternative driving space.

[0045] It should be noted that high-definition maps contain far more data than standard maps. Therefore, in related technologies, when a vehicle needs to replan its route, it is usually impossible to perform route planning directly on the high-definition map. Instead, it is often necessary to return to the standard map for route planning and then map the replanned route back to the high-definition map. This process of replanning and remapping based on the standard map significantly reduces the efficiency of route planning.

[0046] Therefore, this application embodiment marks the driving space type on the high-precision map based on the planned path and the unplanned path, so that the local path can be re-planned directly on the high-precision map based on the driving space type. This eliminates the need to use the standard map for path planning again and re-map to the high-precision map. Only one mapping of the planned path is needed, which can make the path planning more quickly. This achieves efficient and fast path planning on the high-precision map, enabling local planning and decision-making to make better, faster and smarter driving trajectories, and improving the response speed of local planning and decision-making.

[0047] In this embodiment of the application, marking driving space types on high-precision map data based on first lane data and / or second lane data includes: marking drivable space, suggested driving space and non-suggested driving space on high-precision map based on first lane data; and marking alternative driving spaces on high-precision map data based on first lane data and second lane data.

[0048] It is understood that, in the embodiments of this application, the type of driving space can be marked on a high-precision map based on the first lane data of the local planning path, and alternative driving spaces can be marked on a high-precision map based on the first and second lane data.

[0049] In this embodiment of the application, marking drivable space, suggested drivable space and non-suggested drivable space on a high-precision map based on the first lane data includes: identifying target road segments with entrances and exits on the local planned path in the first lane data, wherein the target road segment includes at least one lane; marking lanes with entrances and exits in the target road segment as non-suggested drivable space; marking lanes without entrances and exits in the target road segment as suggested drivable space; and marking road segments other than the target road segments on the planned path as drivable space.

[0050] The target road segment is a road segment with entrances and exits, such as... Figure 2 The black box in the image represents the target road segment.

[0051] It is understood that the embodiments of this application can identify target road segments with entrances and exits in the first lane data. In lanes with entrances and exits, since vehicles can exit at the exit and vehicles can enter at the entrance, lanes with entrances and exits are marked as non-recommended driving spaces, and lanes in the target road segment without entrances and exits are marked as recommended driving spaces, that is, lanes adjacent to the lanes to which the non-recommended driving spaces belong. The remaining road segments are driving spaces.

[0052] For example, such as Figure 2 As shown, in a straight-ahead scenario, this embodiment of the application can query the lane information and mark the driving space type by identifying the road segment relationships between the front and back ends of the data nodes in the first lane node sequence that have entrances and exits. The specific marking method is as follows:

[0053] (1) Non-recommended driving space, such as Figure 2 In the S2 region, the locally planned path follows the map backbone network AEGB. N1-N2-N3 are the locally planned route nodes, and L1, L2, and L3 are the locally planned route lanes. Taking the N2 locally planned route node as an example, E is the front end of N2, and G is the back end of N2. At the front end of N2, besides the N1 node segment of the locally planned route, there is an FE segment as an incoming edge, which affects the driving speed of the solid line portion of lane L3 within the N1 and N2 nodes. Figure 2 The space shown is S2, so space S2 is marked as a non-recommended driving space.

[0054] (2) Driving space, such as Figure 2 In the S1 region, taking the local planning path node N1 as an example, its downstream node E is affected by the S2 non-recommended driving space. That is, when there is already non-recommended driving space in the current node, the remaining parts are marked as driving space when there is driving association.

[0055] (3) Recommended driving space, such as Figure 2 In the S3 area, the space that is not affected by other road sections is marked as the recommended driving space.

[0056] In this embodiment of the application, marking alternative driving spaces on a high-precision map based on first lane data and second lane data includes: creating a spatial network within a target area of ​​the high-precision map, wherein the spatial network includes multiple grids; determining the grid parameters of a target grid among the multiple grids based on the first lane data and second lane data; and marking alternative driving spaces on the high-precision map based on the grid parameters of the target grid.

[0057] The target range of the high-precision map can be set according to specific circumstances. For example, a spatial grid can be created within a range of 100 meters in all directions around the vehicle. The spatial grid includes multiple grids, such as dividing the area into grids every 5 meters in all directions.

