Road trajectory determination method, apparatus and device, and computer storage medium

Abstract

The invention discloses a road track determination method, device and equipment and a computer storage medium, and relates to the technical field of intelligent traffic. The method comprises the following steps: acquiring a user positioning point coordinate, positioning time, a preset road segment coordinate and a topological relation; determining the matching probability of each positioning point and the road segment according to the coordinates of the positioning points, the coordinates of the road segment, the topological relation of the road segment and the relation of the matching probability; according to the matching probability of the positioning points and the road segments, the topological relation of the road segments and the relation of the conditional probability, a prediction track is obtained; and inputting the road segment coordinates of the predicted trajectory and the coordinates of the user positioning point into the trajectory correction model as input parameters of the feature function, and determining a target trajectory according to the relationship among the pre-trained feature weight in the trajectory correction model, the input parameters of the feature function and the target probability. According to the embodiment of the invention, high-precision prediction of the user trajectory can be realized, and the accuracy of the generated road trajectory is improved.

Description

Technical Field

[0001] The present application belongs to the field of intelligent transportation technology, and in particular, relates to a method, device, equipment and computer storage medium for determining a road trajectory. Background Art

[0002] In the field of intelligent transportation technology, operators can obtain measurement reports of user devices containing user location data through base stations. Using user location data obtained from the operator network to generate trajectories and perform analysis has become an important means of gaining insight into user behavior.

[0003] When generating trajectories, existing trajectory generation methods often match positioning data to the nearest road segment through user positioning data, and splice the road segments into continuous trajectories according to time sequence. However, when the road network topology is more complex, the generated road trajectory has low accuracy. Summary of the invention

[0004] The embodiments of the present application provide a method, device, equipment and computer storage medium for determining a road trajectory to solve the problem of low accuracy of road trajectories generated by the prior art.

[0005] In a first aspect, an embodiment of the present application provides a method for determining a road trajectory, the method comprising:

[0006] Acquire user positioning data and road network data, the user positioning data including coordinates of multiple user positioning points and positioning times of the user positioning points, and the road network data including coordinates of multiple preset road segments and preset road segment topological relationships;

[0007] Based on the coordinates of the user's positioning point, the coordinates of the preset road segment and the preset road segment topological relationship, the target matching probability of each user's positioning point and the road segment is determined according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability;

[0008] Based on the target matching probability and the preset road segment topological relationship, according to the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, the predicted conditional probability of the positioning point corresponding to the second positioning time matching each road segment when the positioning point corresponding to the first positioning time matches each road segment is determined in sequence according to the positioning time of the positioning point, and the predicted trajectory consisting of the road segment with the largest predicted conditional probability is obtained, the first positioning time is the positioning time of any user positioning point, and the second positioning time is the positioning time next to the first positioning time;

[0009] The coordinates of the road segments of the predicted trajectory and the coordinates of the user's positioning points are input into the trajectory correction model as input parameters of the feature function. According to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, the target probabilities of the positioning points corresponding to the second positioning time matching each target road segment when the positioning points corresponding to the first positioning time match each target road segment are determined in sequence according to the positioning time sequence of the positioning points. The target trajectory consisting of the road segments with the largest cumulative target probability is obtained, and the target road segment is the road segment in the predicted trajectory.

[0010] In a second aspect, an embodiment of the present application provides a device for determining a road trajectory, the device comprising:

[0011] An acquisition module is used to acquire user positioning data and road network data, wherein the user positioning data includes coordinates of multiple user positioning points and positioning times of the user positioning points, and the road network data includes coordinates of multiple preset road segments and preset road segment topological relationships;

[0012] A determination module, for determining a target matching probability between each user positioning point and a road segment based on the coordinates of the user positioning point, the coordinates of the preset road segment and the preset road segment topological relationship, according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability;

[0013] A prediction module is used to determine, based on the target matching probability and the preset road segment topological relationship, the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, and in sequence according to the positioning time sequence of the positioning points, when the positioning point corresponding to the first positioning time matches the road segment, the prediction conditional probability that the positioning point corresponding to the second positioning time matches the road segment, and obtain a prediction trajectory consisting of the road segment with the largest prediction conditional probability, wherein the first positioning time is the positioning time of any user positioning point, and the second positioning time is the positioning time next to the first positioning time;

[0014] The correction module is used to input the coordinates of the road segments of the predicted trajectory and the coordinates of the user positioning points into the trajectory correction model as input parameters of the feature function, and determine the target probability of the positioning point corresponding to the second positioning time matching each target road segment when the positioning point corresponding to the first positioning time matches each target road segment according to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, so as to obtain a target trajectory consisting of the road segment with the largest cumulative target probability, and the target road segment is the road segment in the predicted trajectory.

