A road network trajectory matching method, device and equipment

Abstract

The application discloses a kind of road network trajectory matching method, device and equipment, the method can include: based on the corresponding relation of single-trip trajectory data and road network data, in road network data, determine the sub-road network data corresponding to single-trip trajectory data;Sub-road network data is split based on split length threshold value, and topological relationship reconstruction is carried out to the sub-road network data after splitting;Based on hidden Markov model, the trajectory point included in single-trip trajectory data and the reconstructed sub-road network data are matched to realize road network trajectory matching.The method starts from road network data itself, reconstructs road network structure and topological relationship, and then constructs the topological relationship of sub-road network data more suitable for trajectory point state, so as to improve the upper limit of HMM algorithm matching ability, improve the accuracy of map matching.

Description

Technical Field

[0001] This invention relates to the field of geographic information technology, and in particular to a method, apparatus, and device for road network trajectory matching. Background Technology

[0002] Map matching refers to the process of associating a series of ordered initial positioning results with the traffic network of an electronic map. It involves matching the trajectory points of a vehicle during its journey to suitable road segments within the network. Its main purposes are to correct positioning results, calculate traffic planning and navigation, and perform traffic information statistics and analysis. Among existing map matching methods, Hidden Markov Models (HMMs) are frequently used for matching.

[0003] The existing map matching method includes the following steps: First, for each trajectory point, candidate roads are searched within a certain range to generate the possible road network segment locations to which it can be matched, i.e., candidate points; then, based on Hidden Markov Models (HMMs), Gaussian distribution probability is used to describe the degree of matching between the trajectory point and the candidate points to calculate the observation probability; based on HMMs, the probability of moving from one candidate point to the next candidate point is described by the shortest path or geometric distance in the candidate road network segments to calculate the state transition probability; finally, the Viterbi algorithm is used to find the state sequence of the trajectory matching path with the highest probability, thereby mapping the trajectory point to the road network location, thus realizing the matching of trajectory points to the road network. Summary of the Invention

[0004] To improve the compatibility between road network data and trajectory point status, thereby enhancing the accuracy of observation probability and ultimately improving map matching accuracy, this invention provides a road network trajectory matching method, apparatus, and device.

[0005] In a first aspect, embodiments of the present invention provide a road network trajectory matching method, which may include:

[0006] Based on the correspondence between single-trip trajectory data and road network data, sub-road network data corresponding to the single-trip trajectory data is determined in the road network data;

[0007] The sub-network data is segmented based on a segmentation length threshold, and the topological relationship of the segmented sub-network data is reconstructed.

[0008] Based on the Hidden Markov Model, the trajectory points included in the single-trip trajectory data are matched with the reconstructed sub-road network data to achieve road network trajectory matching.

[0009] In one embodiment, segmenting the sub-network data based on a segmentation length threshold and reconstructing the topology of the segmented sub-network data may include:

[0010] The sub-network data is segmented based on the segmentation length threshold to obtain several sub-segment point sets;

[0011] Based on the geographical location information of the sub-road segments, the topological structure, spatial relationships, and correspondences of the sub-road segments with the original road network segments are determined in order to reconstruct the topological relationships of the point set of the sub-road segments.

[0012] In another embodiment, before segmenting the sub-network data based on the segmentation length threshold, the method may further include: determining the segmentation length threshold based on the trajectory point spacing between trajectory points included in the single-trip trajectory data.

[0013] In another embodiment, determining the segmentation length threshold based on the trajectory point spacing between trajectory points included in the single-trip trajectory data may include:

[0014] Based on the distance between trajectory points included in the single-trip trajectory data, the sparsity of the trajectory points and / or the degree of change in the heading angle of the trajectory points, at least one segmentation length threshold is determined.

[0015] In another embodiment, the matching of trajectory points included in the single-trip trajectory data and the reconstructed sub-road network data based on the Hidden Markov Model to achieve road network trajectory matching may include:

[0016] The candidate distance of the Hidden Markov Model is set based on the overall offset distance of the road network between the trajectory data and the road network data;

[0017] Based on the candidate distance, a candidate sub-segment search is performed for each trajectory point in the single-trip trajectory data, and the observation probability and state transition probability of the candidate sub-segment are determined based on the hidden Markov model.

