Trajectory-based incremental generation method, device, equipment and medium for road network

By preprocessing and feature extraction, clustering and matching of trajectory data, new lanes are generated, and the problems of large calculation overhead and inaccurate connection relationships of the existing middle-level network extraction method are solved, and high-precision incremental generation of road networks is achieved and map updates.

CN118797377BActive Publication Date: 2024-11-15WUHAN UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411264721.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2024-11-15
Estimated Expiration
2044-09-10

AI Technical Summary

Technical Problem

The existing lane-level road network extraction method based on crowdsourcing trajectory data has a high calculation overhead and relies on a large amount of sample data. The extracted road network connection relationship is not accurate enough to meet the requirements of high-frequency updates.

Method used

By obtaining the original trajectory sequence for preprocessing, calculating motion characteristics and geometric characteristics, random lane-changing trajectory detection, clustering and matching, generating new lanes and updating road network data, improving the accuracy of road network increment generation.

Benefits of technology

It enhances the connection between the trajectory and the road network, improves the accuracy of the incremental generation of the road network, and can update the map in a timely manner.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118797377B_ABST
    Figure CN118797377B_ABST
Patent Text Reader

Abstract

The present application discloses a method, device, electronic device and storage medium for generating a road network increment based on trajectories. The method comprises: obtaining an original trajectory sequence and performing preprocessing; calculating the motion features of the preprocessed trajectory sequence and extracting geometric features, performing random lane change trajectory detection based on the motion features and geometric features, and obtaining a preferred sub-trajectory sequence; obtaining road network data; calculating the single displacement value of the path in the road network data and the corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence; matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence; constructing an inserted road section based on the second sub-trajectory sequence to generate a new lane, and updating the weight of the path corresponding to the first sub-trajectory sequence. The present application can improve the accuracy of road network increment generation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation information technology, and in particular to a method, device, electronic device and storage medium for generating a track-based road network increment. Background Art

[0002] High-precision maps were born out of the demand for autonomous driving technology. As a priori knowledge base and spatial benchmark for autonomous vehicles, they provide data services with high precision, high richness, high dynamics and high reliability. However, domestic roads are long and updated frequently. It is difficult to achieve full coverage and high-frequency updates of high-precision maps for the above roads by relying on highly integrated but limited mobile measurement systems, which restricts its comprehensive implementation and application in the field of autonomous driving. Crowdsourcing provides a new model for high-precision map construction to solve the above problems through group intelligence perception. The crowdsourcing model is based on ubiquitous mobile sensing terminals in urban transportation systems, does not rely on professional mobile measurement systems, and has the characteristics of low cost, wide spatial coverage, and timely updates.

[0003] Vehicle trajectory data records the interaction between vehicles and urban traffic networks from the perspectives of spatial location, time, and mobility patterns. As a "sensor" that interacts most frequently with the urban traffic system, the trajectory data generated by vehicles has a wide coverage, high revisit rate, and is timely. At the same time, high-frequency and high-precision positioning technology provides data support for extracting lane-level road network information based on crowdsourced trajectories. For the analysis and mining of real-time, massive trajectory data, distributed processing, streaming computing, big data analysis, deep learning, etc. provide technical support.

[0004] Existing lane-level road network extraction methods based on crowdsourced trajectory data can be divided into two types. One is to extract lane center points from a large number of trajectory samples based on the probability distribution method and then fit them into lane center lines. This method has high computational overhead and relies on a large amount of sample data. The other is based on the clustering method and is suitable for the extraction and update of lane geometry information. It has high requirements on the sampling frequency and accuracy of data samples. The road network connection relationships extracted by these two methods are not accurate enough and cannot meet the requirements of timely map updates. Summary of the invention

[0005] The embodiments of the present application provide a method, device, electronic device and storage medium for generating incremental road networks based on trajectories, which can improve the accuracy of incremental road network generation.

[0006] The present application provides a method for incrementally generating a road network based on trajectories, including:

[0007] Acquire an original trajectory sequence, and preprocess the original trajectory sequence;

[0008] Calculating motion features of the preprocessed trajectory sequence, extracting geometric features of the preprocessed trajectory sequence, and performing random lane change trajectory detection based on the motion features and the geometric features to obtain a preferred sub-trajectory sequence;

[0009] Obtain road network data;

[0010] Calculating a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence;

[0011] Matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence;

[0012] An inserted road segment is constructed based on the second sub-trajectory sequence to generate a new lane, and a weight of a path corresponding to the first sub-trajectory sequence is updated.

[0013] Furthermore, in the above-mentioned trajectory-based lane-road network incremental generation method, the preprocessing of the original trajectory sequence comprises: trajectory cleaning of the original trajectory sequence;

[0014] Performing trajectory cleaning on the original trajectory sequence includes:

[0015] Get the direction angle sequence corresponding to the original trajectory sequence;

[0016] Calculate a corresponding heading angle sequence according to the original trajectory sequence, calculate an angle difference based on the direction angle sequence and the heading angle sequence, and remove a trajectory sequence whose angle difference exceeds a preset angle threshold;

[0017] The corresponding speed sequence is calculated according to the original trajectory sequence, and the trajectory sequence whose speed exceeds a preset speed threshold is eliminated.

[0018] Furthermore, in the above-mentioned trajectory-based incremental road network generation method, the geometric features include road shape descriptors, and calculating the geometric features of the preprocessed trajectory sequence includes:

[0019] Find the center point of each trajectory sequence;

[0020] Traverse the trajectory points within a range with the trajectory center point as the center and r as the radius, and calculate the relative angle and distance;

[0021] Divide the distance into sections according to a preset angle and a preset length r / 3 to form a plurality of section intervals;

[0022] The number of trajectory points falling into each partition interval is calculated, and the partition interval is encoded based on the number to obtain a road shape descriptor.