[0015] In a third aspect, an embodiment of the present application provides a terminal device, which includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the method for determining a road trajectory as in the first aspect is implemented.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the method for determining a road trajectory as in the first aspect is implemented.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product. When instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method for determining a road trajectory as in the first aspect.

[0018] The embodiments of the present application provide a method, device, equipment and computer storage medium for determining a road trajectory. The method first obtains the coordinates of multiple user positioning points, the positioning time of the user positioning points, the coordinates of multiple preset road segments and the preset road segment topological relationship. The target matching probability of each user positioning point and the road segment is determined based on the relationship between the coordinates of the positioning point, the coordinates of the road segment, the topological relationship of the road segment and the matching probability; the matching degree of each positioning point and the road segment can be quantified, providing a confidence estimate of the road where the positioning point is located. According to the matching probability of the positioning point and the road segment, the relationship between the topological relationship of the road segment and the conditional probability, the predicted conditional probability of the positioning point corresponding to the second positioning time matching each road segment is determined when the positioning point corresponding to the first positioning time matches each road segment, and the predicted trajectory consisting of the road segment with the largest predicted conditional probability is obtained; combined with the topological relationship of the road segment, the road segment with the largest predicted conditional probability is selected to form the predicted trajectory, which effectively improves the accuracy of the prediction. The coordinates of the road segments of the predicted trajectory and the coordinates of the user's positioning point are input into the trajectory correction model as input parameters of the feature function. According to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, the target probability of the positioning point corresponding to the second positioning time matching each target road segment when the positioning point corresponding to the first positioning time matches each target road segment is determined in sequence according to the positioning time sequence of the positioning point, and the target trajectory consisting of the road segment with the largest cumulative target probability is obtained, and the target road segment is the road segment in the predicted trajectory; through the pre-trained feature weights, the model can assign corresponding feature weights to different features and weighted sum them, thereby adjusting the importance of different features, reducing the cumulative error caused by the initial positioning error or prediction error, and improving the prediction accuracy. The embodiment of the present application predicts the matching of the positioning point corresponding to the next positioning time based on the matching of the positioning point corresponding to the current positioning time, which can achieve high-precision prediction of the user trajectory and improve the accuracy of the generated road trajectory. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] Figure 1 is a flowchart of a method for determining a road trajectory provided in an embodiment of the present application;

[0021] Figure 2 It is a flowchart of an implementation method of data preprocessing provided in an embodiment of the present application;

[0022] Figure 3 It is a flowchart of an implementation method for determining target matching probability provided in an embodiment of the present application;

[0023] Figure 4 It is a flowchart of an implementation method for determining a predicted trajectory provided in an embodiment of the present application;

[0024] Figure 5 It is a flowchart of an implementation method of establishing a linear reference system event sequence provided in an embodiment of the present application;

[0025] Figure 6 is a schematic diagram of a linear referencing system event sequence provided by an embodiment of the present application;

[0026] Figure 7 is a schematic diagram of the structure of a device for determining a road trajectory provided in an embodiment of the present application;

[0027] Figure 8 It is a schematic diagram of the structure of the terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0029] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0030] In the prior art, operators can obtain the measurement report (MR) data of user equipment through base stations, which contains the user's location information. First, the road is rasterized and interpolated, and the attributes of the road, building, etc. of the grid where the MR point is located are identified. Then, the positioning points during the commuting period are screened according to the user's residence and workplace, and the grids with high-frequency visits are extracted. Finally, the missing sections are filled along the road topology relationship to obtain a complete commuting trajectory. However, what is generated are discrete motion trajectory fragments, which are difficult to meet the requirements of full-time, continuous and complete trajectory generation. In addition, when generating the trajectory, the positioning data is often matched to the nearest road segment through the user's positioning data, and the road segments are spliced ​​into a continuous trajectory according to the time sequence. However, when the road network topology is more complex, it is difficult to effectively handle the problem of complex road network topology, and the generated road trajectory has low accuracy.