[0018] Based on the observation probability and state transition probability of the trajectory point and the candidate sub-road segment, the matching relationship between the trajectory point and the candidate sub-road segment is determined to achieve road network trajectory matching.

[0019] In another embodiment, after determining the matching relationship between trajectory points and candidate sub-road segments, the process may further include:

[0020] Based on the correspondence between the sub-road segments included in the reconstructed sub-road network data and the original road network, the road segment information of the original road network is matched with the trajectory points.

[0021] In another embodiment, determining the sub-road network data corresponding to the single-trip trajectory data in the road network data based on the correspondence between single-trip trajectory data and road network data may include:

[0022] Line data corresponding to the single-trip trajectory data is constructed based on the trajectory points included in the single-trip trajectory data;

[0023] Using the overall offset distance between the trajectory data and the road network data as the expansion radius, the line data is expanded into area data;

[0024] The sub-road network data is determined based on the overlap relationship between the surface data and the road network data.

[0025] In another embodiment, the method may further include: grouping the trajectory data based on the vehicle ID and / or trajectory route number in the trajectory data to determine single-trip trajectory data.

[0026] In another embodiment, the method may further include: if the trajectory points included in the single-trip trajectory data and the reconstructed sub-road network data fail to match; then, the unmatched trajectory points and trajectory data are counted, and feedback is given on the topological relationship of the road network data based on the statistical results.

[0027] Secondly, embodiments of the present invention provide a road network trajectory matching device, which may include:

[0028] The determination module is used to determine the sub-road network data corresponding to the single-trip trajectory data in the road network data based on the correspondence between single-trip trajectory data and road network data.

[0029] The reconstruction module is used to segment the sub-network data based on the segmentation length threshold, and reconstruct the topology of the segmented sub-network data;

[0030] The matching module is used to match the trajectory points included in the single-trip trajectory data with the reconstructed sub-road network data based on the Hidden Markov Model, so as to achieve road network trajectory matching.

[0031] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the road network trajectory matching method as described in the first aspect.

[0032] Fourthly, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the road network trajectory matching method as described in the first aspect.

[0033] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0034] This invention provides a road network trajectory matching method, apparatus, and device. The method starts from the road network data itself, reconstructs the road network structure and topological relationships, and then constructs a sub-road network data topological relationship that is more suitable for the trajectory point state, thereby improving the upper limit of the HMM algorithm's matching capability and increasing the accuracy of map matching.

[0035] Furthermore, the reconstructed sub-network data topology in this method is high-density sub-network data. This high-density sub-network data provides more candidate points and paths. For the Viterbi algorithm (used to find the most probable state path), it provides more candidate road segments. More candidate paths mean higher matching flexibility and accuracy, making the observed state corresponding to each trajectory point more accurate. Secondly, it can improve the initial state distribution, which has a direct impact on the overall matching quality of the HMM. Finally, candidate road segments in the high-density network are closer to the trajectory points, thereby reducing observation errors (i.e., the distance between the trajectory point and the candidate point). This makes the observation probability of the HMM more accurate and improves the robustness of the algorithm.

[0036] Furthermore, high-density sub-network data provides more detailed sub-network segments, capable of capturing subtle changes and complex structures within the road. Therefore, state transition probabilities can more specifically reflect the actual path migration, improving matching accuracy. Moreover, compared to existing low-density network data, longer road segments simplify many potential transition paths, introducing errors. For the Viterbi algorithm, by improving the accuracy of each state transition and observation probability, accumulated errors are reduced, and the final optimal path is closer to the actual trajectory.