[0023] Furthermore, in the above-mentioned trajectory-based incremental generation method for a road network, the motion features at least include the cumulative heading angle change, the start-end heading angle change and the angular velocity change, and the geometric features at least include the road shape descriptor, the distance from both ends of the trajectory sequence to the road boundary and the lateral displacement of the trajectory sequence relative to the road centerline;

[0024] Calculating the motion features of the preprocessed trajectory sequence, extracting the geometric features of the preprocessed trajectory sequence, and performing lane change trajectory detection based on the motion features and the geometric features to obtain a preferred sub-trajectory sequence, including:

[0025] The accumulated heading angle change, the start-end heading angle change, the angular velocity change, the road shape descriptor, the distances from both ends of the trajectory sequence to the road boundary, and the lateral displacement of the trajectory sequence relative to the road centerline are input into a vector machine for binary classification, and random lane change trajectories are eliminated to obtain an optimal sub-trajectory sequence.

[0026] Furthermore, the above-mentioned trajectory-based incremental generation method for a road network, wherein a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence is calculated, and the trajectory points of the preferred sub-trajectory sequence are moved based on the single displacement value to cluster the preferred sub-trajectory sequence, comprises:

[0027] Calculate the displacement difference and direction difference of the trajectory points in the preferred sub-trajectory sequence and the trajectory points in the surrounding area of ​​the trajectory points relative to the corresponding points in the path in the corresponding road network data;

[0028] Calculating a single displacement value based on the displacement difference and the direction difference;

[0029] All trajectory points in the preferred sub-trajectory sequence are moved according to the single displacement value to cluster the preferred sub-trajectory sequence.

[0030] Furthermore, in the above-mentioned trajectory-based incremental generation method for a road network, the step of matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence comprises:

[0031] Construct a hidden Markov model, traverse the optimal sub-trajectory sequence after clustering, regard the trajectory points as observations, and the real position or real road section as hidden states;

[0032] The trajectory points in the clustered preferred sub-trajectory sequence are calculated by the following formula: The path to the road network data The observed transition probability :

[0033]

[0034] in, For trajectory points To Path The projection distance on is the observation error, exp is the exponential function, Represents trajectory points To Path The projection point on ;

[0035] The transition probability between paths in the road network data is calculated by the following formula:

[0036]

[0037] in,

[0038] in, It is a breadth-first search algorithm for directed graphs with depth. Represents the search network node Depth

[0039] Find the optimal sequence:

[0040] initialization:

[0041] Recursion yields:

[0042]

[0043] in, , is the length of the observation sequence, S is the observed object, Before the Representative The final state is The probability of the state sequence that the observation result most likely corresponds to, is the probability matrix of the observation result of the first node to the matching sequence, is the probability transfer matrix between objects in the matching sequence, For the front The final state is The probability of the state sequence that the observation result most likely corresponds to, is the probability matrix of the observation result of the tth node to the matching sequence;

[0044] The first sub-trajectory sequence is calculated by the following formula { }:

[0045]

[0046] in, For the function Find parameters.

[0047] An unmatched second sub-trajectory sequence is obtained according to the clustered preferred sub-trajectory sequence and the matched first sub-trajectory sequence.

[0048] Furthermore, in the above-mentioned trajectory-based lane network incremental generation method, the construction of the inserted road segment is performed based on the second sub-trajectory sequence to generate a new lane, and the weight of the path corresponding to the first sub-trajectory sequence is updated, including:

[0049] Solving the path of the road network data by using a breadth-first search algorithm with depth of an undirected graph;

[0050] If the result of the solution is 0, matching the front and rear endpoints of the second sub-trajectory sequence to the path of the road network data, constructing an insertion section, and generating a new lane;

[0051] If the solution result is not 0 and does not meet the matching condition, inserting the second sub-trajectory sequence into the graph structure data corresponding to the road network data;

[0052] The weight of the path corresponding to the first sub-trajectory sequence is updated.

[0053] The embodiment of the present application also provides a track-based lane-road network incremental generation device, comprising:

[0054] A first acquisition module is used to acquire an original trajectory sequence;

[0055] A preprocessing module, used for preprocessing the original trajectory sequence;

[0056] a random lane-changing trajectory detection module, configured to extract motion features of the preprocessed trajectory sequence, calculate geometric features of the preprocessed trajectory sequence, perform random lane-changing trajectory detection based on the motion features and the geometric features, and obtain a preferred sub-trajectory sequence;

[0057] The second acquisition module is used to acquire road network data;

[0058] a clustering module, configured to calculate a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and to move the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence;

[0059] A matching module, configured to match the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence;

[0060] The lane construction and weight updating module is used to construct the inserted road segment based on the second sub-trajectory sequence to generate a new lane, and update the weight of the path corresponding to the first sub-trajectory sequence.

[0061] An embodiment of the present application also provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor to execute any of the above-mentioned trajectory-based incremental road network generation methods.

[0062] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used for the steps in any of the above-mentioned methods for incrementally generating a trajectory-based road network.

[0063] The present application provides a method, device, electronic device and storage medium for generating a lane network increment based on trajectories. The present application extracts a preferred sub-trajectory sequence through random lane change trajectory detection, matches and clusters the preferred sub-trajectory sequence with the path in the road network data, obtains an unmatched second sub-trajectory sequence, and constructs an inserted road section through the second sub-trajectory sequence to achieve the generation of a new lane. The embodiments of the present application enhance the connection between the trajectory and the road network, improve the accuracy of the incremental generation of the road network, and can update the map in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The technical solution and other beneficial effects of the present application will be made apparent by describing in detail the specific implementation methods of the present application in conjunction with the accompanying drawings.

