Trajectory point recognition method and device and map making method

Abstract

The embodiment of the present application provides a trajectory point identification method, device and map making method, which comprises the following steps: for a plurality of trajectory points in a target trajectory, determining a matching road section corresponding to each trajectory point; for any trajectory point, determining the connection relationship between the matching road section and a reference road section, wherein the road type of the reference road section is a highway, or the road type of the reference road section is a non-highway; according to the connection relationship, determining the data type corresponding to the trajectory point, wherein the data type is a high-speed type or a non-high-speed type; and dividing the trajectory points corresponding to the same data type into the same trajectory point set, wherein the trajectory point set of the high-speed type is used for determining the map data of the highway, and the trajectory point set of the non-high-speed type is used for determining the map data of the non-highway. The technical scheme of the present application can effectively guarantee the quality of the map data.

Description

Technical Field

[0001] This application relates to the field of high-precision mapping technology, and in particular to a trajectory point recognition method, apparatus, and map production method. Background Technology

[0002] In the process of creating high-precision maps, because the production processes for highways and ordinary roads are quite different, highways and ordinary roads usually need to be processed separately.

[0003] Currently, the creation of high-definition maps typically relies on collected road data. During the data collection process for highways, it's inevitable that some data from ordinary roads will be collected; similarly, during the data collection process for ordinary roads, some data from highways will inevitably be collected. This results in the inclusion of road data outside the designated area when creating high-definition maps for either highways or ordinary roads.

[0004] The presence of road data that exceeds the permitted scope can lead to a mismatch between the road data and the processing scenario, resulting in quality issues with high-precision maps created based on such road data. Summary of the Invention

[0005] This application provides a trajectory point recognition method, apparatus, and map production method to overcome the problem of excessive road data affecting the production quality of high-precision maps.

[0006] In a first aspect, embodiments of this application provide a trajectory point recognition method, including:

[0007] For multiple trajectory points in the target trajectory, determine the matching road segment corresponding to each trajectory point;

[0008] For any of the trajectory points, determine the connection relationship between the matching road segment and the reference road segment, wherein the road type of the reference road segment is a highway, or the road type of the reference road segment is a non-highway;

[0009] Based on the connection relationship, the data type corresponding to the trajectory point is determined, and the data type is either high-speed type or non-high-speed type;

[0010] Trajectory points of the same data type are divided into the same trajectory point set. The trajectory point set of high-speed type is used to determine the map data of high-speed roads, and the trajectory point set of non-high-speed type is used to determine the map data of non-high-speed roads.

[0011] Secondly, embodiments of this application provide a map creation method, including:

[0012] The set of trajectory points is determined according to the method described in the first aspect above;

[0013] If the data type corresponding to the trajectory point set is high-speed, then map data of the high-speed road is created based on the trajectory point set.

[0014] If the data type corresponding to the trajectory point set is non-highway type, then map data for non-highway roads is created based on the trajectory point set.

[0015] Thirdly, embodiments of this application provide a trajectory point recognition device, comprising:

[0016] The determination module is used to determine the matching road segment corresponding to each of the multiple trajectory points in the target trajectory.

[0017] The determining module is further configured to determine the connection relationship between the matching road segment and the reference road segment for any of the trajectory points, wherein the road type of the reference road segment is a highway, or the road type of the reference road segment is a non-highway.

[0018] The determining module is further configured to determine the data type corresponding to the trajectory point based on the connection relationship, wherein the data type is a high-speed type or a non-high-speed type;

[0019] The partitioning module is used to divide trajectory points of the same data type into the same trajectory point set. The trajectory point set of the high-speed type is used to determine the map data of the high-speed road, and the trajectory point set of the non-high-speed type is used to determine the map data of the non-high-speed road.

[0020] Fourthly, embodiments of this application provide an electronic device, including:

[0021] Memory, used to store programs;

[0022] A processor for executing the program stored in the memory, wherein when the program is executed, the processor is configured to perform the methods described in the first and second aspects above.

[0023] Fifthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the methods described in the first and second aspects above.

[0024] Sixthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in the first and second aspects above.

[0025] This application provides a trajectory point identification method, apparatus, and map creation method. The method includes: determining the matching road segment corresponding to each trajectory point in a target trajectory; determining the connection relationship between the matching road segment and the reference road segment for any trajectory point, wherein the road type of the reference road segment is either a highway or a non-highway; determining the data type corresponding to the trajectory point based on the connection relationship, wherein the data type is either highway or non-highway; dividing trajectory points of the same data type into the same trajectory point set, wherein the highway type trajectory point set is used to determine the map data of highways, and the non-highway type trajectory point set is used to determine the map data of non-highway roads. By determining the matching road segment corresponding to each trajectory point in the collected target trajectory, and then determining the data type of the trajectory point based on the connection relationship between the matching road segment and the reference road segment, the highway type trajectory points and non-highway type trajectory points in the target trajectory are distinguished. Then, the trajectory points of the same data type are divided into the same trajectory point set, thereby ensuring that no out-of-range data is included in each data type trajectory point set, thus effectively guaranteeing the quality of the map data. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram illustrating a scenario for collecting road data, provided in an embodiment of this application.

[0028] Figure 2 A flowchart of the trajectory point recognition method provided in the embodiments of this application;

[0029] Figure 3 The flowchart of the trajectory point recognition method provided in the embodiments of this application Figure 2 ;

[0030] Figure 4 A schematic diagram illustrating the implementation of the target trajectory provided in an embodiment of this application;

[0031] Figure 5 This is a schematic diagram of the angles of the sub-trajectories and road segments provided in the embodiments of this application;

[0032] Figure 6 A schematic diagram showing the angle between the driving direction of the sub-trajectory and the road segment provided in the embodiments of this application;

[0033] Figure 7A schematic diagram illustrating the implementation of determining trajectory point pairs in an embodiment of this application;

[0034] Figure 8 A schematic diagram of a multi-level matching process provided in an embodiment of this application;

[0035] Figure 9 The flowchart of the trajectory point recognition method provided in the embodiments of this application Figure 3 ;

[0036] Figure 10 A schematic diagram illustrating the implementation of the fragmented storage matching relationship provided in an embodiment of this application;

[0037] Figure 11 The flowchart of the trajectory point recognition method provided in the embodiments of this application Figure 4 ;

[0038] Figure 12 A schematic diagram of the topological relationship of road segments provided in the embodiments of this application;

[0039] Figure 13 A schematic diagram illustrating the execution process of the trajectory point recognition method provided in this application embodiment;

[0040] Figure 14 This is a schematic diagram of the trajectory point recognition device provided in the embodiments of this application;

[0041] Figure 15 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0043] To better understand the technical solution of this application, the relevant technical background involved in this application will be further described in detail below.

[0044] High-definition (HD) maps are maps that, in contrast to standard (SD) maps, have higher requirements for map detail and richness. For example, standard maps only achieve road-level precision, representing a road only with a single centerline and unable to accurately describe lane information (a road contains multiple lanes). HD maps, however, achieve lane-level precision, representing not only road centerlines but also lane centerlines. Furthermore, while standard maps use topological networks to represent the connections between roads, HD maps can represent not only the topological relationships between roads but also the connections between lanes within a road.

[0045] It is also certain that roads in the real world can include highways and ordinary roads. Highways, also known as expressways, are roads specifically designed for high-speed automobile travel. The definition of a highway can vary in different regions and eras. In this embodiment, the term "highway" refers to a road specifically designed for high-speed automobile travel, as defined during road planning, and can be determined based on actual road planning conditions. Ordinary roads can be understood as ordinary roads in cities or towns, which have certain speed limits and are typically equipped with traffic facilities such as traffic lights.