[0037] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0038] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0039] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0040] Figure 1 This is a flowchart of the road network trajectory matching method provided in the embodiments of the present invention;

[0041] Figure 2 This is a flowchart of the detailed road network trajectory matching method provided in the embodiments of the present invention;

[0042] Figure 3 This is a schematic diagram illustrating the determination of sub-network data provided in an embodiment of the present invention;

[0043] Figure 4 This is a schematic diagram of the sub-network data topology reconstruction provided in an embodiment of the present invention;

[0044] Figure 5 This refers to the matching result of trajectory points on the road network provided in the embodiments of the present invention;

[0045] Figure 6 This is a schematic diagram of the road network trajectory matching device provided in an embodiment of the present invention. Detailed Implementation

[0046] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0047] The inventors discovered in practical work that although the Hidden Markov Model (HMM) method can achieve the matching process from trajectory points to road networks, it has certain limitations in complex scene matching. Firstly, the observation probability of an HMM is often calculated based on the distance between a trajectory point and a candidate point, depending on the distance or number of corresponding candidate points on the road network. Especially for the initial observation probability, its accuracy can affect the matching status of subsequent trajectory points to some extent (for example, if the road network segments are dense, a trajectory point may correspond to multiple road network segments, meaning each road network segment will have one candidate point). Secondly, the state transition probability describes the probability of transitioning from one state (candidate road segment) to another state, depending on the state characteristics of the candidate road segment, such as intersections or curves. Appropriate representation of the road network state is crucial. Therefore, the matching characteristics of the HMM method mean that the degree of fit between the road network data and the trajectory point state affects the overall matching accuracy when calculating the probability model. To address the problem that the accuracy of probability calculation based on HMM in map matching depends on road network data, this invention is proposed to provide a road network trajectory matching method, apparatus, and device that overcomes or at least partially solves the above problems.

[0048] This invention provides a road network trajectory matching method, referring to... Figure 1 As shown, the method may include the following steps:

[0049] Step S11: Based on the correspondence between single-trip trajectory data and road network data, determine the sub-road network data corresponding to the single-trip trajectory data in the road network data.

[0050] Step S12: Divide the sub-network data based on the segmentation length threshold, and reconstruct the topology of the segmented sub-network data.

[0051] Step S13: Based on the Hidden Markov Model, match the trajectory points included in the single-trip trajectory data with the reconstructed sub-road network data to achieve road network trajectory matching.

[0052] The road network trajectory matching method provided in this embodiment of the invention starts from the road network data itself, reconstructs the road network structure and topological relationship, and then constructs a sub-road network data topological relationship that is more suitable for the trajectory point state, thereby improving the upper limit of the HMM algorithm matching ability and improving the accuracy of map matching.

[0053] In one specific embodiment, this invention provides a detailed road network trajectory matching method. Since the distribution of trajectory data varies significantly across different scenarios, different road network data is required for road network trajectory matching in different scenarios. The method provided in this embodiment can perform road network trajectory matching for special types of scenarios such as U-turns, ring roads, turns, and highways / elevated bridges. This involves reconstructing the road network data corresponding to the scenario to form new sub-road network data, and then using this new sub-road network data as a benchmark for road network trajectory matching. (Refer to...) Figure 2 As shown, the method may specifically include the following steps:

[0054] Step S20: Based on the vehicle ID and / or route number in the trajectory data, group the trajectory data to determine the single-trip trajectory data. This step groups the trajectory data. By using the vehicle ID and / or route number in the trajectory data, we can obtain trajectory data that is continuous in time and can describe the state of one trajectory journey, i.e., single-trip trajectory data. In subsequent processing steps, this single-trip trajectory data is matched with the road network data, avoiding data interference when multiple trajectory data are matched simultaneously and improving matching accuracy.

[0055] It should be noted that in this embodiment of the invention, there are two ways to perform road network trajectory matching for each single-trip trajectory data. Method one is to perform road network trajectory data matching for a portion of the aforementioned special scenario within each single-trip trajectory data, while other areas in the single-trip trajectory data not within the aforementioned special scenario can be matched using conventional techniques. Method two is to perform road network trajectory matching for the entire single-trip trajectory data based on the method provided in this embodiment. Of course, those skilled in the art will understand that, regardless of whether method one or method two is used, the specific method employed in matching road network trajectory data embodies the inventive concept of this invention, namely, the central idea of ​​reconstructing sub-road network data. In steps S21 to S24 below, the processing method for matching road networks corresponding to special scenarios and road networks corresponding to non-special scenarios is the same; only the road networks of different scenarios are distinguished when determining the segmentation length threshold. Those skilled in the art should not misunderstand this.