[0065] Figure 1 A schematic flow chart of a method for incrementally generating a track-based road network according to an embodiment of the present application.

[0066] Figure 2 Another flowchart of the trajectory-based lane road network incremental generation method provided in an embodiment of the present application.

[0067] Figure 3 This is a distribution diagram of the original trajectory data provided in an embodiment of the present application.

[0068] Figure 4 This is a distribution diagram of trajectory data after cleaning provided in an embodiment of the present application.

[0069] Figure 5 A schematic diagram comparing the direction angle before cleaning and the heading angle provided in an embodiment of the present application.

[0070] Figure 6 A schematic diagram comparing the direction angle and heading angle after cleaning provided in an embodiment of the present application.

[0071] Figure 7 A schematic diagram of a preprocessed trajectory sequence provided in an embodiment of the present application.

[0072] Figure 8 A schematic diagram of a preferred sub-trajectory sequence after eliminating random lane changes provided in an embodiment of the present application.

[0073] Fig. 9 A schematic diagram of the distribution of trajectory points provided in an embodiment of the present application.

[0074] Fig.10 A schematic diagram of a road shape descriptor for an intersection obtained based on the distribution of trajectory points provided in an embodiment of the present application.

[0075] Fig.11 A schematic diagram of the comparison of trajectories before and after iterative clustering provided in an embodiment of the present application.

[0076] Fig.12 This is a diagram of the structured road map matching effect provided in an embodiment of the present application.

[0077] Fig.13 A schematic diagram of intersection area map matching provided in an embodiment of the present application.

[0078] Fig.14 A schematic diagram of the structure of a trajectory-based lane-road network incremental generation device provided in an embodiment of the present application.

[0079] Fig.15 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0080] Fig.16 Another schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0082] The existing lane-level road network extraction method based on crowdsourced trajectory data can be divided into two steps. The first step is to segment the trajectory data into intersection area trajectories and structured road trajectories through intersection recognition. For the trajectory data in the intersection area, cluster fitting is usually used to complete the construction of the connection relationship. For the lane geometry information extraction in the structured road area, it can be divided into lane geometry information extraction methods based on the local distribution of trajectory points and lane estimation methods based on trajectory clustering algorithms.

[0083] Based on the probability distribution method, the probability model is fitted using the positioning distribution of trajectory points in the road section to solve the position and width of the lane center point, and the specific lane centerline geometry information is obtained through curve fitting. Under the premise that the positioning of the trajectory points in the same lane follows the Gaussian distribution, the number of lanes is iterated, the Gaussian Mixture Model (GMM) is used to model the trajectory point data in the window, and the expectation maximization (EM) algorithm is used to iteratively solve the lane width and center point position. Finally, the chi-square test is used to determine the optimal number of lanes and the location of the lane center point. Some studies use the lane width design specification as a constraint to optimize the iteration strategy to improve the result accuracy and solution rate.

[0084] Based on the clustering method, the trajectory points / trajectory segments belonging to the same lane are clustered by distance measurement, and the corresponding lane centerline is fitted. For example, the optimal solution function for the number of lanes is constructed using the width range, and the constrained K-means clustering algorithm is further combined to cluster the trajectory segments in the same lane. Similarly, some methods use the optimized mean shift clustering algorithm and add post-processing methods to improve the accuracy of the extraction results.

[0085] The probability distribution method is used to extract lane center points from a large number of trajectory samples and then fit them to lane center lines. The computational overhead is large and depends on a large amount of sample data. The clustering method is suitable for extracting and updating lane geometry information, but it has high requirements on the sampling frequency and accuracy of data samples. There is currently no clustering algorithm for scenarios where the number of lanes changes.

[0086] To solve the above problems, the embodiments of the present application provide a method, device, electronic device and storage medium for generating a track-based incremental road network. The embodiment of the present application provides a track-based incremental road network generation device, which can be integrated in an electronic device, which can be a terminal, a server and other devices, wherein the terminal can include a tablet computer, a laptop computer, a personal computer (PC), a micro processing box, or other devices.

[0087] See also Figure 1 and Figure 2 , Figure 1A schematic diagram of a process flow of a method for incrementally generating a track-based road network provided in an embodiment of the present application. Figure 2 Another schematic flow chart of a method for incrementally generating a road network based on trajectories provided in an embodiment of the present application, which is applied to an electronic device, includes the following steps:

[0088] S1, obtain the original trajectory sequence and preprocess the original trajectory sequence.

[0089] In one embodiment, preprocessing the original trajectory sequence includes performing trajectory cleaning on the original trajectory sequence, which specifically includes the following steps:

[0090] S11, performing trajectory cleaning on the original trajectory sequence includes:

[0091] S12, obtaining a direction angle sequence corresponding to the original trajectory sequence.

[0092] S13, calculating a corresponding heading angle sequence according to the original trajectory sequence, calculating an angle difference based on the direction angle sequence and the heading angle sequence, and eliminating trajectory sequences whose angle difference exceeds a preset angle threshold.

[0093] The heading angle refers to the vehicle motion angle calculated by the trajectory, and the direction angle refers to the angle relative to the true north direction measured instantaneously by the device (the inertial measurement unit IMU is used in this embodiment). As an example, the preset angle threshold can be 30°, and the estimation sequence with an angle difference exceeding 30° is eliminated.

[0094] S14, calculating a corresponding speed sequence according to the original trajectory sequence, and eliminating trajectory sequences whose speed exceeds a preset speed threshold.

[0095] As an example, the preset speed threshold may be 30 m / s, and trajectory sequences with an average speed exceeding 30 m / s are removed. Figure 3 The original trajectory data distribution diagram provided in the embodiment of the present application is Figure 4 This is a distribution diagram of the trajectory data after cleaning provided in the embodiment of the present application. Figure 5 This is a schematic diagram comparing the direction angle before cleaning and the heading angle provided in the embodiment of the present application. Figure 6 The above-mentioned effect of cleaning the original trajectory sequence can be referred to as Figure 3-Figure 6 .