[0046] The road conditions of highways and ordinary roads differ significantly. Therefore, the production processes for high-definition maps differ greatly between highways and ordinary roads. Currently, data production and delivery are carried out separately for highways and ordinary roads. The production of high-definition maps relies on road data collected by data acquisition equipment. For example, point cloud data obtained from multiple collection trips can be aligned to create high-definition maps.

[0047] In one scenario, to ensure alignment at the junctions of highways and ordinary roads in a high-precision map, the data collection process for highways typically includes data on connecting roads, ramps, and some ordinary roads. This results in the road data collected for highways including data from roads that are not highways; this latter part of the road data can be considered out-of-scope processing data.

[0048] Similarly, during the field data collection process for ordinary roads, a portion of highways will also be covered. This also means that the road data collected for ordinary roads includes road data that is not ordinary roads. This part of the road data can also be understood as data processed beyond the scope.

[0049] It should be noted that access roads and ramps can be marked as highways or ordinary roads, and their implementation may vary in different maps.

[0050] In another scenario, during data collection on highways, the data acquisition equipment typically travels along the road to collect data. Due to route planning and road intersections, this equipment inevitably traverses some non-highway sections, resulting in the collection of data beyond its intended processing scope. Similarly, during data collection on ordinary roads, the equipment may also traverse some highway sections, again leading to the collection of data beyond its intended processing scope.

[0051] Therefore, based on the above introduction, it can be determined that in the current process of collecting road data for highways and ordinary roads, it is inevitable that data beyond the scope of processing will be collected. For example, we can refer to... Figure 1 To gain a further understanding, Figure 1 This is a schematic diagram of a scenario for collecting road data provided in an embodiment of this application.

[0052] like Figure 1 As shown in Figure 10, assume there are currently two roads: a highway 101 and a regular road 102. Assume that after data collection for highway 101, the following results are obtained: Figure 1 The example number 20 illustrates a scenario where road data for highways was collected, but it also included some road data for ordinary roads. Figure 1 In the case shown in Figure 20, area b represents the covered ordinary road. Also, if a ramp is identified as an ordinary road, then area a in the figure also represents the covered ordinary road.

[0053] And after collecting data on ordinary road 102, we obtained Figure 1 The example in section 30 illustrates a scenario where road data from ordinary roads was collected, but it also included some road data from highways. Figure 1 In the case shown in Figure 30, region d represents the covered highway. Also, if a ramp is identified as a highway, then region c in the figure also represents the covered highway.

[0054] based on Figure 1 The example confirms that road data that does not belong to the current road type can be considered out-of-scope processing data. It should also be noted that... Figure 1 The text merely uses illustrations to demonstrate the processing of data beyond its scope, for example, in... Figure 1In examples 20 and 30, regardless of whether the road is a regular road or a highway, the road coverage for the collected road data includes both two-way lanes. However, in actual implementation, it is possible to collect road data only for one-way lanes. Therefore, in the actual data collection process, the specific road coverage can be determined according to actual needs.

[0055] This excessive data processing causes two problems. First, it increases data production costs. Second, due to different production processes, operators who process highway data may need to process ordinary road data, or vice versa. In either case, it leads to a mismatch between operational skills and the data being processed, resulting in quality issues with the high-precision map.

[0056] Regarding the technical problems described above, it is understandable that a processing method is needed to identify out-of-range data in the collected data. If manual annotation is used for identification, low efficiency and accuracy will result. Therefore, this application proposes the following technical concept: The collected data can correspond to the driving trajectory of the collected vehicle. For each trajectory point in the driving trajectory, a matching road segment corresponding to each trajectory point can be determined in a preset road network structure. Then, based on the connection relationship between the matching road segment corresponding to each trajectory point and the reference road segment in the road network structure, it is marked whether each trajectory point belongs to a highway or a regular road. Afterwards, based on the marking results, trajectory points of the same data type are divided into the same trajectory point set. Then, based on the different types of trajectory point sets, the corresponding type of map data is determined, thereby enabling accurate and suitable subsequent high-precision map production.

[0057] Based on the above introduction, the trajectory point recognition method provided in this application will be described in detail below with reference to specific embodiments. It is worth noting that the execution entity in each embodiment of this application can be a cloud server, local server, processor, chip, or other device with data processing capabilities, or a terminal device such as a computer or tablet with data processing capabilities. This embodiment does not limit the specific implementation method of the execution entity; it can be selected and set according to actual needs.

[0058] The following is a combination of... Figure 2 To introduce, Figure 2 A flowchart of the trajectory point recognition method provided in the embodiments of this application.

[0059] like Figure 2 As shown, the method includes:

[0060] S201. For multiple trajectory points in the target trajectory, determine the matching road segment corresponding to each trajectory point.

[0061] The target trajectory is the path taken by the data acquisition device during the collection of road data, and may include multiple trajectory points. In this embodiment, for example, a preset road network structure may be defined, which could be a road network structure in a high-resolution map, including multiple road segments and the topological relationships between these segments.

[0062] It is understandable that the target trajectory is the trajectory traveled by the data acquisition device, and it is not associated with the actual road. Therefore, in this embodiment, for multiple trajectory points in the target trajectory, the matching road segment corresponding to each trajectory point can be determined in the road network structure, thereby associating the target trajectory with the actual road.

[0063] For example, for any trajectory point 'a', the matching road segment corresponding to trajectory point 'a' is road segment 1, which means that trajectory point 'a' in the target trajectory was actually collected on road segment 1. It can also be understood that one road segment can correspond to multiple trajectory points, therefore, there can be multiple trajectory points whose matching road segments are all road segment 1.

[0064] S202. For any trajectory point, determine the connection relationship between the matching road segment and the reference road segment. The road type of the reference road segment is either a highway or a non-highway.

[0065] Based on the above introduction, it can be determined that the road network structure includes multiple road segments, each of which can correspond to a road type. In one possible implementation, the road segment type can include expressways and non-expressways.

[0066] It is understood that "high-speed" in this embodiment does not actually limit a very high speed value. Road segments classified as highways correspond to road segments that are classified as highways in real-world scenarios. Conversely, road segments classified as non-high-speed correspond to road segments that are not classified as highways in real-world scenarios.

[0067] After determining the matching road segment corresponding to each trajectory point, the connection relationship between the matching road segment corresponding to each trajectory point and the reference road segment in the road network structure can be determined. In this embodiment, the road type of the reference road segment can be a highway, or the road type of the reference road segment in this embodiment can also be a non-highway. The specific implementation of the road type of the reference road segment can be selected according to actual needs.

[0068] The connection relationships can include direct connections and indirect connections. A direct connection means that the connection to the reference segment occurs without passing through any intermediate road segments, while an indirect connection requires passing through at least one intermediate road segment to connect to the reference segment.

[0069] S203. Based on the connection relationship, determine the data type corresponding to each trajectory point. The data type can be either high-speed or non-high-speed.

[0070] After determining the connection relationship between the matching road segment and the reference road segment corresponding to the trajectory point, for example, the data type corresponding to the trajectory point can be determined based on this connection relationship. The data type can be either high-speed or non-high-speed. It should be noted that the data type determined for the trajectory point here is the type re-marked after the trajectory point is collected, and it may be inconsistent with the data type corresponding to the original data collection of the trajectory point.

[0071] In one possible implementation, taking the reference road segment as an example where the road type is a highway, when the connection is direct, it can be determined that the matching road segment corresponding to the trajectory point and the highway segment are directly connected. In this case, it can be determined that the data type corresponding to the trajectory point is highway type, which means that this trajectory point should be processed as highway segment data.

[0072] Alternatively, when the connection is indirect, it can be determined that the matching road segment and the highway segment corresponding to the trajectory point are indirectly connected. In this case, for example, it can be determined that the data type of the trajectory point is non-highway type, which means that the trajectory point should be processed as ordinary road segment data.