[0056] For example, in Method 1 above, it is necessary to identify special scenarios and non-special scenarios for each single trajectory data, as follows:

[0057] (1) U-turn scenario: If the heading angle changes significantly within a short period of time in the trajectory data, and the maximum difference in heading angle within this time range is close to 180 degrees, it is considered a U-turn scenario. Since the speed decreases before the U-turn and gradually increases after the U-turn, the trajectory data distribution in this U-turn scenario may be relatively dense, especially near the U-turn point.

[0058] (2) Loop Scene: If the trajectory heading angle changes continuously with small angles within a continuous time range in the identified trajectory data, it is considered a loop scene. In this loop scene, the trajectory speed is stable but relatively slow, and the data is evenly distributed.

[0059] (3) High-speed / elevated bridge scenario: The heading angle changes are small and gentle in the identified trajectory data, and the trajectory speed is relatively high. It is considered to be an elevated bridge or high-speed scenario. In this high-speed / elevated bridge scenario, the trajectory distribution is uniform and sparse.

[0060] (4) Turning scenario: Identify the position where the heading angle of the trajectory data changes by nearly 90 degrees and consider this position as the entry point of the trajectory turn. During the turn, the trajectory speed slows down and then accelerates, and the acceleration position is considered as the exit point of the trajectory turn. In this turning scenario, the trajectory distribution is relatively dense between the entry point and the exit point.

[0061] (5) Except for the special scenarios mentioned in (1) to (4) above, the other scenarios can be considered as non-special scenarios.

[0062] Step S21: Based on the correspondence between single-trip trajectory data and road network data, determine the sub-road network data corresponding to the single-trip trajectory data in the road network data.

[0063] In the specific implementation of this step, firstly, line data corresponding to the single-trip trajectory data is constructed based on the trajectory points included in the single-trip trajectory data; then, the line data is expanded into area data with the overall offset distance between the trajectory data and the road network data as the expansion radius; finally, the sub-road network data is determined based on the overlap relationship between the area data and the road network data.

[0064] In a specific example, the trajectory point data for each trip is made into a line. A reasonable threshold is set, and the line is expanded into a surface with the offset distance as the threshold width. The surface retains the shape of the original line. Road network data is obtained by calculating whether the expanded surface intersects with the original road network, resulting in the sub-road network used for subsequent matching. (See reference...) Figure 3 As shown, trajectory points in each trajectory data are converted into line data. A reasonable threshold is set, and the lines are expanded into surfaces with a width equal to the threshold, while retaining the shape of the original lines. The threshold size can be selected based on the deviation distance of all trajectory data from the overall road network, aiming to ensure that the expanded surface encompasses road network data within a certain range. However, the threshold should not be too large, as the included road network is used as a sub-road network for trajectory matching, and road networks far from the trajectory data are not considered in the matching process. Road network data is obtained by calculating whether the expanded surface intersects with the original road network (overlapping relationship), thus obtaining the sub-road network data used for subsequent matching. Figure 3 (The red road network in China).

[0065] It should be noted that the method for determining the overlap relationship between surface data and road network data can be based on the geographic information (coordinates) data contained in the attributes of the surface data and road network data themselves. For example, the surface data and road network data can be unified under the same coordinate system for judgment. Those skilled in the art can implement this in existing ways, and the embodiments of the present invention will not be described in detail here.

[0066] It should also be noted that the offset distance is an empirical value, dependent on the overall trajectory data. The offset distance of a single trajectory relative to the road network may have special cases / localities. The macroscopic offset distance between the overall trajectory data and the road network is a value that can describe the distance from the trajectory data to the corresponding road network as a whole. Currently, there is no specific operation for obtaining this value; it is merely an empirical value based on observation. In this embodiment of the invention, the offset distance between each trajectory and the road network data is used to expand the scope before performing reconstruction, matching, and other operations to achieve cyclical traversal of the overall trajectory data.