[0096] In one embodiment, preprocessing the original trajectory sequence further includes simplification of the original trajectory sequence, which specifically includes:

[0097] Given a trajectory T={P1, P2,…, Pn}, let the direction angle of a trajectory point Pi in the trajectory T be αi, then the direction change of the adjacent point Pi+1 of the trajectory point Pi relative to the trajectory point Pi is δ(Pi+1|Pi), and the direction change of any point Pj (j>i) relative to point Pi is the cumulative sum of the direction changes of any two adjacent points in Pi, Pi+1,…, Pj-1, Pj.

[0098] The variable Ps is used to save the value of the previous point to be identified (the feature point refers to the backbone trajectory point used to represent the trajectory sequence, which can be understood as the retained trajectory point after the trajectory is simplified). The variable Pi is used to save the value of the previous adjacent point (the previous trajectory point of the current trajectory point). For the starting point of the trajectory, it is directly regarded as a feature point. At this time, the Ps variable is assigned to P1, and Pi is assigned to P1. Similar compression processing is continued for the next trajectory point. According to the inference, the cumulative direction change of the current point Pi relative to the previous feature point is equal to the cumulative sum of Ps and Pi. It is only necessary to save the current point, the previous feature point Ps, the previous adjacent point Pi, and the direction change of the current point relative to the previous feature point. And in the same way, the cumulative distance is calculated by variable sampling. Finally, the variable sampling of the trajectory is realized by the two thresholds of cumulative direction 5° and cumulative distance 10m, and the trajectory segments with cumulative directions greater than 5° and cumulative distances greater than 10m are eliminated.

[0099] Furthermore, trajectory segmentation is performed.

[0100] S2, calculating the motion features of the preprocessed trajectory sequence, extracting the geometric features of the preprocessed trajectory sequence, performing random lane change trajectory detection based on the motion features and the geometric features, and obtaining an optimal sub-trajectory sequence.

[0101] In one embodiment, the motion features include at least the cumulative heading angle change, the start-end heading angle change and the angular velocity change, and the geometric features include at least the road shape descriptor, the distance from the two ends of the trajectory sequence to the road boundary and the lateral displacement of the trajectory sequence relative to the center line of the road.

[0102] Figure 7 A schematic diagram of a preprocessed trajectory sequence provided in an embodiment of the present application, Figure 8 For a schematic diagram of a preferred sub-trajectory sequence after eliminating random lane changes provided in an embodiment of the present application, please refer to Figure 7 and Figure 8, step S2 specifically includes: calculating the cumulative heading angle change, the start-end heading angle change and the angular velocity change based on the preprocessed trajectory sequence (i.e., the segmented trajectory), extracting the road shape descriptor based on the preprocessed trajectory sequence, and calculating the distances from both ends of the trajectory sequence to the road boundary and the lateral displacement of the trajectory sequence relative to the road centerline, inputting the above features into a support vector machine for binary classification to eliminate random lane change trajectories and obtain an optimal sub-trajectory sequence.

[0103] In one embodiment, the process of extracting road descriptors may include the following steps:

[0104] Find the center point of each trajectory sequence;

[0105] Traverse the trajectory points within a range with the trajectory center point as the center and r as the radius, and calculate the relative angle and distance;

[0106] Divide the distance into sections according to a preset angle and a preset length r / 3 to form a number of section intervals;

[0107] The number of trajectory points falling into each partition interval is calculated, and the partition interval is encoded based on the number to obtain a road shape descriptor.

[0108] Specifically, the road shape descriptor is used to distinguish the intersection and non-intersection areas. First, the center point of each trajectory is found. , traverse the trajectory points at a certain radius r and calculate the relative angle and distance, which is calculated by the following formula:

[0109]

[0110]

[0111] in, Represents the traversed trajectory points, for and The relative angle between for and The distance between Is the square root.

[0112] like Fig. 9 and Fig.10 As shown, Fig. 9 A schematic diagram of the distribution of trajectory points provided in an embodiment of the present application, Fig.10A schematic diagram of a road shape descriptor for a cross intersection obtained according to the distribution of trajectory points provided in an embodiment of the present application. The distance is divided into 36 partitions every 30° clockwise and every r / 3, and the number of trajectory points falling into each partition is calculated and normalized. When the value of the partition is greater than the preset number threshold, it is encoded as 1, otherwise it is encoded as 0, and a 1×36 vector matrix is ​​obtained, which is the corresponding shape descriptor.

[0113] In one embodiment, the lateral displacement of the trajectory sequence relative to the centerline of the road may include:

[0114] A rectangular buffer is constructed for the trajectory segments in the trajectory sequence. The radius of the rectangular buffer is d. The angle difference between the first and last endpoints of the trajectory segment in the buffer and the direction difference between the first and last endpoints of the trajectory segment to be calculated is calculated. The trajectory segments exceeding the angle threshold are eliminated. The geometric center of the remaining trajectory segments, that is, the road centerline, is calculated. The lateral displacement of the trajectory segment relative to the road centerline is calculated using the following formula:

[0115]

[0116] in, represents the trajectory segment distance between two trajectory points, Represents the angular difference between the trajectory point and its projection to the road centerline.

[0117] S3, obtain road network data.

[0118] Specifically, read the existing road network data and convert the road network data into The form of storage (G represents the graph structure, V represents the node, E represents the edge, and the nodes are connected by the edges).

[0119] S4, calculating the single displacement value of the path in the road network data and the corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence.