[0031] In order to solve the existing technology, the embodiments of the present application provide a method, device, equipment and computer storage medium for determining a road trajectory. The method first obtains the coordinates of multiple user positioning points, the positioning time of the user positioning points, the coordinates of multiple preset road segments and the preset road segment topological relationship. The target matching probability of each user positioning point and the road segment is determined according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the topological relationship of the road segment and the matching probability; the matching degree of each positioning point and the road segment can be quantified, and a confidence estimate of the road where the positioning point is located is provided. According to the matching probability of the positioning point and the road segment, the relationship between the topological relationship of the road segment and the conditional probability, the predicted conditional probability of the positioning point corresponding to the second positioning time matching each road segment is determined when the positioning point corresponding to the first positioning time matches each road segment, and the predicted trajectory consisting of the road segment with the largest predicted conditional probability is obtained; combined with the topological relationship of the road segment, the road segment with the largest predicted conditional probability is selected to form the predicted trajectory, which effectively improves the accuracy of the prediction. The coordinates of the road segments of the predicted trajectory and the coordinates of the user's positioning point are input into the trajectory correction model as input parameters of the feature function. According to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, the target probability of the positioning point corresponding to the second positioning time matching each target road segment when the positioning point corresponding to the first positioning time matches each target road segment is determined in sequence according to the positioning time sequence of the positioning point, and the target trajectory consisting of the road segment with the largest cumulative target probability is obtained, and the target road segment is the road segment in the predicted trajectory; through the pre-trained feature weights, the model can assign corresponding feature weights to different features and weighted sum them, thereby adjusting the importance of different features, reducing the cumulative error caused by the initial positioning error or prediction error, and improving the prediction accuracy. The embodiment of the present application predicts the matching of the positioning point corresponding to the next positioning time based on the matching of the positioning point corresponding to the current positioning time, which can achieve high-precision prediction of the user trajectory and improve the accuracy of the generated road trajectory.

[0032] The following first introduces the method for determining the road trajectory provided in the embodiment of the present application.

[0033] Figure 1 FIG. 1 is a flow chart of a method for determining a road trajectory provided by an embodiment of the present application. Figure 1 As shown, the method may include the following steps: S101 to S104.

[0034] S101, obtaining user positioning data and road network data, wherein the user positioning data includes coordinates of multiple user positioning points and positioning times of the user positioning points, and the road network data includes coordinates of multiple preset road segments and preset road segment topological relationships.

[0035] The user location data is related data for indicating the user's location, the coordinates of the user's location point are the coordinate values ​​of the user's specific location in the geographic space, and the location time of the user's location point is the specific time when the user's location point is acquired.

[0036] Road network road data is related data used to indicate road network information. The coordinates of the preset road segments may include the coordinates of key points such as the starting point, end point and connection point of the road segment. The preset road segment topological relationship describes the connection relationship between the road segments.

[0037] In some embodiments, the user positioning data can be obtained by acquiring the measurement report (MR) data of the user equipment through the base station. The coordinates of the user positioning point obtained through the MR data are longitude and latitude coordinates, and the road network data can be obtained through a digital map service provider.

[0038] S102, based on the coordinates of the user's positioning point, the coordinates of the preset road segments and the preset road segment topological relationship, determine the target matching probability of each user's positioning point and the road segment according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability.

[0039] The target matching probability is used to indicate the matching degree between each user positioning point and each road segment. The larger the target matching probability is, the greater the matching possibility between the user positioning point and the corresponding road segment is.

[0040] In some embodiments, the road segment topological relationship represents the connectivity on the road segment, 1 represents connected, and 0 represents disconnected.

[0041] In some embodiments, determining the target matching probability between each user positioning point and the road segment according to the coordinates of the positioning point, the coordinates of the road segment, the topological relationship of the road segment, and the relationship between the matching probabilities may include:

[0042] Calculate the vertical distance from the positioning point to the road segment through the coordinates of the positioning point and the coordinates of the road segment;

[0043] The vertical distance and the sum of the topological relationships between the road segment and the road segments that match the positioning points corresponding to other adjacent positioning times are multiplied and normalized to obtain the target matching probability.

[0044] The closer the positioning point is to the road segment, the greater the probability that the user point is matched to the candidate matching road. At the same time, the topological relationship between the matching roads of the positioning points corresponding to adjacent positioning times is introduced. If the adjacent matching roads are not connected, the matching probability is reduced accordingly.