[0058] It is understood that the embodiments of this application can create a spatial grid within the target area of ​​the high-precision map, and further determine the grid parameters of the target grid in multiple grids based on the first lane data of the local planned path and the second lane data of the target path, and then mark the alternative driving space on the high-precision map according to the grid parameters of the target grid, wherein the target grid includes the local planned path and the non-planned path related to the local planned path.

[0059] In this embodiment, the path in the first lane data and the second lane data is divided into multiple road segments. The grid parameters include at least one road segment, and each road segment has multiple road segment attributes. Candidate driving spaces are marked on a high-precision map based on the grid parameters of the target grid. This includes: calculating the candidate value for each target grid based on the relevant data of each road segment attribute, where the candidate value represents a reference value for the road segment within the target grid to meet vehicle traffic conditions; constructing a first matrix based on the candidate values ​​of all target grids for each road segment attribute; obtaining a second matrix by weighting the first matrix of each road segment attribute with the weights of each road segment attribute; and using the candidate path planned based on the second matrix as the candidate driving space.

[0060] The road segment attributes include at least one of the following: road segment direction, road segment width, road segment level, and number of lanes; the path in the first lane data and the second lane data is divided into multiple road segments, and the grid parameters include at least one road segment.

[0061] It is understood that, in the embodiments of this application, the alternative values ​​of each target grid can be calculated based on the relevant data of each road segment attribute, and a first matrix can be constructed based on the alternative values ​​of all target grids. The first matrix of each road segment attribute and the weight of each road segment attribute are weighted to obtain a second matrix. The alternative paths planned in the second matrix are used as alternative driving spaces. The alternative values ​​represent the reference values ​​of the road segments in the target grid that meet the conditions for vehicle passage.

[0062] The formula for calculating the second matrix is ​​as follows: l represents the number of road segment attributes, k represents the weight of the road segment attribute, and M represents the number of road segment attributes. nmF(r) is the first matrix of road segment attributes, and F(r) is the weight matrix of multiple road segment attributes, i.e., the second matrix.

[0063] It should be noted that since a spatial grid is created within the target area of ​​the high-precision map, the spatial grid does not contain the grid containing the first lane data of the local planning path and the second lane data of the target planning path. The alternative value can be set to a target value, such as 0, 0.5 or 1, which is a value that will not affect subsequent planning.

[0064] As one possible approach, the relevant data for each road segment attribute can be the actual data and reference data for each road segment attribute. First, the candidate value for each target grid is calculated based on the actual data and reference data for each road segment attribute. Second, a first matrix is ​​constructed based on the candidate values ​​of all target grids for each road segment attribute. Then, a second matrix is ​​obtained by weighting the first matrix and the weights of each road segment attribute. Finally, the candidate paths planned based on the second matrix serve as the candidate driving space.

[0065] For example, taking road segment width as a road segment attribute, the reference data can be understood as the standard width. When the width is greater than or equal to the standard width, vehicles can ensure safety when passing quickly. When the width is less than the standard width, vehicles will usually reduce their speed to ensure safety, making it impossible to pass quickly and safely. Therefore, the embodiments of this application can determine the difference or ratio between the actual data and the parameter data, determine the candidate value relationship table for the road segment width through calibration, and determine the candidate value of the target grid through the candidate value relationship table. It can be specifically set that a larger candidate value indicates a wider road segment, or a smaller candidate value indicates a wider road segment, without specific limitations.

[0066] The road segment attributes, such as road segment direction, road segment level, and number of lanes, are similar to those of the road segment width mentioned above. A corresponding alternative value relationship table can also be defined. For example, the reference data is the vehicle travel direction, and the actual data is the actual road segment direction. Based on the vehicle travel direction, the actual road segment direction, and the corresponding alternative value relationship table, alternative values ​​for the target grid are determined. It can be specifically set that a larger alternative value indicates the same or identical direction, or that a smaller alternative value indicates the same or identical direction; no specific limitation is imposed. Another example is that the reference data is the reference road segment level, which can be an expressway, and the actual data is the actual road segment level within the target grid, such as a secondary arterial road. Based on the reference... The candidate values ​​for the target grid are determined by a table relating road segment level, actual road segment level, and corresponding alternative values. Specifically, a larger alternative value can indicate a higher road segment level, or a smaller alternative value can indicate a higher road segment level; no specific limitation is imposed. For example, the reference data might be the number of reference lanes (e.g., two or three lanes as a parameter), while the actual data is the actual number of lanes that vehicles can drive on. The candidate values ​​for the target grid are determined based on the table relating the number of lanes as a parameter, the actual number of lanes, and corresponding alternative values. Again, a larger alternative value can indicate a greater number of drivable lanes, or a smaller alternative value can indicate a greater number of drivable lanes; no specific limitation is imposed.