[0120] In one embodiment, step S4 includes the following steps:

[0121] S41, calculating the displacement difference and direction difference of the trajectory points in the preferred sub-trajectory sequence and the trajectory points in the surrounding area of ​​the trajectory points relative to the corresponding points in the path in the corresponding road network data.

[0122] Exemplarily, the points in the surrounding area of ​​the trajectory point may be trajectory points whose distance d≤3.75 from the trajectory point.

[0123] S42, calculating a single displacement value based on the displacement difference and the direction difference.

[0124] Specifically, it can be calculated by the following formula:

[0125]

[0126] in, is the displacement difference of the corresponding points, is the direction difference of the corresponding points, is the standard deviation of the distances between corresponding points, is the standard deviation of the heading angles of the corresponding points, is a single displacement value, where the distribution of the displacement difference and the direction difference is estimated by the Gaussian kernel.

[0127] S43, moving all trajectory points in the preferred sub-trajectory sequence according to the single displacement value to cluster the preferred sub-trajectory sequence.

[0128] Fig.11 A schematic diagram of the comparison of trajectories before and after iterative clustering provided in an embodiment of the present application is shown in FIG. Fig.11 As shown, specifically, all trajectory points are iteratively moved according to a single displacement value, wherein each complete traversal of all trajectory points in the preferred sub-trajectory sequence is regarded as one iteration, and the iteration can be stopped when the maximum number of iterations is 10 or the average displacement of the iteration is less than the limit error of 0.1m.

[0129] S5, matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence.

[0130] Fig.12 This is a structured road map matching effect diagram provided in the embodiment of the present application. Fig.13 A schematic diagram of intersection area map matching provided in an embodiment of the present application is shown in FIG. Fig.12 and Fig.13 As shown, the horizontal axis represents the index of each track point in the matching track sequence in the map matching, and the vertical axis represents the matched road network nodes, and the overall representation of the connected road network nodes has connectivity. In one embodiment, step S5 includes the following steps:

[0131] S51, construct a hidden Markov model, traverse the optimal sub-trajectory sequence after clustering, regard the trajectory points as observations, and the real positions or real road sections as hidden states, solve the optimal hidden state sequence, and the solution result is the matching result.

[0132] Specifically, the hidden state S is: the path sequence of the road network;

[0133] The observed state O is: the optimal sub-trajectory sequence after clustering;

[0134] Initial state probability matrix π: initial hidden state S probability matrix;

[0135] Observation state transition probability matrix : The probability relationship between trajectory points and matching road sections;

[0136] Hidden state transition probability matrix : Connections between paths, topological structure of road network.

[0137] Compared with traditional HMM-based map matching, the embodiment of the present application makes the following improvements: missing matching paths in hidden states are allowed, that is, the original full match is now allowed for partial matching of trajectory sequences. This is changed during the iterative solution of the Viterbi algorithm. Considering that the topological relationship at the lane level is more complex than that at the road level, a deep directed graph breadth-first search algorithm is introduced to calculate the hidden state transition probability.

[0138] S52, calculate the trajectory points in the clustered preferred sub-trajectory sequence by the following formula Path to the road network data The observed transition probability :

[0139]

[0140] in, For trajectory points To Path The projection distance on is the observation error, exp is the exponential function, Represents trajectory points To Path The projection point on .

[0141] S53, calculating the transition probability between paths in the road network data by the following formula;

[0142]

[0143] in,

[0144] in, It is a breadth-first search algorithm for directed graphs with depth. Represents the search network node The current depth is 10.

[0145] express If the connection distance between If it is 0, it means There is no connectivity between them. For solving the optimal hidden state sequence, the Viterbi algorithm is used to reduce the computational complexity. The conventional Viterbi algorithm for decoding hidden Markov models does not allow the observed transition probability or hidden state transition probability of a node to be 0. However, for the scenario of updating the road network with trajectory increments, the above situation must exist, that is, there is a problem that there is no candidate path for the trajectory point or the candidate path topological relationship between two trajectory points does not exist. is 0 or When is 0, the Viterbi algorithm is appropriately improved.

[0146] S54, solve the optimal sequence through the Viterbi algorithm:

[0147] initialization:

[0148] Recursion yields:

[0149]

[0150] in, , is the length of the observation sequence, S is the observed object, Before the Representative The final state is The probability of the state sequence that the observation result most likely corresponds to, is the probability matrix of the observation result of the first node to the matching sequence, is the probability transfer matrix between objects in the matching sequence, For the front The final state is The probability of the state sequence that the observation result most likely corresponds to, is the probability matrix of the observation result of the tth node to the matching sequence.

[0151] S55, the first sub-trajectory sequence is calculated by the following formula { }:

[0152]

[0153] in, For the function Find parameters.

[0154] S56 , obtaining an unmatched second sub-trajectory sequence according to the clustered preferred sub-trajectory sequence and the matched first sub-trajectory sequence.

[0155] S6: construct an insertion section based on the second sub-trajectory sequence to generate a new lane, and update the weight of the path corresponding to the first sub-trajectory sequence.

[0156] In one embodiment, step S6 comprises the following steps:

[0157] S61, solving the paths of the road network data using a breadth-first search algorithm with depth in an undirected graph.

[0158] Specifically, the front and rear endpoints of the unmatched second sub-trajectory sequence are matched to the paths of the existing road network. , On, then respectively , Perform a breadth-first search algorithm with depth on an undirected graph. The formulas include:

[0159]

[0160] in, It is a breadth-first search algorithm for undirected graphs with depth. Usually 10.

[0161] S62: If the result of the solution is 0, the front and rear endpoints of the second sub-trajectory sequence are matched to the path of the road network data, and the insertion section is constructed to generate a new lane.