[0045] When calculating the matching probability, by introducing the topological relationship between the matching roads of the positioning points corresponding to adjacent positioning times, that is, checking whether these road segments are interconnected in the actual road network, it is possible to avoid erroneously matching the user's positioning points to these unconnected roads, thereby improving the accuracy of the matching probability calculation.

[0046] S103, based on the target matching probability and the preset road segment topological relationship, according to the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, in accordance with the positioning time sequence of the positioning point, in the case where the positioning point corresponding to the first positioning time matches each road segment, the predicted conditional probability of the positioning point corresponding to the second positioning time matching each road segment is determined, and a predicted trajectory consisting of road segments with the largest predicted conditional probability is obtained, the first positioning time is the positioning time of any user positioning point, and the second positioning time is the positioning time next to the first positioning time.

[0047] In some embodiments, the relationship between the matching probability of the positioning point and the road segment, the topological relationship of the road segment and the conditional probability can be: the conditional probability that the positioning point corresponding to the second positioning time matches the specific road segment is equal to the probability that the positioning point corresponding to the first positioning time matches the specific road segment, the result of multiplying the matching probability of the second positioning point and the specific road segment and the value corresponding to the topological relationship of the road segment.

[0048] In some embodiments, when determining the predicted trajectory, a hidden Markov model (HMM) can be used to model the matching probability between the positioning point and the road segment, the relationship between the road segment topology and the conditional probability, and obtain the state transfer matrix, the observation probability matrix and the initial state probability vector. The Viterbi algorithm is used to calculate the probability based on the state transfer matrix, the observation probability matrix and the initial state probability vector, so as to infer the optimal predicted trajectory.

[0049] Among them, the state transition probability matrix is ​​defined according to the topological relationship of the road segments. Each element represents the connection relationship between the road segments represented by the row elements and the column elements. Only the interconnected road segments have non-zero probabilities. The interconnected road segments can take the value of 1, and the non-interconnected road segments take the value of 0. Each element in the observation probability matrix represents the matching probability of the positioning point and the road segment. The initial state probability vector is a preset value, which represents the matching probability of the initial positioning point and the road segment.

[0050] In one example, the process of obtaining the predicted trajectory consisting of the road segments with the maximum prediction conditional probability may specifically be:

[0051] Based on the probability that the positioning point corresponding to the initial positioning time matches each road segment, the probability of the positioning point matching each road segment is respectively multiplied by the probability that the positioning point corresponding to the second positioning time matches each road segment, and the road segment topological relationship of the road segments matched by the two positioning points, to obtain the conditional probability that the positioning point corresponding to the second positioning time matches each road segment when the positioning point corresponding to the initial positioning time matches each road segment;

[0052] Based on the probability that the positioning points corresponding to the initial positioning time and the second positioning time match the corresponding road segments, the probabilities of the positioning points corresponding to the third positioning time matching the respective road segments are multiplied respectively, and the road segment topological relationship of the road segments matching the second positioning point and the third positioning point is multiplied, so as to obtain the conditional probability that the positioning points corresponding to the third positioning time match the respective road segments when the positioning points corresponding to the initial positioning time and the second positioning time match the respective road segments;

[0053] And so on, until the conditional probability that the positioning point corresponding to the last positioning time matches each road segment is obtained when the positioning points corresponding to all positioning times before the last positioning time match each road segment;

[0054] The road segment with the largest conditional probability that the positioning point corresponding to the last positioning time matches each road segment is selected, and the road segments matched with the preceding matching condition of the conditional probability, that is, each positioning point corresponding to each positioning time before the last positioning time, are formed into a predicted trajectory.

[0055] By considering the target matching probability and the road topological relationship and selecting the predicted trajectory consisting of the road segments with the highest probability, the accuracy of the predicted trajectory generation can be improved while ensuring that the predicted trajectory is coherent.

[0056] S104, input the coordinates of the road segments of the predicted trajectory and the coordinates of the user positioning points into the trajectory correction model as input parameters of the feature function, and determine the target probabilities of the positioning points corresponding to the second positioning time matching the target road segments when the positioning points corresponding to the first positioning time match the target road segments according to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, and obtain the target trajectory consisting of the road segments with the largest cumulative target probability, wherein the target road segments are the road segments in the predicted trajectory.