[0067] Step S22: Divide the sub-network data based on the segmentation length threshold, and reconstruct the topology of the segmented sub-network data.

[0068] In practice, this step first involves dividing the sub-road network data based on a segmentation length threshold to obtain several sub-road segment point sets. Then, based on the geographical location information of the sub-road segments, the topological structure, spatial relationships, and correspondence with the original road network segments are determined to reconstruct the topological relationships of the sub-road segment point sets.

[0069] In this step, road segments in the road network data are divided into more sub-segments according to the segmentation length threshold. This means that different scenarios identified in the trajectory data will be reconstructed using different methods when reconstructing the sub-road network data in the subsequent topology relationship.

[0070] In an optional embodiment, before segmenting the sub-network data based on the segmentation length threshold, the method further includes: determining the segmentation length threshold based on the trajectory point spacing between trajectory points included in a single-trip trajectory data. Specifically, at least one segmentation length threshold is determined based on the trajectory point spacing between trajectory points included in a single-trip trajectory data, as well as the sparsity of the trajectory points and / or the degree of change in the heading angle of the trajectory points.

[0071] Trajectory data is grouped according to the aforementioned special scenarios, and sub-road networks are reconstructed using different methods based on these scenarios, primarily reflected in the difference in sub-network division distance. In scenarios with dense trajectory distribution, the division distance is smaller, while in scenarios with sparse trajectory distribution, the division distance is larger. The division principle is to ensure that each trajectory point corresponds to a different road network segment as much as possible, while minimizing the number of trajectory points corresponding to the same road network segment. This allows for a more accurate and detailed description of the path changes corresponding to each trajectory point. In this embodiment, the sparsity of trajectory points and / or the degree of change in the heading angle of trajectory points fully consider the changes in trajectory points in special scenarios, thereby generating different segmentation length thresholds for different special scenarios. These segmentation length thresholds, after segmentation, ensure that each trajectory point corresponds to one road network segment as much as possible, and that no multiple trajectory points correspond to the same road network segment.

[0072] After obtaining the point sets of the divided sub-road segments, the corresponding topological structure, spatial information, original road network information, and number of sub-road segments are obtained. This yields a topological relationship network of the associated sub-road segment point sets and their geographic coordinates for each sub-road segment point set. High-density road networks can describe complex road network structures, such as intersections and roundabouts, in more detail, better capturing the details of these complex structures. (Refer to...) Figure 4 As shown, each sub-segment point set corresponds to the preceding and following topological information of this segment, thereby obtaining the reconstructed sub-network data.

[0073] The reconstructed sub-network topology in this method is high-density sub-network data, which provides more candidate points and paths. For the Viterbi algorithm (used to find the most probable state path), it provides more candidate road segments. More candidate paths mean higher matching flexibility and accuracy, making the observed state corresponding to each trajectory point more accurate. Secondly, it can improve the initial state distribution, which has a direct impact on the overall matching quality of the HMM. Finally, candidate road segments in the high-density network are closer to the trajectory points, thereby reducing observation error (i.e., the distance between the trajectory point and the candidate point). This makes the observation probability of the HMM more accurate and improves the robustness of the algorithm.

[0074] Step S23: Based on the Hidden Markov Model, match the trajectory points included in the single-trip trajectory data with the reconstructed sub-road network data to achieve road network trajectory matching.

[0075] This step uses the Hidden Markov Model (HMM) to match trajectory point data with the reconstructed sub-network, setting candidate distances for calculating observation probabilities by searching for candidate road segments. The candidate distance should be similar to the threshold (offset distance) set in step S21 above when obtaining the sub-network, ensuring that a large number of closely spaced road segments are found, ultimately resulting in... Figure 5 The matching results of the trajectory points on the road network are shown. In this embodiment, the offset distance is used as the candidate distance, which can ensure that while obtaining the road segment corresponding to the trajectory point (the trajectory point can search the sub-road network based on this distance, and thus search the road network segment corresponding to each trajectory point), it does not involve road segments that are too far away.