[0067] Based on the above method, a first matrix can be calculated for each road segment attribute. A second matrix is ​​obtained by weighting each road segment attribute according to its weight. The alternative routes planned based on the second matrix are then used as alternative driving spaces. For example, if the weight of road segment width is k1, the first matrix is ​​M1; the weight of road segment direction is k2, the first matrix is ​​M2; the weight of road segment level is k3, the first matrix is ​​M3; and the weight of the number of lanes is k4, the first matrix is ​​M4. Then, the second matrix = k1*M1 + k2*M2 + k3*M3 + k4*M4. The maximum value in the second matrix can be used as the alternative driving space. The weights of road segment attributes such as road segment direction, road segment level, road segment width, and number of lanes can be specifically set or calibrated, without specific limitations.

[0068] For example, using such Figure 2 In the straight-ahead scenario, alternative driving spaces are marked, and the local planning and decision-making methods based on high-precision maps are as follows:

[0069] 1. Mapping spatial grids to road segments, such as... Figure 3 As shown.

[0070] (1) Calculate the spatial grid based on the current absolute position of the vehicle and the preset spatial network parameters, wherein the preset spatial grid parameters are the target range based on the high-precision map.

[0071] (2) Search for lane data of the road backbone in the high-precision map within the spatial grid range.

[0072] (3) Identify each target grid in the spatial grid and obtain the index row and column of the spatial grid.

[0073] (4) Figure 4 As shown, an n*m matrix is ​​constructed based on the grid index rows and columns n*m. Each grid cell contains alternative values ​​for different road segment attributes. The grid parameters of the target grid are determined based on the lane data. For other grid cells that do not have alternative values ​​for corresponding road segments, 1 is filled in.

[0074] (5) Multiple n*m ​​matrices are obtained based on various road segment attributes.

[0075] 2. Calculation of weight matrix of multiple road segment attributes.

[0076] The final weight matrix of a road segment is obtained by multiplying an n*m matrix of multiple road segment attributes by a mode factor k and summing the results. The formula is as follows: l represents the number of road segment attributes, k represents the weight of the road segment attribute, and M represents the number of road segment attributes. nm F(r) is the first matrix of road segment attributes, and F(r) is the weight matrix of multiple road segment attributes for the current road segment, i.e., the second matrix.

[0077] For example, in a high-precision map, a target area 65m long and 40m wide is defined with the vehicle as the center. A spatial network is created, with a grid defined at 5m intervals, thus constructing a 13*8 grid. A 13*8 matrix is ​​then constructed. For example, the weight of the road segment width is k1, the first matrix is ​​M1; the weight of the road segment direction is k2, the first matrix is ​​M2; the weight of the road segment level is k3, the first matrix is ​​M3; and the weight of the number of lanes is k4, the first matrix is ​​M4. Then, the second matrix = k1*M1 + k2*M2 + k3*M3 + k4*M4. The maximum value in the second matrix can be used as the alternative driving space.

[0078] For example, the globally planned path from the standard map can be mapped to a high-definition map, and the driving space type can be marked on the high-definition map, such as... Figure 2In the straight-ahead scenario shown, the local planned path is AEGB. The local planned path is a part of the global planned path. According to the driving space type marking method in the above embodiment, GHD can be marked as a candidate driving space. If there is congestion or traffic accident on the GB path, the vehicle can make local planning and decisions based on the actual situation and the marked driving space type. The replanned driving path is AEGHD. The vehicle is controlled to continue driving from the candidate driving space GHD, thereby avoiding the congested sections on the original planned path. Moreover, the local planning and decision can be realized directly using the driving space type marked on the high-precision map. There is no need to replan the path using the standard map and then map it to the high-precision map. This effectively reduces the time of repeated mapping, improves the response speed of local path planning and decision, and enhances the intelligence of the vehicle and the user experience.