[0162] S63: If the solution result is not 0 and the matching condition is not satisfied, the second sub-trajectory sequence is inserted into the graph structure data corresponding to the road network data.

[0163] That is, if =0, then construct the inserted section. ≠0, then further obtain the returned path node sequence The similarity calculation based on Hausdorff distance is performed with the second sub-trajectory sequence. If the distance exceeds a certain threshold of 2.2m, the beginning and end of the sub-trajectory sequence are compared. , Separately formed , , and insert the second sub-trajectory sequence into .

[0164] S64: Update the weight of the path corresponding to the first sub-trajectory sequence.

[0165] Specifically, for the matched first sub-trajectory sequence, the weight of the corresponding path is increased.

[0166] The embodiment of the present application first completes the preprocessing of trajectory data by trajectory cleaning, trajectory simplification, and trajectory segmentation. Then, the existing road network data can be input, and on this basis, the motion features of the trajectory sequence are extracted for the segmented trajectory, and the features of the trajectory sequence in spatial geometry are calculated at the same time, and the support vector machine is combined for binary classification to realize the identification of random lane change trajectory segments in random structured areas. After the trajectory segments of random lane changes are eliminated, the preferred sub-trajectory sequence is subjected to mean shift iterative clustering to realize the clustering of trajectory points belonging to the center line of the same lane. Then, lane-level local map matching is performed on the clustered preferred sub-trajectory sequence, and the matching results are clustered to complete the classification of trajectory sequences belonging to the new lane and the old lane. On this basis, the topological relationship of the new lane is judged and constructed to realize the generation of the new lane, and the weight of the old lane is updated. Repeat the above steps until all trajectory sequences are processed and the incremental update of the lane-level road network is completed. The embodiment of the present application can identify the trajectory and match the existing path with the trajectory with the help of road network information to achieve the generation of new lanes, enhance the connection between the trajectory and the road network, improve the accuracy of road network incremental generation, and enable timely update of the map.

[0167] According to the method described in the above embodiment, this embodiment will be further described from the perspective of a trajectory-based vehicle road network incremental generation device. The trajectory-based vehicle road network incremental generation device can be implemented as an independent entity or integrated in an electronic device. The electronic device can be a terminal, a server, etc., wherein the terminal can include a tablet computer, a laptop computer, a personal computer (PC), a micro processing box, or other devices.

[0168] See also Fig.14 , Fig.14 The invention specifically describes a device for generating a lane-road network increment based on a trajectory provided in an embodiment of the present application, which is applied to an electronic device. The device for generating a lane-road network increment based on a trajectory may include:

[0169] A first acquisition module is used to acquire an original trajectory sequence;

[0170] A preprocessing module, used to preprocess the original trajectory sequence;

[0171] A random lane change trajectory detection module is used to extract the motion features of the preprocessed trajectory sequence, calculate the geometric features of the preprocessed trajectory sequence, perform random lane change trajectory detection based on the motion features and the geometric features, and obtain an optimal sub-trajectory sequence;

[0172] The second acquisition module is used to acquire road network data;

[0173] A clustering module, used to calculate the single displacement value of the path in the road network data and the corresponding point of the preferred sub-trajectory sequence, and move the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence;

[0174] A matching module, used for matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence;

[0175] The lane construction and weight updating module is used to construct the inserted road segment based on the second sub-trajectory sequence to generate a new lane and update the weight of the path corresponding to the first sub-trajectory sequence.

[0176] During specific implementation, the above modules and / or units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above modules and / or units can refer to the previous method embodiments. The specific beneficial effects that can be achieved can also refer to the beneficial effects in the previous method embodiments, which will not be repeated here.

[0177] In addition, the embodiment of the present application also provides an electronic device, which may be a computer, a tablet computer, or other device. Fig.15 As shown, the electronic device 400 includes a processor 401 and a memory 402. The processor 401 is electrically connected to the memory 402.

[0178] The processor 401 is the control center of the electronic device 400. It uses various interfaces and lines to connect various parts of the entire electronic device. It executes various functions of the electronic device and processes data by running or loading applications stored in the memory 402 and calling data stored in the memory 402, thereby monitoring the electronic device as a whole.

[0179] In this embodiment, the processor 401 in the electronic device 400 loads instructions corresponding to the processes of one or more application programs into the memory 402 according to the following steps, and the processor 401 runs the application programs stored in the memory 402 to implement various functions:

[0180] Acquire an original trajectory sequence, and preprocess the original trajectory sequence;

[0181] Calculating motion features of the preprocessed trajectory sequence, extracting geometric features of the preprocessed trajectory sequence, and performing random lane change trajectory detection based on the motion features and the geometric features to obtain a preferred sub-trajectory sequence;

[0182] Obtain road network data;

[0183] Calculating a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence;

[0184] Matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence;

[0185] An inserted road segment is constructed based on the second sub-trajectory sequence to generate a new lane, and a weight of a path corresponding to the first sub-trajectory sequence is updated.

[0186] The electronic device can implement the steps in any embodiment of the method for incrementally generating a track-based road network provided in the embodiments of the present application. Therefore, it can achieve the beneficial effects that can be achieved by any method for incrementally generating a track-based road network provided in the embodiments of the present invention. For details, please refer to the previous embodiments and will not be repeated here.

[0187] Fig.16 The specific structural block diagram of the electronic device provided in the embodiment of the present invention is shown, and the electronic device can be used to implement the track-based incremental generation method of the road network provided in the above embodiment. The electronic device 500 can be a terminal, a server, etc., wherein the terminal can include a tablet computer, a laptop computer, a personal computer (PC, Personal Computer), a micro processing box, or other devices.