[0073] In another possible implementation, taking the reference road segment as a non-highway as an example, when the connection relationship is direct, it can be determined that the matching road segment and non-highway segment corresponding to the trajectory point are directly connected. In this case, it can be determined that the data type corresponding to the trajectory point is non-highway type, which means that this trajectory point should be processed as non-highway road segment data.

[0074] Alternatively, for example, when the connection relationship is indirect, it can be determined that the matching road segment and non-highway segment corresponding to the trajectory point are indirectly connected. In this case, for example, it can be determined that the data type of the trajectory point is high-speed type, which means that this trajectory point should be processed as high-speed road segment data.

[0075] S204. Divide trajectory points of the same data type into the same trajectory point set, wherein the trajectory point set of high-speed type is used to determine the map data of high-speed roads, and the trajectory point set of non-high-speed type is used to determine the map data of non-high-speed roads.

[0076] After labeling each trajectory point with its corresponding data type, for example, trajectory points with the same data type can be further divided into the same trajectory point set. Accordingly, a high-speed type trajectory point set and a non-high-speed type trajectory point set can be obtained.

[0077] In the high-speed type trajectory point set, all trajectory points correspond to high-speed data types. Therefore, the high-speed type trajectory point set does not include out-of-scope work data. Then, by determining the high-speed map data based on the high-speed type trajectory point set, it can be effectively ensured that the work data and the work skills of the workers are matched, thus effectively improving the quality of high-speed map data production.

[0078] Furthermore, the data types corresponding to the trajectory points included in the non-high-speed type trajectory point set are all non-high-speed type. Therefore, the non-high-speed type trajectory point set does not include out-of-range operation data. Then, by determining the non-high-speed road map data based on the non-high-speed type trajectory point set, it can be effectively ensured that the operation data and the operation skills of the operators are matched, thus effectively improving the production quality of non-high-speed road map data.

[0079] The method of determining map data based on a set of trajectory points, as described above, can be understood as creating high-precision or standard-precision maps based on the set of trajectory points. The specific implementation method of the map data can be selected according to actual needs.

[0080] The trajectory point identification method provided in this application includes: determining the matching road segment corresponding to each trajectory point in a target trajectory; determining the connection relationship between the matching road segment and the reference road segment for any trajectory point, wherein the road type of the reference road segment is either a highway or a non-highway; determining the data type corresponding to the trajectory point based on the connection relationship, wherein the data type is either a highway type or a non-highway type; dividing trajectory points of the same data type into the same trajectory point set, wherein the highway type trajectory point set is used to determine the map data of highways, and the non-highway type trajectory point set is used to determine the map data of non-highway roads. By determining the matching road segment corresponding to each trajectory point in the collected target trajectory, and then determining the data type of the trajectory point based on the connection relationship between the matching road segment and the reference road segment, the highway type trajectory points and non-highway type trajectory points in the target trajectory are distinguished. Then, the trajectory points of the same data type are divided into the same trajectory point set, thereby ensuring that no out-of-range data is included in each data type trajectory point set, thus effectively guaranteeing the quality of the map data.

[0081] Based on the above introduction, the following section provides a more detailed explanation of how to determine the matching road segment corresponding to each trajectory point, using specific embodiments. Figures 3 to 8 To explain, Figure 3 The flowchart of the trajectory point recognition method provided in the embodiments of this application Figure 2 , Figure 4 This is a schematic diagram illustrating the implementation of the target trajectory provided in an embodiment of this application. Figure 5 This is a schematic diagram of the angles of the sub-trajectories and road segments provided in the embodiments of this application. Figure 6 This is a schematic diagram showing the angle between the driving direction of the sub-trajectory and the road segment provided in the embodiments of this application. Figure 7 This is a schematic diagram illustrating the implementation of determining trajectory point pairs according to an embodiment of this application. Figure 8 This is a schematic diagram of a multi-level matching process provided in an embodiment of this application.

[0082] like Figure 3 As shown, the method includes:

[0083] S301. For any trajectory point, determine a sub-trajectory in the target trajectory, the sub-trajectory including the trajectory point.

[0084] In the process of determining the matching road segment corresponding to the trajectory point, for example, we can first match the sub-trajectories including the trajectory point and each road segment to initially filter out some trajectory points with a very high matching degree.

[0085] Therefore, in one possible implementation, this embodiment can determine the sub-trajectories of each trajectory point in the target trajectory. Since the processing methods for each trajectory point are similar, the following explanation uses any one of the multiple trajectory points as an example.

[0086] In this embodiment, a sub-trajectory can be determined within the target trajectory, and this sub-trajectory includes the currently corresponding trajectory point. For example, multiple trajectory points near the current trajectory point can be selected, and these trajectory points together constitute the sub-trajectory. For instance, this can be combined with... Figure 4 To understand, such as Figure 4 As shown, for example, the current target trajectory includes trajectory points a to j, and we assume that a sub-trajectory is being determined for trajectory point d.

[0087] In one possible implementation, for example, multiple trajectory points preceding trajectory point d can be selected, such as selecting... Figure 4 The trajectory points a, b, and c shown in the figure are then used to determine the trajectory of the target trajectory consisting of trajectory points a, b, c, and d as the sub-trajectory corresponding to trajectory point d.

[0088] Alternatively, for example, you could select multiple trajectory points after trajectory point d, such as selecting... Figure 4 The trajectory points e, f, and g shown in the figure are then used to determine the trajectory composed of trajectory points d, e, f, and g in the target trajectory as the sub-trajectory corresponding to trajectory point d.

[0089] Alternatively, for example, you could select multiple trajectory points before and after trajectory point d, such as selecting... Figure 4 The trajectory points b, c, e, and f shown in the figure are then used to determine the trajectory composed of trajectory points b, c, d, e, and f in the target trajectory as the sub-trajectory corresponding to trajectory point d.

[0090] In actual implementation, the specific method for determining the sub-trajectory and the number of trajectory points in the sub-trajectory can be selected and set according to actual needs, as long as the determined sub-trajectory is a trajectory that includes the current trajectory point in the target trajectory.

[0091] S302. For any road segment, determine the matching information between the sub-trajectory and the road segment based on the trajectory information of the sub-trajectory and the road segment information.

[0092] After determining the sub-trajectory corresponding to the trajectory point, for example, the matching information between the sub-trajectory and multiple road segments in the road network structure can be determined separately. The processing method for each road segment in the road network structure is similar. Therefore, the following explanation will take any road segment as an example, and the implementation method for other road segments will not be repeated.

[0093] In this embodiment, matching information between a sub-trajectory and a road segment can be determined based on the trajectory information of the sub-trajectory and the road segment information. This matching information indicates the matching status between the sub-trajectory and the road segment.

[0094] In one possible implementation, the trajectory information of a sub-track may include, for example, the angle between the sub-track line and a preset direction (e.g., due north), the position of the sub-track, its altitude, its direction of travel, and its length. Similarly, the road segment information may include, for example, the angle between the road segment line and a preset direction (e.g., due north), its position, altitude, direction of travel, and its length. In practice, the implementation methods for the trajectory information of sub-tracks and the road segment information can be selected and set according to actual needs.

[0095] In this embodiment, the matching information between the sub-track and the road segment may include at least one of the following: the angle between the sub-track and the road segment, the distance between the sub-track and the road segment, the height difference between the sub-track and the road segment, and the angle of the driving direction between the sub-track and the road segment.