[0079] For example Figure 5 In the roundabout scenario shown, the local planning path is JBCDEM (e.g., Figure 5 In the driving space 1), the local planning path is a part of the global planning path. According to the driving space type marking method in the above embodiment, EFGHAB is marked as a candidate driving space (e.g., Figure 5 In the driving space 2), if there is congestion or traffic accident on the EM route, the vehicle can make local planning and decisions based on the actual situation and the marked driving space type. The replanned driving route is EFGHABCDEM. The vehicle is controlled to drive around the roundabout again and exit from EM. This can avoid the congested sections on the original planned route and can directly use the marked driving space type on the high-precision map to realize local planning and decisions. There is no need to replan the route using the standard map and then map it to the high-precision map. This effectively reduces the time of repeated mapping, improves the response speed of local route planning and decision-making, and enhances the intelligence of the vehicle and the user experience.

[0080] Taking the roundabout scenario as an example, such as Figure 6 As shown, the local planning path is JBCDEM (e.g., Figure 6 In the driving space 1), according to the driving space type marking method of the above embodiment, EFGO is marked as a candidate driving space (e.g., driving space 1). Figure 6 In the driving space 2), if there is congestion or traffic accident in the EM path, the vehicle can make local planning and decisions based on the actual situation and the marked driving space type, and replan the driving path as EFGO, and control the vehicle to drive out through GO.

[0081] It should be noted that the relevant technologies use standard map data and load high-precision map data to provide local planning and decision-making, but they cannot observe or calculate the impact of branch roads entering or leaving the planned path on local planning and decision-making, nor can they provide multiple road segment attribute methods to measure the optimal alternative driving space.

[0082] This application embodiment can help local planning and decision-making quickly produce high-quality vehicle driving trajectories by marking the driving space type. Furthermore, alternative routes can be selected by considering the influence of multiple dimensions and road segment attributes, enabling local planning and decision-making to produce better, faster, and smarter driving trajectories.

[0083] The map driving space type marking method proposed in the embodiments of this application can map the globally planned path on the standard map to the high-precision map, mark the driving space type on the high-precision map based on the planned path and the unplanned path, and directly re-plan the path on the high-precision map based on the driving space type. Therefore, it is not necessary to re-plan the path using the standard map and re-map it to the high-precision map. This allows for faster path re-planning, which is more conducive to making better, faster and smarter path planning in subsequent local planning and decision-making processes, and meets the local planning and decision-making needs of different complex scenarios.

[0084] This application also provides a driving route planning and decision-making method. It should be noted that the embodiments and descriptions of the map driving space type marking method in the above embodiments are also applicable to this embodiment.

[0085] like Figure 7 As shown, this driving route planning and decision-making method includes the following steps:

[0086] In step S201, the global planned path of the vehicle on the standard map is obtained.

[0087] In step S202, the global planning path is mapped to the high-precision map, and the first lane data of the local planning path and the second lane data related to the local planning path are identified within the target range of the high-precision map.

[0088] In step S203, driving space types are marked on a high-precision map based on the first lane data and / or the second lane data, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space, and alternative driving space.

[0089] In step S204, local planning and decision-making are performed based on the driving space type and high-precision map.

[0090] It is understood that the embodiments of this application can perform local planning and decision-making for vehicles based on the driving space type marked above and high-precision maps, and can make better, faster and more intelligent path planning to meet the local planning and decision-making needs of different complex scenarios.

[0091] In summary, the driving path planning and decision-making method of this application can solve the problem of the impact of planned path-related roads on local planning and decision-making in complex road network topologies. It can search and load path-related roads to expand the field of view of high-precision map data based on global planning, and calculate and search the expanded field of view of high-precision map data through various road segment attributes, and calculate and mark driving areas to reduce local planning and decision-making calculations. This enables local planning and decision-making to make better, faster and smarter driving trajectories.

[0092] The specific process is as follows: Figure 8 As shown, it mainly includes two process branches: process A calculates the driving space type, and process B calculates the alternative driving spaces.

[0093] Figure 8 The input data includes: standard map data, high-precision map data, global planning path node data, various road segment attributes and their weight coefficient configurations, and preset spatial network parameters; the output data includes: lane data (including attribute information and shape points), lane driving space markings (suggested, drivable, alternative, and non-suggested), and lane connection topology.