[0076] Specifically, firstly, candidate distances for the Hidden Markov Model (HMM) are set based on the overall offset distance between the trajectory data and the road network data. Then, candidate sub-segments are searched for each trajectory point in the single-trip trajectory data based on the candidate distances, and the observation probability and state transition probability of the candidate sub-segments are determined based on the HMM. Finally, the matching relationship between the trajectory point and the candidate sub-segments is determined based on the observation probability and state transition probability of the trajectory point and the candidate sub-segments to achieve road network trajectory matching.

[0077] It should be noted that the HMM algorithm is not modified in this embodiment of the invention. However, the road network is reconstructed in this embodiment of the invention, which improves the accuracy of the state transition probabilities and observation probabilities used in the HMM algorithm.

[0078] In this step, high-density sub-network data provides more detailed sub-network segments, capable of capturing subtle changes and complex structures within the roads. Therefore, during road network trajectory matching, the state transition probabilities can more specifically reflect the actual path migration, improving matching accuracy. Furthermore, compared to existing low-density road network data, longer road segments simplify many potential transition paths, introducing errors. For the Viterbi algorithm, by improving the precision of each state transition and observation probability, accumulated errors are reduced, and the final optimal path is closer to the actual trajectory.

[0079] In an optional embodiment, after determining the matching relationship between trajectory points and candidate sub-road segments, the method may further include: matching the original road network segment information with the trajectory points based on the correspondence between the sub-road segments included in the reconstructed sub-road network data and the original road network. In this embodiment, the matched sub-road segment information is traced back to the original road network information, and the corresponding original road network segment information is added to the trajectory point information, thereby achieving the final road network trajectory data matching. In this embodiment, the matched sub-road segments are associated with the original road network information, which can determine which part of the original road network the matched sub-road segment belongs to. The final trajectory point matching information should reflect the road segment information matched to the original road network, rather than the reconstructed sub-road network in this embodiment.

[0080] Step S24: If the trajectory points included in the single-trip trajectory data and the reconstructed sub-road network data do not match successfully, then count the unmatched trajectory points and trajectory data, and provide feedback on the topological relationship of the road network data based on the statistical results.

[0081] In this embodiment of the invention, step S24 can be seen as another purpose of matching. During the map matching process, trajectory points can be mapped to road network segments to collect information on the matched road network segments, such as traffic flow on a road segment during a certain time period. Trajectory data that does not match a road network segment during this matching process can indirectly reflect newly added road information. Many trajectories pass through this location, but there is no corresponding road network segment, indicating that the road network needs to be updated. Specifically, this can be divided into two types: ① If map data can be matched based on the trajectory data, sub-road network data is obtained, and the sub-road network data is segmented and reconstructed; ② If map data cannot be matched, it indicates that this part is new data and needs to be updated into the main map data, which is another application of trajectory data matching in another situation.

[0082] During HMM matching, some trajectories may fail to find corresponding sub-road networks, or some trajectory data may lose matching information at certain locations. These trajectories are considered to have no matching road network segment within a defined range, and the geographical location corresponding to the trajectory path is a suspected newly added road segment. All unmatched data in this trajectory data is collected and statistically analyzed, which can be used to describe the suspected newly added locations of the road network or the topological relationship of newly added locations. In specific implementation, if a trajectory point fails to find a corresponding candidate point or the entire trajectory fails to find a corresponding road network segment, these trajectory points are marked with a matching status of 0. After the entire matching process is completed, all trajectory data with a matching status of 0 are collected. The location of the data can be used to determine the suspected location of a newly added road network segment. If the number of trajectory data passing through that location exceeds a set threshold, the confidence level of that location as newly added is considered high. By obtaining which trajectory data the unmatched trajectory point belongs to, the start and end positions of the unmatched trajectory where the matching relationship was lost can be determined, thus obtaining the topological relationship of the newly added road network segment and achieving the goal of identifying newly added road network segments through trajectory matching.