[0057] Among them, the trajectory correction model is used to correct the input predicted trajectory to determine the corrected target trajectory.

[0058] In some embodiments, the relationship between the pre-trained feature weights, the input parameters of the feature function and the target probability can be that the target probability is equal to the result of inputting the input parameters into the feature function and performing weighted summation using the feature weights.

[0059] In some embodiments, the pre-trained feature data is obtained by training the trajectory correction model with the training data and selecting the parameter value that maximizes the probability that the predicted result will appear as the labeled result of the training data. It is used to evaluate the importance of the feature function to the trajectory correction and reflects the importance of different features to the trajectory correction.

[0060] In some embodiments, the trajectory correction model can be obtained by performing maximum likelihood estimation training on the CRF model based on training data, a conditional random field (CRF) model and a limited memory BFGS optimization algorithm.

[0061] In some embodiments, the feature function may include a transition feature and a state feature, wherein the transition feature is a position feature between positioning points at adjacent positioning times, and the state feature is a position feature between a positioning point and a road segment.

[0062] In some embodiments, the trajectory correction model corrects the predicted trajectory through the characteristic function and characteristic weights in the trajectory correction model. The trajectory correction model inputs the input parameters into the characteristic function, and then calculates the weighted sum of the characteristic function values ​​and normalizes it according to the characteristic weights of the characteristic function. The sum is used as the probability value of the positioning point at the positioning time matching each target road segment, and the trajectory consisting of road segments corresponding to the cumulative maximum probability value of each positioning point matching each target road segment is selected as the target trajectory.

[0063] By correcting the matching probability between the positioning points and the matching road segments according to the features and weight functions between the positioning points and the positioning points, and between the positioning points and the matching road segments, the predicted trajectory can be accurately adjusted to improve the prediction accuracy.

[0064] The target matching probability of each user's positioning point and road segment is determined based on the relationship between the coordinates of the positioning point, the coordinates of the road segment, the topological relationship of the road segment and the matching probability; the matching degree of each positioning point and the road segment can be quantified, providing a confidence estimate of the road where the positioning point is located. According to the matching probability of the positioning point and the road segment, the relationship between the topological relationship of the road segment and the conditional probability, the predicted conditional probability of the positioning point corresponding to the second positioning time matching each road segment is determined when the positioning point corresponding to the first positioning time matches each road segment, and the predicted trajectory consisting of the road segment with the largest predicted conditional probability is obtained; combined with the topological relationship of the road segment, the road segment with the largest predicted conditional probability is selected to form the predicted trajectory, which effectively improves the accuracy of the prediction. The coordinates of the road segments of the predicted trajectory and the coordinates of the user's positioning points are input into the trajectory correction model as input parameters of the feature function. According to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, the target probability of the positioning point corresponding to the second positioning time matching each target road segment when the positioning point corresponding to the first positioning time matches each target road segment is determined in sequence according to the positioning time of the positioning points. The target trajectory consisting of the road segment with the largest cumulative target probability is obtained, and the target road segment is the road segment in the predicted trajectory. Through the pre-trained feature weights, the model can assign corresponding feature weights to different features and perform weighted summation, thereby adjusting the importance of different features, reducing the cumulative error caused by initial positioning error or prediction error, and improving prediction accuracy.

[0065] In some embodiments, before determining the target matching probability between each user location point and the road segment, the method may further include:

[0066] Obtain preset road segments whose coordinates are within the preset range of the user positioning point coordinates, calculate the vertical distance from each user positioning point coordinate to each road segment within the preset range, and obtain road segments whose distance does not exceed the set threshold as candidate matching roads for each user positioning point.

[0067] By selecting only road segments whose distance does not exceed the set threshold as candidate matching roads, the candidate range is effectively narrowed, the calculation process and amount are simplified, and the prediction efficiency is improved.

[0068] In some embodiments, Figure 2 As shown, before determining the target matching probability of each positioning point and the road segment according to the relationship between the coordinates of the positioning point, the coordinates of the road segment and the matching probability, the method may further include: S201 to S202.

[0069] S201, cleaning the user location data to remove missing data and duplicate data in the fields of the user location data;

[0070] S202, cleaning the road network road data to remove duplicate coordinates of preset road segments and topological errors in the topological relationship of preset road segments.

[0071] In some embodiments, field missing data may include user identification ID, positioning time, longitude and latitude, etc., and topological errors may include illegal hanging points and self-intersections of road nodes in the road network.