[0094] Process A sub-branch process: Using the first lane data of the local planned path within the target range of the high-precision map, the target road segment with entrances and exits, and the relationship between the lanes of the entrances and exits in the target road segment and the planned path, mark the driving space type.

[0095] Process B sub-branch process: Using preset spatial network parameters and multiple road segment attribute candidate values, a second matrix is ​​calculated in the high-precision map by combining the spatial network with the candidate values ​​of multiple road segment attributes (first matrix) and their weights. The candidate paths planned by the second matrix are marked as candidate driving spaces for reference. Figure 5 Roundabout scenarios are not limited to this one; other scenarios include parallel interchanges and main / auxiliary roads.

[0096] After the A sub-branch process and the B sub-branch process are completed, the driving space data is merged and broadcast to the local planning and decision-making module.

[0097] According to the driving route planning and decision-making method proposed in the embodiments of this application, the globally planned route on the standard map can be mapped to the high-precision map. Based on the planned and unplanned routes, the driving space type is marked on the high-precision map. Based on the driving space type, the route can be re-planned directly on the high-precision map, so that the route planning is not required to be performed again using the standard map and re-mapped to the high-precision map. The route planning can be re-made more quickly, which facilitates better, faster and more intelligent route planning in the local planning and decision-making process, and meets the local planning and decision-making needs of different complex scenarios.

[0098] Next, referring to the accompanying drawings, a map driving space type marking device according to an embodiment of this application is described.

[0099] Figure 9 This is a block diagram of a map driving space type marking device according to an embodiment of this application.

[0100] like Figure 9 As shown, the driving space type marking device 10 of the map includes: a first acquisition module 100, a first mapping module 200 and a first marking module 300.

[0101] The first acquisition module 100 is used to acquire the global planned path of the vehicle on the standard map; the first mapping module 200 is used to map the global planned path to the high-precision map and identify the first lane data and the second lane data related to the local planned path within the target range of the high-precision map; the first marking module 300 is used to mark the driving space type on the high-precision map data according to the first lane data and / or the second lane data of the planned path, and to perform path planning on the high-precision map according to the driving space type, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space and alternative driving space.

[0102] In this embodiment of the application, the first marking module 300 is further configured to: mark driving space types on a high-precision map based on first lane data and / or second lane data, including: marking drivable space, suggested driving space and non-suggested driving space on the high-precision map based on the first lane data; and marking alternative driving spaces on the high-precision map data based on the first lane data and the second data.

[0103] In this embodiment of the application, the first marking module 300 is further configured to: identify target road segments with entrances and exits on the local planning path in the first lane data, wherein the target road segment includes at least one lane; mark the lanes with entrances and exits in the target road segment as non-recommended driving spaces; mark the lanes without entrances and exits in the target road segment as recommended driving spaces; and mark road segments other than the target road segment on the planning path as driving spaces.

[0104] In this embodiment of the application, the first marking module 300 is further configured to: create a spatial network within the target area of ​​the high-precision map, wherein the spatial network includes multiple grids; determine the grid parameters of the target grid among the multiple grids based on the first lane data and the second lane data; and mark alternative driving spaces on the high-precision map based on the grid parameters of the target grid.

[0105] In this embodiment, the path in the first lane data and the second lane data is divided into multiple road segments, and the grid parameters include at least one road segment, and each road segment has multiple road segment attributes; the first marking module 300 is further used to: calculate the candidate value of each target grid according to the relevant data of each road segment attribute, wherein the candidate value represents the reference value of the road segment in the target grid that meets the vehicle passage conditions; construct a first matrix according to the candidate values ​​of all target grids of each road segment attribute; obtain a second matrix by weighting the first matrix of each road segment attribute and the weight of each road segment attribute; and use the candidate path planned based on the second matrix as the candidate driving space.

[0106] In this application embodiment, the data of multiple road segment attributes includes at least one of road segment direction, road segment width, road segment grade, and number of lanes.

[0107] It should be noted that the foregoing explanation of the method embodiment for marking the driving space type of a map also applies to the driving space type marking device of this embodiment, and will not be repeated here.

[0108] The map driving space type marking device proposed in the embodiments of this application can map the globally planned path on the standard map to the high-precision map, mark the driving space type on the high-precision map based on the planned path and the unplanned path, and directly re-plan the path on the high-precision map based on the driving space type. Therefore, it is not necessary to re-plan the path using the standard map and re-map it to the high-precision map. The path planning can be made more quickly, so as to make better, faster and more intelligent path planning in the subsequent local planning and decision-making process, and meet the local planning and decision-making needs of different complex scenarios.