[0083] Based on the same inventive concept, this invention also provides a road network trajectory matching device, referring to... Figure 6 As shown, the device may include a determining module 11, a reconstructing module 12, and a matching module 13, and its working principle is as follows:

[0084] The determining module 11 is used to determine the sub-road network data corresponding to the single-trip trajectory data in the road network data based on the correspondence between the single-trip trajectory data and the road network data;

[0085] The reconstruction module 12 is used to segment the sub-network data based on the segmentation length threshold, and to reconstruct the topology of the segmented sub-network data;

[0086] The matching module 13 is used to match the trajectory points included in the single-trip trajectory data with the reconstructed sub-road network data based on the Hidden Markov Model, so as to achieve road network trajectory matching.

[0087] In one embodiment, the reconstruction module 12 is specifically used for:

[0088] The sub-network data is segmented based on the segmentation length threshold to obtain several sub-segment point sets;

[0089] Based on the geographical location information of the sub-road segments, the topological structure, spatial relationships, and correspondences of the sub-road segments with the original road network segments are determined in order to reconstruct the topological relationships of the point set of the sub-road segments.

[0090] In another embodiment, the reconstruction module 12 can also be used to: determine the segmentation length threshold based on the trajectory point spacing between trajectory points included in the single-pass trajectory data.

[0091] In a specific embodiment, the reconstruction module 12 can also be used to: determine at least one segmentation length threshold based on the trajectory point spacing between trajectory points included in the single trajectory data, the sparsity of the trajectory points and / or the degree of change in the heading angle of the trajectory points.

[0092] In another embodiment, the matching module 13 is specifically used for:

[0093] The candidate distance of the Hidden Markov Model is set based on the overall offset distance of the road network between the trajectory data and the road network data;

[0094] Based on the candidate distance, a candidate sub-segment search is performed for each trajectory point in the single-trip trajectory data, and the observation probability and state transition probability of the candidate sub-segment are determined based on the hidden Markov model.

[0095] Based on the observation probability and state transition probability of the trajectory point and the candidate sub-road segment, the matching relationship between the trajectory point and the candidate sub-road segment is determined to achieve road network trajectory matching.

[0096] In another embodiment, after the matching module 13 determines the matching relationship between the trajectory point and the candidate sub-road segment, the matching module 13 can also be used to: match the original road network segment information with the trajectory point based on the correspondence between the sub-road segments included in the reconstructed sub-road network data and the original road network.

[0097] In another embodiment, the determining module 11 is specifically used for:

[0098] Line data corresponding to the single-trip trajectory data is constructed based on the trajectory points included in the single-trip trajectory data;

[0099] Using the overall offset distance between the trajectory data and the road network data as the expansion radius, the line data is expanded into area data;

[0100] The sub-road network data is determined based on the overlap relationship between the surface data and the road network data.

[0101] In another embodiment, refer to Figure 6 As shown, the device may further include a grouping module 10, which is used to group the trajectory data based on the vehicle ID and / or trajectory route number in the trajectory data to determine single-trip trajectory data.

[0102] In another embodiment, reference is also made to Figure 6As shown, the device may further include: a feedback module 14, which, if the trajectory points included in the single-trip trajectory data and the reconstructed sub-road network data fail to match based on the matching module 13, then the feedback module 14 counts the unmatched trajectory points and trajectory data, and provides feedback on the topological relationship of the road network data based on the statistical results.

[0103] Based on the same inventive concept, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described road network trajectory matching method.

[0104] Based on the same inventive concept, this embodiment of the invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described road network trajectory matching method.

[0105] The principle of the problem solved by the above-mentioned device, medium and related equipment in the embodiments of the present invention is similar to that of the aforementioned road network trajectory matching method. Therefore, its implementation can refer to the implementation of the aforementioned road network trajectory matching method, and the repeated parts will not be described again.