[0072] By cleaning up duplicate data, unnecessary calculations can be reduced, computing resources can be saved, processing speed can be improved, and missing data and topological errors can be removed to avoid these erroneous data from misleading the matching algorithm, thereby improving the reliability of the matching results.

[0073] In some embodiments, the coordinates of the user's location point or the coordinates of the preset road segment are longitude and latitude coordinates. After obtaining the user's location data and the road network data, the method may further include:

[0074] The coordinates of the user positioning point and the coordinates of the preset road segment are converted into a plane rectangular coordinate system based on the road centerline as new coordinates of the user positioning point and the coordinates of the preset road segment.

[0075] The conversion formula is:

[0076]

[0077]

[0078] Among them, x and y are the horizontal and vertical coordinates of the coordinates in the plane rectangular coordinate system, respectively. and λ are the latitude and longitude of the original longitude and latitude data respectively, and R is the radius of the earth.

[0079] Unifying the coordinates of the user's positioning point and the coordinates of the preset road segment into the same plane rectangular coordinate system can ensure data consistency, eliminate errors caused by different coordinate systems, and improve calculation efficiency.

[0080] In some embodiments, Figure 3 As shown, based on the coordinates of the user's positioning point, the coordinates of the preset road segment and the preset road segment topological relationship, the target matching probability of each positioning point and the road segment is determined according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability, which can include: S301 to S302.

[0081] S301, based on the coordinates of the user positioning point and the coordinates of the candidate road segment, according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, and the vertical distance between the positioning point and the road segment, determining the vertical distance between the user positioning point and the candidate road segment, the candidate road segment being a road segment whose coordinates are within a preset range of the coordinates of the user positioning point;

[0082] S302, based on the vertical distance between the user positioning point and the candidate road segment, the preset road segment topological relationship, and the relationship between the vertical distance between the positioning point and the road segment, the road segment topological relationship and the matching probability, determine the target matching probability between each user positioning point and the road segment.

[0083] By considering factors such as the coordinates of the user's positioning point and the road segment, the topological relationship of the road segment and the vertical distance, the matching probability between the user's positioning point and the road segment can be determined more accurately, thereby improving the matching accuracy.

[0084] In some embodiments, the calculation formula for the matching probability is:

[0085]

[0086] Among them, f is the mapping function of Sigmoid form:

[0087]

[0088] Among them, λ is the balance factor, c i,j represents the i-th positioning point p i Its adjacent point p j The topological connectivity of the matching road segment, 1 means connected, and 0 means disconnected. i For p i The collection of adjacent points of .

[0089] In some embodiments, Figure 4 As shown, based on the target matching probability and the preset road segment topological relationship, according to the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, the predicted conditional probability of the positioning point corresponding to the second positioning time matching each road segment when the positioning point corresponding to the first positioning time matches each road segment is determined in sequence according to the positioning time of the positioning point, and a predicted trajectory consisting of the road segments with the largest predicted conditional probability is obtained, including: S401 to S403.

[0090] S401, using the target matching probability between the initial user positioning point and the road segment corresponding to the initial positioning time as the initial prediction condition probability of matching the initial user positioning point to each road segment;

[0091] S402, based on the target matching probability between each user's positioning point and the road segment, the preset road segment topological relationship and the initial prediction condition probability, according to the matching probability between the positioning point and the road segment, the relationship between the road segment topological relationship and the prediction probability, in accordance with the positioning time sequence of the positioning points, sequentially determine the prediction condition probability that the positioning point corresponding to the second positioning time matches each road segment when the positioning point corresponding to the first positioning time matches each road segment;

[0092] S403: Select a trajectory consisting of a road segment with the largest prediction condition probability matched by the positioning point corresponding to the second positioning time and all road segments matched by the positioning points corresponding to the first positioning time as a predicted trajectory.

[0093] The matching situation is determined in sequence according to the positioning time, and prediction is made based on conditional probability. Prediction is made in the positioning time sequence of the positioning points to ensure the continuity and temporal consistency of trajectory prediction. The most likely trajectory can be obtained by selecting the predicted trajectory composed of the road segments with the largest prediction conditional probability. Time series prediction of user trajectories can be achieved.