[0096] The angle between the sub-trajectory and the road segment can be, for example, the angle between the trajectory line of the sub-trajectory and the line of the road segment. For instance, it can be combined with... Figure 5 To understand, such as Figure 5 As shown, assume there exists a sub-trajectory consisting of trajectory points a, b, c, and d, and a road segment 501. The angle between the sub-trajectory and road segment 501 is... Figure 5 The β shown is used to determine the angle between a sub-track and a road segment. For example, the angle between the sub-track line and the preset direction, and the angle between the road segment line and the preset direction, can be determined.

[0097] Furthermore, the distance between sub-trajectories and road segments can be considered. For example, it could be the horizontal distance between a sub-trajectory and a road segment. Specifically, the horizontal distance between each trajectory point in the sub-trajectory and the road segment can be determined, and then the distance between the sub-trajectory and the road segment can be determined based on the horizontal distance between each trajectory point. Alternatively, the distance between the sub-trajectory and the road segment can be determined based on the horizontal distance between the midpoint of the sub-trajectory and the road segment.

[0098] Additionally, the elevation difference between a sub-track and a road segment can be the difference between the elevation of the sub-track and the elevation of the road segment.

[0099] Furthermore, sub-trajectories correspond to different driving directions, and road segments also correspond to different driving directions. Therefore, the angle between the driving directions of sub-trajectories and road segments can also be determined. For example, one can refer to... Figure 6 To understand, such as Figure 6 As shown, assume there is a sub-trajectory consisting of trajectory points a, b, c, and d, and road segment 601 currently exists. (Refer to...) Figure 6 It is certain that road segment 601 and the sub-track are parallel, so in this case, the angle between the road segment and the sub-track is 0. However, in Figure 6 In the illustration, the driving direction of the road segment and the driving direction of the sub-track are opposite, so the angle between the driving directions can be understood as 180 degrees.

[0100] In actual implementation, the matching information is not limited to the content described above. Any information that can indicate the degree of matching between sub-trajectories and road segments can be used as the matching information in this embodiment. Therefore, the specific implementation of the matching information can be selected and set according to actual needs.

[0101] S303. If the matching information between the sub-trajectory and the road segment satisfies the initial threshold condition, then the road segment that satisfies the initial threshold condition is determined to be the matching road segment corresponding to the trajectory point, and the trajectory point is determined to be the trajectory point in the initial trajectory point set.

[0102] The matching information between sub-trajectories and road segments described above can also be understood as the matching information between trajectory points and road segments. Therefore, the matching information can describe the degree of matching between trajectory points and road segments. In this embodiment, it can be further determined whether the initial threshold condition meets the matching information, thereby determining whether the current road segment and trajectory point match.

[0103] In one possible implementation, the initial threshold condition includes at least one of the following: the angle is less than or equal to a preset angle threshold, the distance is less than or equal to a preset distance threshold, the height difference is less than or equal to a preset height difference threshold, and the included angle of the driving direction is less than or equal to a preset included angle threshold.

[0104] Understandably, when the angle between the sub-track and the road segment is less than or equal to a preset angle threshold, it can be determined that the sub-track and the road segment are relatively consistent in direction. When the distance between the sub-track and the road segment is less than or equal to a preset distance threshold, it can be determined that the sub-track and the road segment are relatively close in distance. When the height difference between the sub-track and the road segment is less than or equal to a preset height difference threshold, it can be determined that the sub-track and the road segment are relatively close in altitude. When the angle between the driving directions of the sub-track and the road segment is less than or equal to a preset angle threshold, it can be determined that the sub-track and the road segment are relatively consistent in driving direction.

[0105] Therefore, when the matching information between a trajectory point and a road segment meets the initial threshold condition, the matching degree between the current trajectory point and the road segment can be considered relatively high. Thus, the road segment that meets the initial threshold condition is the matching road segment corresponding to the trajectory point.

[0106] If there are multiple road segments that meet the initial threshold conditions, then, for example, we can further determine the matching degree of each road segment based on the matching information corresponding to each of the multiple road segments that meet the initial threshold conditions (for example, by performing mathematical operations or weighted processing, etc.), and then determine the road segment with the highest matching degree as the matching road segment corresponding to the trajectory point.

[0107] In one possible implementation, the values ​​of the preset angle threshold, preset distance threshold, preset height difference threshold, and preset included angle threshold in this embodiment can all be limited to a relatively small range, thereby ensuring that the matching relationship between the currently determined trajectory points and road segments can be accurate.

[0108] Furthermore, in this embodiment, the trajectory points of the matching road segments determined according to the initial threshold conditions are identified as trajectory points in the initial trajectory point set, thereby obtaining the initial trajectory point set. Due to the strict limitation of the initial threshold conditions, it can be guaranteed that the trajectory points in the initial trajectory point set can guarantee the accuracy of the corresponding matching road segments.

[0109] S304. Based on the topological relationship between multiple road segments and the initial trajectory point set, the matching road segment corresponding to the target trajectory point is completed to obtain the target trajectory point set. The target trajectory point is at least a portion of the trajectory points other than those in the initial trajectory point set.

[0110] After determining the initial set of trajectory points, for example, based on the initial set of trajectory points and the topological relationships between multiple road segments, target trajectory points that can uniquely determine the corresponding matching road segments can be inferred. Then, the matching information of these inferred target trajectory points is supplemented. Therefore, in this embodiment, the target trajectory points are at least a portion of the trajectory points other than those in the initial set of trajectory points. Whether trajectory points can be supplemented depends on the topological relationships between the initial set of trajectory points and the road segments. Therefore, the specific number of trajectory points that will be identified as target trajectory points, excluding those in the initial set, needs to be determined based on the actual situation.

[0111] The following describes possible methods for determining target trajectory points. For example, at least one pair of trajectory points can be determined based on the matching road segments and topological relationships corresponding to each trajectory point in the initial trajectory point set. These trajectory point pairs must meet certain conditions.

[0112] For example, it can be based on the following: Figure 7 To understand:

[0113] Specifically, it requires that the two trajectory points in a trajectory point pair correspond to the same common matching road segment. For example, refer to... Figure 7 Assuming Figure 7 Trajectory points a and b in the dataset belong to the initial set of trajectory points, and the matching road segment corresponding to trajectory point a is... Figure 7 The matching road segment corresponding to trajectory point b is also shown as road segment 701. Figure 7 If the road segment 701 shown is used, then it can be determined that trajectory point a and trajectory point b correspond to the same common matching road segment 701.

[0114] Furthermore, it is required that there exists at least one intermediate trajectory point between the two trajectory points in the target trajectory, and this intermediate trajectory point is not included in the initial trajectory point set. For example, referencing... Figure 7 In the target trajectory, there are two intermediate trajectory points between trajectory point a and trajectory point b, namely trajectory point b and trajectory point c, where trajectory point b and trajectory point c are not included in the trajectory point set.

[0115] Furthermore, it requires that only a common matching road segment exists between the locations corresponding to the two trajectory points in the road network structure. For example, refer to... Figure 7In the road network structure, there is only road segment 701 between the location corresponding to trajectory point a and the location corresponding to trajectory point b, and there are no other road segments. Figure 7 To illustrate the possibility of other road sections, road section 702 is drawn as a dashed line.

[0116] Two trajectory points that meet the conditions described above are identified as a trajectory point pair. Thus, at least one trajectory point pair is determined based on the matching road segments corresponding to each trajectory point in the initial trajectory point set.

[0117] Then, for any pair of trajectory points, the intermediate trajectory points between the two trajectory points in the pair can be determined as the target trajectory points described above. The matching road segment corresponding to the intermediate trajectory point is the common matching road segment described above.