[0106] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0107] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0108] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0109] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0110] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A road network trajectory matching method, characterized in that, include: Based on the correspondence between single-trip trajectory data and road network data, sub-road network data corresponding to the single-trip trajectory data is determined in the road network data; The sub-network data is segmented based on a segmentation length threshold, and the topological relationship of the segmented sub-network data is reconstructed. Based on the Hidden Markov Model, the trajectory points included in the single-trip trajectory data are matched with the reconstructed sub-road network data to achieve road network trajectory matching.

2. The method according to claim 1, characterized in that, The step of segmenting the sub-network data based on a segmentation length threshold and reconstructing the topology of the segmented sub-network data includes: The sub-network data is segmented based on the segmentation length threshold to obtain several sub-segment point sets; Based on the geographical location information of the sub-road segments, the topological structure, spatial relationships, and correspondences of the sub-road segments with the original road network segments are determined in order to reconstruct the topological relationships of the point set of the sub-road segments.

3. The method according to claim 2, characterized in that, Before segmenting the sub-network data based on the segmentation length threshold, the method further includes: determining the segmentation length threshold based on the trajectory point spacing between trajectory points included in the single-trip trajectory data.

4. The method according to claim 3, characterized in that, Determining the segmentation length threshold based on the distance between trajectory points included in the single-trip trajectory data includes: Based on the distance between trajectory points included in the single-trip trajectory data, the sparsity of the trajectory points and / or the degree of change in the heading angle of the trajectory points, at least one segmentation length threshold is determined.

5. The method according to claim 1, characterized in that, The method of matching the trajectory points included in the single-trip trajectory data with the reconstructed sub-road network data based on the Hidden Markov Model to achieve road network trajectory matching includes: The candidate distance of the Hidden Markov Model is set based on the overall offset distance of the road network between the trajectory data and the road network data; Based on the candidate distance, a candidate sub-segment search is performed for each trajectory point in the single-trip trajectory data, and the observation probability and state transition probability of the candidate sub-segment are determined based on the hidden Markov model. Based on the observation probability and state transition probability of the trajectory point and the candidate sub-road segment, the matching relationship between the trajectory point and the candidate sub-road segment is determined to achieve road network trajectory matching.

6. The method according to claim 5, characterized in that, After determining the matching relationship between trajectory points and candidate sub-road segments, the following steps are also included: Based on the correspondence between the sub-road segments included in the reconstructed sub-road network data and the original road network, the road segment information of the original road network is matched with the trajectory points.

7. The method according to claim 1, characterized in that, The step of determining the sub-road network data corresponding to the single-trip trajectory data in the road network data based on the correspondence between single-trip trajectory data and road network data includes: Line data corresponding to the single-trip trajectory data is constructed based on the trajectory points included in the single-trip trajectory data; Using the overall offset distance between the trajectory data and the road network data as the expansion radius, the line data is expanded into area data; The sub-road network data is determined based on the overlap relationship between the surface data and the road network data.

8. The method according to any one of claims 1 to 7, characterized in that, Also includes: Based on the vehicle ID and / or trajectory route number in the trajectory data, the trajectory data is grouped to determine the single-trip trajectory data.

9. The method according to any one of claims 1 to 7, characterized in that, Also includes: If the trajectory points included in the single-trip trajectory data and the reconstructed sub-road network data do not match successfully, then the unmatched trajectory points and trajectory data are counted, and the topological relationship of the road network data is fed back based on the statistical results.

10. A road network trajectory matching device, characterized in that, include: The determination module is used to determine the sub-road network data corresponding to the single-trip trajectory data in the road network data based on the correspondence between single-trip trajectory data and road network data. The reconstruction module is used to segment the sub-network data based on the segmentation length threshold, and reconstruct the topology of the segmented sub-network data; The matching module is used to match the trajectory points included in the single-trip trajectory data with the reconstructed sub-road network data based on the Hidden Markov Model, so as to achieve road network trajectory matching.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the road network trajectory matching method as described in any one of claims 1 to 9.

12. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the road network trajectory matching method as described in any one of claims 1 to 9.

Images