[0094] In some embodiments, the user positioning data further includes the speed of the user positioning point, and the road network road data further includes the direction of a preset road segment and a speed threshold of a preset road segment. After obtaining the predicted trajectory consisting of the road segment with the maximum prediction condition probability, the method may further include:

[0095] Determine the distance, average speed and first change angle of adjacent user positioning points corresponding to adjacent positioning times according to the coordinates, positioning time and speed of the user positioning points;

[0096] Determine the average speed difference between the user positioning point and the corresponding road segment and the second change angle according to the average speed of the user positioning point, the first change angle, the direction of the road segment and the speed threshold of the road segment;

[0097] The distance between adjacent user positioning points, the average speed, the first change angle, the average speed difference between the positioning point and the corresponding road segment, and the second change angle are used as input parameters of the feature function. Based on the input parameters of the feature function and the relationship between the feature weights pre-trained in the trajectory correction model, the input parameters of the feature function and the target probability, the target probability of the positioning point corresponding to the second positioning time matching each target road segment when the positioning point corresponding to the first positioning time matches each target road segment is determined in sequence according to the positioning time of the positioning points, and the target trajectory consisting of the road segment with the largest cumulative target probability is obtained.

[0098] A feature function is introduced as an input parameter. Based on the pre-trained feature weights and the input parameters of the feature function, the importance of different features can be adjusted, the relationship between the user's positioning point and the road segment can be more flexibly represented, and the accuracy and robustness of trajectory correction can be improved.

[0099] In some embodiments, the characteristic function includes a distance transfer characteristic, a speed transfer characteristic, an angle transfer characteristic, a distance state characteristic, a topological state characteristic, a speed state characteristic, and an angle state characteristic.

[0100] In some embodiments, the characteristic functions of the distance transfer feature, the speed transfer feature, and the angle transfer feature are respectively:

[0101] t dis (y i ,y i-1 )=|y i -y i-1 |

[0102]

[0103] where |·| represents the Euclidean distance, y i represents the i-th positioning point, t i is the positioning time of the ith positioning point.

[0104] The characteristic functions of distance state characteristics, topological state characteristics, velocity state characteristics and angle state characteristics are:

[0105] s dist (y i ,q i )=d(y i ,q i )

[0106] s topo (y i ,q i ,q i-1 ,q i+1 )=c(q i ,q i-1 )+c(q i ,q i+1 )

[0107] s spd (y i ,q i )=|v(y i ,y i-1 )-v limit (q i )|

[0108]

[0109] Where d(·) is the vertical distance from the positioning point to the road, c(·) is the topological connectivity of the adjacent roads, and its value is 0 or 1. v(·) is the speed of the positioning point, v limit (·) is to match the road speed limit. The coordinates of the start and end nodes of the matching road.

[0110] In some embodiments, Figure 5 As shown, after obtaining the target trajectory consisting of the road segments with the largest cumulative target probability, the method may further include: S501 to S503.

[0111] S501, determining the road segment matched by each positioning point according to the target trajectory, and determining the starting and ending mileages of the target trajectory in the road segment according to the coordinates of the positioning point and the coordinates of the road segment;

[0112] S502, determining the topological relationship of road segments matched by positioning points corresponding to adjacent positioning times;

[0113] S503, establishing a linear reference system event sequence for representing the target trajectory, the linear reference system event sequence includes multiple linear reference system events, each linear reference system event includes a preset identification ID of a road segment matched by a corresponding positioning point of the event, the start and end mileages of the target trajectory segment on the road segment, and a topological relationship of the road segment.

[0114] The linear referencing system event sequence (LRS) includes the road segment identification, start and end mileage and topological relationship of each linear referencing system event. By establishing a linear referencing system event sequence to represent the target trajectory, the trajectory can be better described, which facilitates subsequent path analysis, simulation and planning.

[0115] In some embodiments, specifically, the method may further include:

[0116] For each point y in the optimization trajectory i , calculate the ID r of the matching road id , relative position x on the road i and the accumulated mileage from the starting point i , and obtain the projection coordinates (r id ,x i ,l i ):

[0117]

[0118] Where d(·) is the distance from the positioning point to the road, r start ,r endare the coordinates of the starting and ending nodes of the road, and |r| is the length of the road. i It is calculated by accumulating the distance between the previous positioning points.

[0119] Reconstruct the topological connectivity between the anchor points and check the adjacent anchor points y i and j The matching road r i and r j , record its topological relationship c i,i+1 :

[0120]

[0121] c i,i+1 =1 means r i and r j Connected end to end, forward connected; c i,i+1 =0 means no connection or reverse connection.