[0118] Specifically, two trajectory points must meet certain conditions to be identified as a trajectory point pair. The intermediate trajectory points between these pairs that meet the aforementioned conditions must correspond to the matching road segments of the trajectory points within the pair. Therefore, in this embodiment, the intermediate trajectory points between the two trajectory points in a pair can be identified as target trajectory points, and the matching road segments corresponding to the target trajectory points can be identified as the common matching road segments of the trajectory point pair. This allows for accurate completion of the matching road segments corresponding to the target trajectory points based on the initial set of trajectory points and the topological relationships between road segments.

[0119] For example in Figure 7 In the example, based on the trajectory point pair formed by trajectory point a and trajectory point b, it can be determined that trajectory points b and c between trajectory points a and b must correspond to road segment 701, because there are no other road segments in between that can be traveled. Therefore, the matching relationship between trajectory point b and road segment 701, as well as the matching relationship between trajectory point c and road segment 701, determined in this way, can guarantee accuracy. Trajectory points b and c are the target trajectory points mentioned above. By completing the matching road segment corresponding to the target trajectory points, it is determined that the matching road segment corresponding to trajectory points b and c is road segment 701.

[0120] S305. For any remaining trajectory point, determine the matching information corresponding to each road segment.

[0121] After determining the matching road segments corresponding to some trajectory points based on the methods described above, the trajectory points other than those in the initial trajectory point set and those other than those in the target trajectory point set become the remaining trajectory points. Next, we will determine the corresponding matching road segments for each remaining trajectory point. The processing method for each remaining trajectory point is similar; therefore, we will use any one remaining trajectory point as an example below, and will not elaborate on the processing methods for the remaining trajectory points.

[0122] In this embodiment, the matching information of the remaining trajectory points corresponding to each road segment can be determined based on the trajectory information of the sub-trajectories including the remaining trajectory points in the target road segment and the road segment information of each road segment. The implementation method for determining the matching information is similar to that described above, and will not be repeated here.

[0123] S306. Determine at least one road segment whose matching information meets the advanced threshold condition as the candidate road segment corresponding to the remaining trajectory points.

[0124] In this embodiment, an advanced threshold condition is set for the remaining trajectory points. The advanced threshold condition is similar to the initial threshold condition described above, except that the advanced threshold condition is more lenient than the initial threshold condition described above.

[0125] In one possible implementation, the initial threshold conditions include at least one of the following: angle less than or equal to a preset angle threshold, distance less than or equal to a preset distance threshold, height difference less than or equal to a preset height difference threshold, and the included angle of the driving direction less than or equal to a preset included angle threshold. Since all thresholds in the advanced threshold conditions are greater than those in the initial threshold conditions, the advanced threshold conditions in this embodiment are more lenient, allowing for the selection of more suitable road segments from the remaining trajectory points.

[0126] In this embodiment, at least one road segment whose matching information meets the advanced threshold condition can be identified as a candidate road segment corresponding to the remaining trajectory points. It is understood that because the advanced threshold condition is set relatively leniently, there may be multiple road segments whose matching information meets the advanced threshold condition.

[0127] S307. Determine the matching road segment corresponding to the remaining trajectory points in the candidate road segment.

[0128] Next, the matching road segment corresponding to the remaining trajectory points is determined from the candidate road segments. In one possible implementation, for example, for any candidate road segment, the target probability of the remaining trajectory points matching the candidate road segment can be determined, and then the candidate road segment with the highest target probability can be determined as the matching road segment corresponding to the remaining trajectory points.

[0129] In one possible implementation, when determining the target probability of a match between the remaining trajectory points and the candidate road segments, an implicit Markov matching process can be used. This process calculates the observation probability and transition probability of the candidate road segment corresponding to the remaining trajectory points, and then determines the target probability based on a combination of the observation and transition probabilities. For example, the observation and transition probabilities can be summed or weighted, and the specific implementation method can be chosen according to actual needs.

[0130] The following section further explains how to determine the observation probability and transition probability.

[0131] In this embodiment, the observation probability corresponding to the candidate road segment can be determined, where the observation probability indicates the probability that the remaining trajectory points are located on the candidate road segment. The remaining trajectory points being located on the candidate road segment can be further understood as the remaining trajectory points being located within the road segment area of ​​the candidate road segment.

[0132] In this embodiment, the transition probability corresponding to the candidate road segment can also be determined. The transition probability is used to indicate the probability that the next road segment adjacent to the target road segment is the candidate road segment in the road segment matching result. The target road segment is the matching road segment corresponding to the previous trajectory point adjacent to the remaining trajectory point. The road segment matching result includes the matching road segment corresponding to each trajectory point.

[0133] Here's an example illustrating the transition probability. Suppose there are two consecutive trajectory points in the target trajectory, trajectory point a and trajectory point b. For trajectory point a, it's assumed that the matching road segment is already determined to be road segment 1, and for trajectory point b, three candidate road segments are determined: road segment 2, road segment 3, and road segment 4.

[0134] Taking road segment 2 as an example, the transition probability for road segment 2 refers to the probability that the next road segment after road segment 1 is road segment 2 in the road segment matching results. It can be simply understood as the probability of walking from road segment 1 to road segment 2. For example, you can understand it as having three intersections to choose from after road segment 1; the transition probability for road segment 2 is the probability of taking road segment 2. The transition probabilities for road segments 3 and 4 are similar and will not be elaborated upon here.

[0135] After determining the corresponding matching road segment for each remaining trajectory point, it becomes possible to determine the corresponding matching probability for each trajectory point in the target trajectory.

[0136] In this embodiment, refer to Figure 8It can be further understood that this embodiment actually performs a three-level matching process. First, the first-level matching process filters out several trajectory points that can guarantee the correctness of the matched road segments based on the matching information between the sub-trajectories and road segments corresponding to the trajectory points, and then determines these trajectory points as the initial trajectory point set. Then, in the second-level matching process, the matching road segments corresponding to the target trajectory points are completed based on the trajectory points in the initial trajectory point set. Finally, the third-level matching process is performed. For the remaining trajectory points, a relaxed threshold range is adopted to first determine some candidate road segments that may match them, and then the road segments with a higher probability of matching the target are determined as the matching road segments. This allows for the determination of a corresponding matching road segment for each trajectory point. At the same time, the multi-level matching process described above can achieve the orderly and efficient determination of the matching road segments for trajectory points, while also effectively ensuring the accuracy of the correspondence between trajectory points and matching road segments, thus providing an accurate foundation for subsequent data processing.

[0137] Based on the above description, in this embodiment, after determining the matching road segment corresponding to each trajectory point, it can be stored in a storage unit. Further optimization can be performed during the storage process. The following section will combine... Figures 9 to 10 The implementation method of matching road segments corresponding to stored trajectory points is introduced.

[0138] Figure 9 The flowchart of the trajectory point recognition method provided in the embodiments of this application Figure 3 , Figure 10 This is a schematic diagram illustrating the implementation of the fragmented storage matching relationship provided in an embodiment of this application.

[0139] like Figure 9 As shown, the method includes:

[0140] S901. Based on the generation time of each trajectory point and multiple consecutive time periods, the target trajectory is divided into multiple data segments, where the duration of each time period is a preset duration, and the generation time of the multiple trajectory points included in each data segment belongs to the corresponding time period.

[0141] In this embodiment, multiple consecutive time periods can be determined using a preset duration. Then, based on the generation time of each trajectory point in the target trajectory and the multiple consecutive time periods, the target time is divided into multiple data slices. The number of data slices is the same as the number of time periods; that is, each time period corresponds to a data slice. Each data slice can include multiple trajectory points, and the generation time of each trajectory point in the data slice belongs to the time period corresponding to that data slice.