[0188] The RF circuit 510 is used to receive and send electromagnetic waves, realize the mutual conversion between electromagnetic waves and electrical signals, and thus communicate with a communication network or other devices. The RF circuit 510 may include various existing circuit elements for performing these functions, such as antennas, radio frequency transceivers, digital signal processors, encryption / decryption chips, user identity module (SIM) cards, memories, etc. The RF circuit 510 can communicate with various networks such as the Internet, corporate intranets, wireless networks, or communicate with other devices through wireless networks. The above-mentioned wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The above-mentioned wireless networks may use various communication standards, protocols and technologies, including but not limited to Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as Institute of Electrical and Electronics Engineers standards IEEE 802.11a, IEEE 802.11b, IEEE802.11g and / or IEEE802.11n), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging and short messages, and any other suitable communication protocols, even those that have not yet been developed.

[0189] The memory 520 can be used to store software programs and modules, such as the corresponding program instructions / modules in the above embodiments. The processor 580 executes various functional applications and data processing by running the software programs and modules stored in the memory 520, that is, realizing the functions of taking pictures with the front camera, processing the captured images, and switching the display color of the display content on the display screen. The memory 520 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 520 may further include a memory remotely arranged relative to the processor 580, and these remote memories may be connected to the electronic device 500 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0190] The input unit 530 may be used to receive input digital or character information, and generate a keyboard and a mouse related to user settings and function control.

[0191] The display unit 540 may be used to display information input by the user or information provided to the user and various graphical user interfaces, which may be composed of graphics, text, icons, videos, and any combination thereof. The display unit 540 may include a display panel 541, which may be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like.

[0192] The audio circuit 560, the speaker 561, and the microphone 562 can provide an audio interface between the user and the electronic device 500. The audio circuit 560 can transmit the electrical signal converted from the received audio data to the speaker 561, which is converted into a sound signal for output; on the other hand, the microphone 562 converts the collected sound signal into an electrical signal, which is received by the audio circuit 560 and converted into audio data, and then the audio data is output to the processor 580 for processing, and then sent to another terminal through the RF circuit 510, or the audio data is output to the memory 520 for further processing. The audio circuit 560 may also include an earplug jack to provide communication between an external headset and the electronic device 500.

[0193] The electronic device 500 can help the user receive requests, send information, etc. through the transmission module 570 (such as a Wi-Fi module), which provides the user with wireless broadband Internet access. Although the transmission module 570 is shown in the figure, it can be understood that it is not a necessary component of the electronic device 500 and can be omitted as needed without changing the essence of the invention.

[0194] The processor 580 is the control center of the electronic device 500. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and / or modules stored in the memory 520, and calling data stored in the memory 520, it executes various functions of the electronic device 500 and processes data, thereby monitoring the electronic device as a whole. Optionally, the processor 580 may include one or more processing cores; in some embodiments, the processor 580 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communications. It can be understood that the above-mentioned modem processor may not be integrated into the processor 580.

[0195] The electronic device 500 also includes a power supply 590 (such as a battery) for supplying power to various components. In some embodiments, the power supply can be logically connected to the processor 580 through a power management system, so that the power management system can manage charging, discharging, and power consumption management. The power supply 590 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0196] Although not shown, the electronic device 500 also includes a camera (such as a front camera, a rear camera), a Bluetooth module, etc., which will not be described in detail here. Specifically in this embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal also includes a memory, and one or more programs, wherein one or more programs are stored in the memory, and are configured to be executed by one or more processors. One or more programs include instructions for performing the following operations:

[0197] Acquire an original trajectory sequence, and preprocess the original trajectory sequence;

[0198] Calculating motion features of the preprocessed trajectory sequence, extracting geometric features of the preprocessed trajectory sequence, and performing random lane change trajectory detection based on the motion features and the geometric features to obtain a preferred sub-trajectory sequence;

[0199] Obtain road network data;

[0200] Calculating a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence;

[0201] Matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence;

[0202] An inserted road segment is constructed based on the second sub-trajectory sequence to generate a new lane, and a weight of a path corresponding to the first sub-trajectory sequence is updated.

[0203] In specific implementation, the above modules can be implemented as independent entities, or can be arbitrarily combined and implemented as the same or several entities. The specific implementation of the above modules can be found in the previous method embodiments, which will not be repeated here.

[0204] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling related hardware through instructions, and the instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. To this end, an embodiment of the present invention provides a storage medium, which stores a plurality of instructions, which can be loaded by a processor to execute the steps of any embodiment of the method for incrementally generating a vehicle road network based on trajectories provided in an embodiment of the present invention.

[0205] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0206] Since the instructions stored in the storage medium can execute the steps in any embodiment of the method for incrementally generating a track-based road network provided in the embodiments of the present invention, the beneficial effects that can be achieved by any method for incrementally generating a track-based road network provided in the embodiments of the present invention can be achieved. For details, please refer to the previous embodiments and will not be repeated here.