[0109] Figure 10 This is a block diagram of the driving path planning and decision-making device according to an embodiment of this application.

[0110] like Figure 10 As shown, the driving path planning and decision-making device 20 includes: a second acquisition module 400, a second mapping module 500, a second marking module 600, and a planning module 700.

[0111] The second acquisition module 400 is used to acquire the global planned path of the vehicle on the standard map; the second mapping module 500 is used to map the global planned path to the high-precision map and identify the first lane data and the second lane data related to the local planned path within the target range of the high-precision map; the second marking module 600 is used to mark the driving space type on the high-precision map data according to the first lane data and / or the second lane data of the planned path, and to perform path planning on the high-precision map according to the driving space type, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space and alternative driving space; the planning module 700 is used to perform local planning and decision-making based on the driving space type and the high-precision map.

[0112] It should be noted that the foregoing explanation of the embodiments of the driving path planning and decision-making method also applies to the driving path planning and decision-making device of this embodiment, and will not be repeated here.

[0113] According to the driving route planning and decision-making device proposed in the embodiments of this application, the globally planned route on the standard map can be mapped to the high-precision map. Based on the planned and unplanned routes, the driving space type is marked on the high-precision map. Based on the driving space type, the route can be re-planned directly on the high-precision map, so that the route planning is not required to be performed again using the standard map and re-mapped to the high-precision map. The route planning can be re-made more quickly, which facilitates better, faster and more intelligent route planning in the local planning and decision-making process, and meets the local planning and decision-making needs of different complex scenarios.

[0114] Figure 11 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include:

[0115] The memory 1101, the processor 1102, and the computer program stored on the memory 1101 and executable on the processor 1102.

[0116] When the processor 1102 executes the program, it implements the map driving space type marking method or the driving path planning and decision-making method provided in the above embodiments.

[0117] Furthermore, the vehicle also includes:

[0118] Communication interface 1103 is used for communication between memory 1101 and processor 1102.

[0119] The memory 1101 is used to store computer programs that can run on the processor 1102.

[0120] The memory 1101 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0121] If the memory 1101, processor 1102, and communication interface 1103 are implemented independently, then the communication interface 1103, memory 1101, and processor 1102 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 11 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0122] Optionally, in a specific implementation, if the memory 1101, processor 1102, and communication interface 1103 are integrated on a single chip, then the memory 1101, processor 1102, and communication interface 1103 can communicate with each other through an internal interface.

[0123] The processor 1102 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0124] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0125] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0126] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0127] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0128] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0129] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for marking driving space types on a map, characterized in that, Includes the following steps: Obtain the vehicle's globally planned path on the standard map; The global planning path is mapped to a high-precision map, and the first lane data of the local planning path and the second lane data related to the local planning path are identified within the target area of ​​the high-precision map. Marking driving space types on the high-precision map based on the first lane data and / or the second lane data, wherein the driving space type includes at least one of drivable space, suggested driving space, unsuggested driving space, and alternative driving space; marking driving space types on the high-precision map based on the first lane data and / or the second lane data includes: marking the drivable space, the suggested driving space, and the unsuggested driving space on the high-precision map based on the first lane data; marking the alternative driving space on the high-precision map based on the first lane data; marking the drivable space, the suggested driving space, and the unsuggested driving space on the high-precision map based on the first lane data includes: identifying target road segments with entrances / exits on the local planned path in the first lane data, wherein the target road segment includes at least one lane; marking lanes with entrances / exits in the target road segment as unsuggested driving spaces; marking lanes without entrances / exits in the target road segment as suggested driving spaces; and marking road segments other than the target road segment on the planned path as unsuggested driving spaces. The process of marking the candidate driving space on the high-precision map based on the first lane data and the second lane data includes: creating a spatial network within the target area of ​​the high-precision map, wherein the spatial network includes multiple grids; determining the grid parameters of the target grid among the multiple grids based on the first lane data and the second lane data; marking the candidate driving space on the high-precision map based on the network parameters of the target grid, wherein the path in the first lane data and the second lane data is divided into multiple road segments, the grid parameters include at least one road segment, and each road segment has multiple road segment attributes; marking the candidate driving space on the high-precision map based on the grid parameters of the target grid includes: calculating the candidate value of each target grid based on the relevant data of each road segment attribute, wherein the candidate value represents a reference value for the road segment within the target grid to meet the vehicle passage conditions; constructing a first matrix based on the candidate values ​​of all target grids for each road segment attribute; obtaining a second matrix by weighting the first matrix of each road segment attribute and the weights of each road segment attribute; and using the candidate path planned based on the second matrix as the candidate driving space.