[0142] For example, it can be combined Figure 10 To understand, such as Figure 10 As shown, assume the current target trajectory includes Figure 10 The trajectory points shown are a to p. Assuming a preset duration of 30 seconds, multiple consecutive time segments are defined in 30-second increments. Then, based on these consecutive time segments and the generation time of each trajectory point, the multiple consecutive trajectory points are sequentially divided. Trajectory points whose generation time falls within a given time segment are identified as the trajectory points included in the data slice corresponding to that time segment.

[0143] For example, if the first time segment is 9:00:00 to 9:00:30, and assuming trajectory point a was generated at 9:00:00, trajectory point d at 9:00:29, and trajectory point e at 9:00:31, then trajectory points a, b, c, and 1 can be grouped into one data slice, corresponding to the first time segment. The determination of other data slices is similar. (Refer to...) Figure 10 For example, obtained Figure 10 The partitioning results shown are data fragment 1, data fragment 2, data fragment 3 and data fragment 4, each of which includes multiple trajectory points.

[0144] In one possible implementation, setting a longer preset duration for the time slots can effectively reduce the number of mapping relationships that need to be stored, but the corresponding query performance will be worse. Conversely, setting a shorter preset duration for the time slots can effectively improve query performance, but will increase the number of mapping relationships that need to be stored. Therefore, the preset duration for the time slots can be determined according to actual needs to achieve a balance between query performance and storage requirements.

[0145] S902. For any data segment, identify multiple trajectory points in the data segment that have the same matching road segment and are continuous in the target trajectory as sub-segments to obtain at least one sub-segment.

[0146] After dividing the data into multiple data segments, the trajectory points that are consecutive in the data segments and correspond to the same matching road segment can be further divided into sub-segments. That is, multiple trajectory points that correspond to the same matching road segment and are consecutive in the target trajectory are identified as sub-segments to obtain at least one sub-segment.

[0147] For example, you can refer to Figure 10To understand this, taking data segment 1 as an example, trajectory points a, b, c, and d all correspond to road segment 1, and these four trajectory points are consecutive within the target trajectory. Therefore, trajectory points a, b, c, and d can be defined as a sub-segment, represented as follows: Figure 10 The sub-fragment 1.1 is shown.

[0148] For example, taking data segment 2 as an example, where trajectory points e and f both correspond to road segment 1, and these two trajectory points are consecutive in the target trajectory, then trajectory points e and f can be defined as a sub-segment, represented as follows: Figure 10 Sub-segment 2.1 is shown. Since trajectory points g and h both correspond to road segment 2, and these two trajectory points are consecutive within the target trajectory, trajectory points g and h can be defined as a single sub-segment, represented as follows: Figure 10 The sub-fragment shown is 2.2. Figure 10 The determination method for the remaining sub-segments is similar and will not be repeated here.

[0149] S903. For any sub-segment, store a mapping relationship. The mapping relationship is used to indicate the correspondence between multiple trajectory points in the sub-segment and the same matching road segment that they all correspond to.

[0150] Based on the above introduction, it can be determined that the trajectory points in the same sub-segment are continuous in the target trajectory, and these trajectory points all correspond to the same matching road segment. Therefore, for each sub-segment, only one mapping relationship needs to be stored to represent the correspondence between these trajectory points and matching road segments in the sub-segment. The mapping relationship is used to indicate the correspondence between multiple trajectory points in the sub-segment and the same matching road segment that they all correspond to.

[0151] For example, you can refer to Figure 10 To understand this, for sub-segment 1.1, a mapping relationship can be stored. This mapping relationship indicates the correspondence between the four trajectory points and the same matching road segment 1. Compared to the implementation method of storing four mapping relationships for each of the four trajectory points, the storage method in this embodiment can effectively save storage space.

[0152] For example, referencing Figure 10 For sub-segment 2.1, a mapping relationship can be stored. This mapping relationship indicates the correspondence between two trajectory points (trajectory point e and trajectory point f) and the same matching road segment 1. Compared to the implementation method of storing two mapping relationships for each of these two trajectory points, the storage method in this embodiment can effectively save storage space.

[0153] Therefore, in the trajectory point recognition method provided in this application, the generation time of trajectory points is divided into multiple consecutive time periods to obtain multiple data slices. Then, multiple consecutive trajectory points with the same matching road segment in each data slice are determined as a sub-slice. Then, a mapping relationship is stored for each sub-slice, which can realize the merging and storage of common mapping relationships, effectively saving storage space. At the same time, it is understood that if too many mapping relationships are merged and stored, the query efficiency may be low. Therefore, this application first divides multiple trajectory points into data slices with a preset time period, thereby avoiding the situation of merging and storing too many mapping relationships together. Therefore, the implementation method of storing mapping relationships in this application takes into account both saving storage space and ensuring query efficiency, and can effectively provide services for subsequent query requests.

[0154] Next, let's combine... Figure 11 The method for determining the connection relationship between matching road segments and reference road segments is described in further detail. Figure 11 The flowchart of the trajectory point recognition method provided in the embodiments of this application Figure 4 , Figure 12 This is a schematic diagram of the topological relationship of road segments provided in an embodiment of this application.

[0155] like Figure 11 As shown, the method includes:

[0156] S1101. Based on the topological relationship between multiple road segments, determine the road segment set, which includes at least one road segment directly connected to the matching road segment corresponding to the trajectory point.

[0157] In this embodiment, the topological relationship between multiple road segments reflects the specific connection relationship between each road segment. Therefore, based on the topological relationship, at least one road segment directly connected to the matching road segment corresponding to the trajectory point can be determined, thereby obtaining the road segment set.

[0158] For example, you can refer to Figure 12 To understand, Figure 12 The diagram illustrates the topological relationships between eight road segments, from segment 1 to segment 8, where each pair of points represents a road segment. Assume the matching road segment corresponding to the current trajectory point is... Figure 12 Section 5 shown in the diagram, refer to Figure 12 It can be determined that the road segments directly connected to road segment 5 include road segment 1, road segment 2, road segment 3, road segment 6, and road segment 7, and these road segments can constitute a road segment set.

[0159] In one possible implementation, when querying the road segments directly connected to the matching road segments corresponding to the trajectory points based on topological relationships, a depth-first search approach can be used for the query.

[0160] S1102. Remove road segments with the road attribute of ramp from the road segment set to obtain the updated road segment set.

[0161] In this embodiment, the road attribute of the road segment corresponding to the ramp can be set to ramp, because a ramp connects a highway segment and a regular road segment. Furthermore, the specific labeling of a ramp as a highway segment or a regular road segment varies across different maps. Therefore, road segments with the road attribute of ramp will affect the subsequent determination of connection relationships. Further, in this embodiment, road segments with the road attribute of ramp can be removed from the road segment set to obtain an updated road segment set.

[0162] S1103. If a reference road segment exists in the updated road segment set, then the connection relationship between the matching road segment and the reference road segment is determined to be a direct connection.

[0163] Then we can determine whether there is a reference road segment in the updated road segment set. If there is a reference road segment in the updated road segment set, then we can determine that the matching road segment and the reference road segment are directly connected. Therefore, we can determine that the connection relationship between the matching road segment and the reference road segment is a direct connection.

[0164] S1104. If there is no reference road segment in the updated road segment set, then the connection relationship between the matching road segment and the reference road segment is determined to be an indirect connection.

[0165] Alternatively, if there is no reference road segment in the updated road segment set, then it can be determined that the matching road segment does not have any directly connected reference road segments, and therefore the connection relationship between the matching road segment and the reference road segment can be determined to be non-direct connection.

[0166] In this embodiment, by obtaining and matching road segments that are directly connected to each other to obtain a road segment set, then removing ramp road segments from the road segment set, and then determining whether there is a reference road segment in the updated road segment set, the connection relationship between the matching road segment and the reference road segment can be determined simply and effectively.