[0207] The above is a detailed introduction to a trajectory-based road network incremental generation method, device, storage medium and electronic device provided in an embodiment of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for incremental generation of a road network based on trajectories, characterized in that: The method comprises: Acquire an original trajectory sequence, and preprocess the original trajectory sequence; Calculating motion features of the preprocessed trajectory sequence, extracting geometric features of the preprocessed trajectory sequence, and performing random lane change trajectory detection based on the motion features and the geometric features to obtain a preferred sub-trajectory sequence; Obtain road network data; Calculating a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence; Calculating a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and moving the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence, comprising: calculating a displacement difference and a direction difference between a trajectory point in the preferred sub-trajectory sequence and a trajectory point in a surrounding area of ​​the trajectory point relative to a corresponding point in the path in the corresponding road network data; calculating a single displacement value based on the displacement difference and the direction difference; and moving all trajectory points in the preferred sub-trajectory sequence according to the single displacement value to cluster the preferred sub-trajectory sequence; Matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence; matching the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence, including: Construct a hidden Markov model, traverse the optimal sub-trajectory sequence after clustering, regard the trajectory points as observations, and the real position or real road section as hidden states; The trajectory points in the clustered preferred sub-trajectory sequence are calculated by the following formula: The path to the road network data The observed transition probability : in, For trajectory points To Path The projection distance on is the observation error, exp is the exponential function, Represents trajectory points To Path The projection point on ; The transition probability between paths in the road network data is calculated by the following formula: in, in, It is a breadth-first search algorithm for directed graphs with depth. Represents the search network node Depth Find the optimal sequence: initialization: Recursion yields: in, , is the length of the observation sequence, S is the observed object, Before the Representative The final state is The probability of the state sequence that the observation result most likely corresponds to, is the probability matrix of the observation result of the first node to the matching sequence, is the probability transfer matrix between objects in the matching sequence, For the front The final state is The probability of the state sequence that the observation result most likely corresponds to, is the probability matrix of the observation result of the tth node to the matching sequence; The first sub-trajectory sequence is calculated by the following formula { }: in, For the function Find parameters; Obtaining an unmatched second sub-trajectory sequence according to the clustered preferred sub-trajectory sequence and the matched first sub-trajectory sequence; An inserted road segment is constructed based on the second sub-trajectory sequence to generate a new lane, and a weight of a path corresponding to the first sub-trajectory sequence is updated.

2. The method for incremental generation of a road network based on trajectories according to claim 1, characterized in that: The preprocessing of the original trajectory sequence includes: Get the direction angle sequence corresponding to the original trajectory sequence; Calculate a corresponding heading angle sequence according to the original trajectory sequence, calculate an angle difference based on the direction angle sequence and the heading angle sequence, and remove a trajectory sequence whose angle difference exceeds a preset angle threshold; The corresponding speed sequence is calculated according to the original trajectory sequence, and the trajectory sequence whose speed exceeds a preset speed threshold is eliminated.

3. The method for incremental generation of a track-based road network according to claim 1, characterized in that: The geometric features include a road shape descriptor, and calculating the geometric features of the preprocessed trajectory sequence includes: Find the center point of each trajectory sequence; Traverse the trajectory points within a range with the trajectory center point as the center and r as the radius, and calculate the relative angle and distance; Divide the distance into sections according to a preset angle and a preset length r / 3 to form a plurality of section intervals; The number of trajectory points falling into each partition interval is calculated, and the partition interval is encoded based on the number to obtain a road shape descriptor.

4. The method for incremental generation of a track-based road network according to claim 1, characterized in that: The motion features at least include the cumulative heading angle change, the start-end heading angle change and the angular velocity change, and the geometric features at least include the road shape descriptor, the distance from the two ends of the trajectory sequence to the road boundary and the lateral displacement of the trajectory sequence relative to the road centerline; Calculating the motion features of the preprocessed trajectory sequence, extracting the geometric features of the preprocessed trajectory sequence, and performing lane change trajectory detection based on the motion features and the geometric features to obtain a preferred sub-trajectory sequence, including: The accumulated heading angle change, the start-end heading angle change, the angular velocity change, the road shape descriptor, the distances from both ends of the trajectory sequence to the road boundary, and the lateral displacement of the trajectory sequence relative to the road centerline are input into a vector machine for binary classification, and random lane change trajectories are eliminated to obtain an optimal sub-trajectory sequence.

5. The method for incremental generation of a road network based on trajectories according to claim 1, characterized in that: Constructing an inserted road segment based on the second sub-trajectory sequence to generate a new lane, and updating a weight of a path corresponding to the first sub-trajectory sequence, including: Solving the path of the road network data by using a breadth-first search algorithm with depth of an undirected graph; If the result of the solution is 0, matching the front and rear endpoints of the second sub-trajectory sequence to the path of the road network data, constructing an insertion section, and generating a new lane; If the solution result is not 0 and does not meet the matching condition, inserting the second sub-trajectory sequence into the graph structure data corresponding to the road network data; The weight of the path corresponding to the first sub-trajectory sequence is updated.

6. A device for generating a road network increment based on a trajectory, the device for generating a road network increment based on a trajectory being used to implement the method for generating a road network increment based on a trajectory according to claim 1, characterized in that: include: A first acquisition module is used to acquire an original trajectory sequence; A preprocessing module, used for preprocessing the original trajectory sequence; a random lane-changing trajectory detection module, configured to extract motion features of the preprocessed trajectory sequence, calculate geometric features of the preprocessed trajectory sequence, perform random lane-changing trajectory detection based on the motion features and the geometric features, and obtain a preferred sub-trajectory sequence; The second acquisition module is used to acquire road network data; a clustering module, configured to calculate a single displacement value between a path in the road network data and a corresponding point of the preferred sub-trajectory sequence, and to move the trajectory points of the preferred sub-trajectory sequence based on the single displacement value to cluster the preferred sub-trajectory sequence; A matching module, configured to match the road network data with the clustered preferred sub-trajectory sequence to obtain a matched first sub-trajectory sequence and an unmatched second sub-trajectory sequence; The lane construction and weight updating module is used to construct the inserted road segment based on the second sub-trajectory sequence to generate a new lane, and update the weight of the path corresponding to the first sub-trajectory sequence.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the trajectory-based incremental road network generation method according to any one of claims 1 to 5.

8. An electronic device, characterized in that: It includes a processor and a memory, the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used to execute the steps in the trajectory-based road network incremental generation method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Road network completion method and device, electronic equipment and readable storage medium

    CN112818075A

  • High-sampling-rate trajectory data map matching method based on road network connectivity

    CN114485692A

  • Road data updating method and apparatus, device, and storage medium

    WO2024141037A1