2. The method for marking driving space types on a map according to claim 1, characterized in that, The road segment attributes include at least one of the following: road segment direction, road segment width, road segment class, and number of lanes.

3. A method for planning and deciding on a driving route, characterized in that, Includes the following steps: Obtain the vehicle's globally planned path on the standard map; The global planning path is mapped to a high-precision map, and the first lane data of the local planning path and the second lane data related to the local planning path are identified within the target area of ​​the high-precision map. Marking driving space types on the high-precision map based on the first lane data and / or the second lane data, wherein the driving space type includes at least one of drivable space, suggested driving space, unsuggested driving space, and alternative driving space; marking driving space types on the high-precision map based on the first lane data and / or the second lane data includes: marking the drivable space, the suggested driving space, and the unsuggested driving space on the high-precision map based on the first lane data; marking the alternative driving space on the high-precision map based on the first lane data; marking the drivable space, the suggested driving space, and the unsuggested driving space on the high-precision map based on the first lane data includes: identifying target road segments with entrances / exits on the local planned path in the first lane data, wherein the target road segment includes at least one lane; marking lanes with entrances / exits in the target road segment as unsuggested driving spaces; marking lanes without entrances / exits in the target road segment as suggested driving spaces; and marking road segments other than the target road segment on the planned path as unsuggested driving spaces. The process of marking the candidate driving space on the high-precision map based on the first lane data and the second lane data includes: creating a spatial network within the target area of ​​the high-precision map, wherein the spatial network includes multiple grids; determining the grid parameters of the target grid among the multiple grids based on the first lane data and the second lane data; marking the candidate driving space on the high-precision map based on the grid parameters of the target grid, wherein the path in the first lane data and the second lane data is divided into multiple road segments, the grid parameters include at least one road segment, and each road segment has multiple road segment attributes; marking the candidate driving space on the high-precision map based on the grid parameters of the target grid includes: calculating the candidate value of each target grid based on the relevant data of each road segment attribute, wherein the candidate value represents a reference value for the road segment within the target grid to meet the vehicle passage conditions; constructing a first matrix based on the candidate values ​​of all target grids for each road segment attribute; obtaining a second matrix by weighting the first matrix of each road segment attribute and the weights of each road segment attribute; and using the candidate path planned based on the second matrix as the candidate driving space. Local planning and decision-making are based on driving space type and high-precision maps.

4. A map navigation space type marking device, characterized in that, The apparatus is used to perform the map driving space type marking method as described in any one of claims 1-2, wherein the apparatus comprises: The first acquisition module is used to acquire the global planned path of the vehicle on the standard map; The first mapping module is used to map the global planning path to a high-precision map, and identify the first lane data and the second lane data related to the local planning path within the target range of the high-precision map. The first marking module is used to mark the driving space type on the high-precision map according to the first lane data and / or the second lane data of the planned route, and to perform route planning on the high-precision map according to the driving space type, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space and alternative driving space.

5. A driving path planning and decision-making device, characterized in that, The device is used to execute the driving path planning and decision-making method as described in claim 3, wherein the device comprises: The second acquisition module is used to acquire the vehicle's global planned path on the standard map; The second mapping module is used to map the global planning path to a high-precision map, and to identify the first lane data of the local planning path and the second lane data related to the local planning path within the target range of the high-precision map. The second marking module is used to mark the driving space type on the high-precision map according to the first lane data and / or the second lane data of the planned route, and to perform route planning on the high-precision map according to the driving space type, wherein the driving space type includes at least one of drivable space, suggested driving space, non-suggested driving space and alternative driving space; The planning module is used to perform local planning and decision-making based on the driving space type and the high-precision map.

6. A vehicle, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, the processor executing the program to implement the map driving space type marking method according to any one of claims 1-2, or the driving path planning and decision method according to claim 3.

Images