[0167] Based on the above introduction, the following can be combined with... Figure 13 The overall implementation process of the trajectory point recognition method provided in this application will be further described in sequence. Figure 13 This is a schematic diagram illustrating the execution process of the trajectory point recognition method provided in the embodiments of this application.

[0168] like Figure 13As shown, the road network matching service can determine the matching road segment corresponding to each trajectory point in the road network structure for multiple trajectory points in the target trajectory. Then, the matching information is stored in the data management center in a segmented storage manner. The specific implementation method can be referred to the description of the above embodiment, and will not be repeated here.

[0169] Afterwards, data types can be labeled for trajectory points. Specifically, the matching road segments corresponding to each trajectory point can be queried in the data management center. Then, the connection relationship between the matching road segments corresponding to each trajectory point and the expressway segments in the road network structure can be determined. Based on the connection relationship, the data types corresponding to each trajectory point can be labeled. Then, the data types labeled for the trajectory points can be stored in the data management center.

[0170] Then, based on the data type of each trajectory point, trajectory points of the same data type can be grouped into the same trajectory point set. It's understandable that the type currently labeled for a trajectory point may differ from the type labeled during data collection. Therefore, in this embodiment, the actual data type corresponding to the trajectory point can be labeled, and corresponding trajectory point sets can be obtained for subsequent map processing. For example, a task push method can be used to send the trajectory point set to the corresponding terminal device.

[0171] Specifically, if the data type corresponding to the trajectory point set is highway, then highway map data can be created based on the trajectory point set. Conversely, if the data type corresponding to the trajectory point set is non-highway, then non-highway map data can be created based on the trajectory point set.

[0172] Furthermore, for trajectory points whose labeled data type and collected data type are inconsistent, this embodiment can also store them in the data management center, so that they can be retrieved in the event of a mismatch in the algorithm.

[0173] In summary, the trajectory point recognition method provided in this application can mark each trajectory point, thereby enabling the subsequent creation of different types of trajectory point sets, and then producing corresponding types of map data based on the trajectory point sets, which can effectively ensure the quality of the map data.

[0174] Figure 14 This is a schematic diagram of the trajectory point recognition device provided in an embodiment of this application. Figure 14 As shown, the device 140 includes: a determining module 1401 and a dividing module 1402.

[0175] The determining module 1401 is used to determine the matching road segment corresponding to each of the multiple trajectory points in the target trajectory.

[0176] The determining module 1401 is further configured to determine the connection relationship between the matching road segment and the reference road segment for any of the trajectory points, wherein the road type of the reference road segment is a highway, or the road type of the reference road segment is a non-highway.

[0177] The determining module 1401 is further configured to determine the data type corresponding to the trajectory point based on the connection relationship, wherein the data type is a high-speed type or a non-high-speed type;

[0178] The partitioning module 1402 is used to partition trajectory points of the same data type into the same trajectory point set. The trajectory point set of the high-speed type is used to determine the map data of the high-speed road, and the trajectory point set of the non-high-speed type is used to determine the map data of the non-high-speed road.

[0179] In one possible design, the determining module 1401 is specifically used for:

[0180] Based on the matching information of each trajectory point corresponding to multiple road segments, an initial trajectory point set is determined, wherein the matching information corresponding to each trajectory point in the initial trajectory point set satisfies the initial threshold condition.

[0181] Based on the topological relationship between the multiple road segments and the initial trajectory point set, the matching road segments corresponding to the target trajectory points are completed to obtain the target trajectory point set. The target trajectory points are at least some of the trajectory points other than the trajectory points in the initial trajectory point set among the multiple trajectory points.

[0182] Based on the matching information of the remaining trajectory points corresponding to the plurality of road segments, at least one candidate road segment corresponding to each of the remaining trajectory points is determined. Among the candidate road segments, the matching road segment corresponding to the remaining trajectory point is determined. The remaining trajectory points are the trajectory points other than those in the initial trajectory point set and those other than those in the target trajectory point set.

[0183] In one possible design, the determining module 1401 is specifically used for:

[0184] For any of the trajectory points, a sub-trajectory corresponding to the trajectory point is determined in the target trajectory, and the sub-trajectory includes the trajectory point;

[0185] For any of the road segments, matching information between the sub-trajectory and the road segment is determined based on the trajectory information of the sub-trajectory and the road segment information.

[0186] If the matching information between the sub-trajectory and the road segment satisfies the initial threshold condition, then the road segment that satisfies the initial threshold condition is determined to be the matching road segment corresponding to the trajectory point, and the trajectory point is determined to be a trajectory point in the initial trajectory point set.

[0187] In one possible design, the matching information includes at least one of the following: the angle between the sub-track and the road segment, the distance between the sub-track and the road segment, the height difference between the sub-track and the road segment, and the angle of the driving direction between the sub-track and the road segment;

[0188] The initial threshold condition includes at least one of the following: the angle is less than or equal to a preset angle threshold, the distance is less than or equal to a preset distance threshold, the height difference is less than or equal to a preset height difference threshold, and the included angle of the driving direction is less than or equal to a preset included angle threshold.

[0189] In one possible design, the determining module 1401 is specifically used for:

[0190] For any one of the remaining trajectory points, determine the matching information corresponding to each of the road segments;

[0191] At least one road segment whose matching information satisfies the advanced threshold condition is identified as the candidate road segment corresponding to the remaining trajectory points.

[0192] In one possible design, the determining module 1401 is specifically used for:

[0193] For any one of the candidate road segments, determine the observation probability corresponding to the candidate road segment, and the observation probability is used to indicate the probability that the remaining trajectory points are located on the candidate road segment;

[0194] For any of the candidate road segments, determine the transition probability corresponding to the candidate road segment. The transition probability is used to indicate the probability that the next road segment adjacent to the target road segment in the road segment matching result is the candidate road segment. The target road segment is the matching road segment corresponding to the previous trajectory point adjacent to the remaining trajectory point.

[0195] The target probability is determined based on the transition probability and observation probability corresponding to each of the candidate road segments.

[0196] The candidate road segment with the highest target probability is determined as the matching road segment corresponding to the remaining trajectory points.

[0197] In one possible design, the determining module 1401 is specifically used for:

[0198] Based on the topological relationship between the multiple road segments, a road segment set is determined, wherein the road segment set includes at least one road segment directly connected to the matching road segment corresponding to the trajectory point;

[0199] Remove road segments with the road attribute of ramp from the road segment set to obtain the updated road segment set;

[0200] If the reference road segment exists in the updated road segment set, then the connection relationship between the matching road segment and the reference road segment is determined to be a direct connection;

[0201] If the reference road segment is not present in the updated road segment set, then the connection between the matching road segment and the reference road segment is determined to be an indirect connection.

[0202] In one possible design, the processing module 1402 is specifically used for:

[0203] If the connection relationship is a direct connection, then the data type corresponding to the trajectory point is determined to be high-speed type; or,

[0204] If the connection relationship is not a direct connection, then the data type corresponding to the trajectory point is determined to be a non-high-speed type.

[0205] In one possible design, the determining module 1401 is specifically used for:

[0206] If the connection is a direct connection, then the data type corresponding to the trajectory point is determined to be a non-high-speed type; or,

[0207] If the connection relationship is not a direct connection, then the data type corresponding to the trajectory point is determined to be high-speed type.

[0208] The apparatus provided in this embodiment can be used to execute the technical solutions of the above method embodiments. Its implementation principle and technical effects are similar, and will not be described again here.

[0209] Figure 15 A schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application, such as... Figure 15 As shown, the electronic device 150 of this embodiment includes: a processor 1501 and a memory 1502; wherein

[0210] Memory 1502 is used to store instructions executed by the computer;

[0211] The processor 1501 is used to execute computer execution instructions stored in the memory to implement the various steps of the trajectory point recognition method and map creation method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing method embodiments.

[0212] Alternatively, the memory 1502 can be either standalone or integrated with the processor 1501.

[0213] When the memory 1502 is set up independently, the electronics also include a bus 1503 for connecting the memory 1502 and the processor 1501.

[0214] This application also provides a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the trajectory point recognition method executed by the above-mentioned electronic device.

[0215] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0216] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0217] The integrated modules implemented as software functional modules described above can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application.

[0218] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0219] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0220] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0221] The aforementioned storage medium can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage medium can be any available medium accessible to general-purpose or special-purpose computers.

[0222] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0223] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for trajectory point recognition, characterized in that, include: For multiple trajectory points in the target trajectory, the matching road segment corresponding to each trajectory point is determined from the multiple road segments included in the road network structure; For any of the trajectory points, the connection relationship between the matching road segment and the reference road segment is determined, wherein the road type of the reference road segment is a highway or the road type of the reference road segment is a non-highway, and the connection relationship includes direct connection and indirect connection. The direct connection refers to the connection with the reference road segment without passing through an intermediate road segment, and the indirect connection refers to the connection with the reference road segment through at least one intermediate road segment. Based on the connection relationship, the data type corresponding to the trajectory point is determined, and the data type is either high-speed type or non-high-speed type; Trajectory points of the same data type are divided into the same trajectory point set. The trajectory point set of high-speed type is used to determine the map data of high-speed roads, and the trajectory point set of non-high-speed type is used to determine the map data of non-high-speed roads.

2. The method according to claim 1, characterized in that, The step of determining the matching road segment corresponding to each of the multiple trajectory points in the target trajectory includes: Based on the matching information of each trajectory point corresponding to multiple road segments, an initial trajectory point set is determined, wherein the matching information corresponding to each trajectory point in the initial trajectory point set satisfies the initial threshold condition. Based on the topological relationship between the multiple road segments and the initial trajectory point set, the matching road segments corresponding to the target trajectory points are completed to obtain the target trajectory point set. The target trajectory points are at least some of the trajectory points other than the trajectory points in the initial trajectory point set among the multiple trajectory points. Based on the matching information of the remaining trajectory points corresponding to the plurality of road segments, at least one candidate road segment corresponding to each of the remaining trajectory points is determined. Among the candidate road segments, the matching road segment corresponding to the remaining trajectory point is determined. The remaining trajectory points are the trajectory points other than those in the initial trajectory point set and those other than those in the target trajectory point set.

3. The method according to claim 2, characterized in that, The step of determining the initial set of trajectory points based on the matching information of each trajectory point corresponding to multiple road segments includes: For any of the trajectory points, a sub-trajectory corresponding to the trajectory point is determined in the target trajectory, and the sub-trajectory includes the trajectory point; For any of the road segments, matching information between the sub-trajectory and the road segment is determined based on the trajectory information of the sub-trajectory and the road segment information. If the matching information between the sub-trajectory and the road segment satisfies the initial threshold condition, then the road segment that satisfies the initial threshold condition is determined to be the matching road segment corresponding to the trajectory point, and the trajectory point is determined to be a trajectory point in the initial trajectory point set.

4. The method according to claim 3, characterized in that, The matching information includes at least one of the following: the angle between the sub-track and the road segment, the distance between the sub-track and the road segment, the height difference between the sub-track and the road segment, and the angle of the driving direction between the sub-track and the road segment; The initial threshold condition includes at least one of the following: the angle is less than or equal to a preset angle threshold, the distance is less than or equal to a preset distance threshold, the height difference is less than or equal to a preset height difference threshold, and the included angle of the driving direction is less than or equal to a preset included angle threshold.

5. The method according to any one of claims 2-4, characterized in that, The step of determining at least one candidate road segment corresponding to each of the remaining trajectory points based on the matching information corresponding to the multiple road segments includes: For any one of the remaining trajectory points, determine the matching information corresponding to each of the road segments; At least one road segment whose matching information satisfies the advanced threshold condition is identified as the candidate road segment corresponding to the remaining trajectory points.

6. The method according to any one of claims 2-4, characterized in that, The step of determining the matching road segment corresponding to the remaining trajectory points in the candidate road segments includes: For any one of the candidate road segments, determine the observation probability corresponding to the candidate road segment, and the observation probability is used to indicate the probability that the remaining trajectory points are located on the candidate road segment; For any of the candidate road segments, determine the transition probability corresponding to the candidate road segment. The transition probability is used to indicate the probability that the next road segment adjacent to the target road segment in the road segment matching result is the candidate road segment. The target road segment is the matching road segment corresponding to the previous trajectory point adjacent to the remaining trajectory point. The target probability is determined based on the transition probability and observation probability corresponding to each of the candidate road segments. The candidate road segment with the highest target probability is determined as the matching road segment corresponding to the remaining trajectory points.

7. The method according to any one of claims 2-4, characterized in that, Determining the connection relationship between the matching road segment and the reference road segment for any of the trajectory points includes: Based on the topological relationship between the multiple road segments, a road segment set is determined, wherein the road segment set includes at least one road segment directly connected to the matching road segment corresponding to the trajectory point; Remove road segments with the road attribute of ramp from the road segment set to obtain the updated road segment set; If the reference road segment exists in the updated road segment set, then the connection relationship between the matching road segment and the reference road segment is determined to be a direct connection; If the reference road segment is not present in the updated road segment set, then the connection between the matching road segment and the reference road segment is determined to be an indirect connection.

8. The method according to claim 7, characterized in that, The road type of the reference road segment is a highway; determining the data type corresponding to each trajectory point based on the connection relationship includes: If the connection relationship is a direct connection, then the data type corresponding to the trajectory point is determined to be high-speed type; or, If the connection relationship is not a direct connection, then the data type corresponding to the trajectory point is determined to be a non-high-speed type.

9. The method according to claim 7, characterized in that, The road type of the reference road segment is a non-expressway; determining the data type corresponding to each trajectory point based on the connection relationship includes: If the connection is a direct connection, then the data type corresponding to the trajectory point is determined to be a non-high-speed type; or, If the connection relationship is not a direct connection, then the data type corresponding to the trajectory point is determined to be high-speed type.

10. A map production method, characterized in that, include: The method according to any one of claims 1 to 9 determines the set of trajectory points; If the data type corresponding to the trajectory point set is high-speed, then map data of the high-speed road is created based on the trajectory point set. If the data type corresponding to the trajectory point set is non-highway type, then map data for non-highway roads is created based on the trajectory point set.

11. A trajectory point recognition device, characterized in that, include: The determination module is used to determine the matching road segment corresponding to each trajectory point in the target trajectory among multiple road segments included in the road network structure. The determining module is further configured to determine the connection relationship between the matching road segment and the reference road segment for any of the trajectory points, wherein the road type of the reference road segment is a highway or the road type of the reference road segment is a non-highway, and the connection relationship includes direct connection and indirect connection. The direct connection refers to the connection with the reference road segment without passing through an intermediate road segment, and the indirect connection refers to the connection with the reference road segment through at least one intermediate road segment. The determining module is further configured to determine the data type corresponding to the trajectory point based on the connection relationship, wherein the data type is a high-speed type or a non-high-speed type; The partitioning module is used to divide trajectory points of the same data type into the same trajectory point set. The trajectory point set of the high-speed type is used to determine the map data of the high-speed road, and the trajectory point set of the non-high-speed type is used to determine the map data of the non-high-speed road.

12. An electronic device, characterized in that, include: Memory, used to store programs; A processor for executing the program stored in the memory, wherein when the program is executed, the processor is configured to perform the method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, Includes instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 10.

Images