[0122] In some embodiments, the trajectory is represented as a sequence of LRS events e1, ..., e k , each event is:

[0123]

[0124] in is the road ID matched by the kth trajectory, is the starting and ending mileage of the track on the current road, Identifies the direction from the current segment to the next segment, 1 means forward, 0 means backward or not connected.

[0125] In one example, if Figure 6 As shown in the figure, event 1 indicates that the trajectory starts at 0m on Road1 and ends at 100m. Event 2 indicates that the trajectory starts at 100m on Road1 and ends at 200m. Event 3 indicates that the trajectory starts at 0m on Road2 and ends at 150m, and enters the next road segment in the forward direction. Event 4 indicates that the trajectory starts at 0m on Road3 and ends at 120m, and enters the next road segment in the forward direction.

[0126] Figure 7 FIG. 7 is a schematic diagram showing a structure of a device 700 for determining a road trajectory provided in an embodiment of the present application. Figure 7 As shown, the device may include: an acquisition module 701 , a determination module 702 , a prediction module 703 and a correction module 704 .

[0127] The acquisition module 701 is used to acquire user positioning data and road network data, wherein the user positioning data includes coordinates of multiple user positioning points and positioning times of the user positioning points, and the road network data includes coordinates of multiple preset road segments and preset road segment topological relationships;

[0128] A determination module 702 is used to determine the target matching probability between each user positioning point and the road segment based on the coordinates of the user positioning point, the coordinates of the preset road segment and the preset road segment topological relationship, according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability;

[0129] The prediction module 703 is used to determine, based on the target matching probability and the preset road segment topological relationship, the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, and in sequence according to the positioning time sequence of the positioning points, when the positioning point corresponding to the first positioning time matches the road segment, the prediction conditional probability that the positioning point corresponding to the second positioning time matches the road segment, and obtain a prediction trajectory consisting of the road segment with the largest prediction conditional probability, wherein the first positioning time is the positioning time of any user positioning point, and the second positioning time is the positioning time next to the first positioning time;

[0130] The correction module 704 is used to input the coordinates of the road segments of the predicted trajectory and the coordinates of the user positioning points into the trajectory correction model as input parameters of the feature function, and determine the target probability of the positioning points corresponding to the second positioning time matching each target road segment when the positioning points corresponding to the first positioning time match each target road segment according to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, so as to obtain the target trajectory consisting of the road segments with the largest cumulative target probability, and the target road segment is the road segment in the predicted trajectory.

[0131] In some embodiments, the road trajectory determination device 700 may further include a cleaning module, which is used to clean the user positioning data to remove missing data and duplicate data in the user positioning data before determining the target matching probability of each positioning point and the road segment according to the relationship between the coordinates of the positioning point, the coordinates of the road segment and the matching probability;

[0132] Correspondingly, the cleaning module is also used to clean the road network road data to remove duplicate coordinates of preset road segments and topological errors in the topological relationship of preset road segments.

[0133] In some embodiments, the determination module 702 is further configured to determine the vertical distance between the user positioning point and the candidate road segment based on the coordinates of the user positioning point and the coordinates of the candidate road segment, according to the coordinates of the positioning point, the coordinates of the road segment, and the relationship between the vertical distance between the positioning point and the road segment, wherein the candidate road segment is a road segment whose coordinates are within a preset range of the coordinates of the user positioning point;

[0134] Correspondingly, the determination module 702 is also used to determine the target matching probability between each user positioning point and the road segment based on the vertical distance between the user positioning point and the candidate road segment, the preset road segment topological relationship, and the relationship between the vertical distance between the positioning point and the road segment, the road segment topological relationship and the matching probability.

[0135] In some embodiments, the prediction module 703 is further used to use the target matching probability of the initial user positioning point and the road segment corresponding to the initial positioning time as the initial prediction condition probability of matching the initial user positioning point to each road segment;

[0136] Correspondingly, the determination module 702 is further used to determine, based on the target matching probability of each user positioning point and road segment, the preset road segment topological relationship and the initial prediction condition probability, according to the relationship between the matching probability of the positioning point and the road segment, the road segment topological relationship and the prediction probability, in accordance with the positioning time sequence of the positioning points, in the case where the positioning point corresponding to the first positioning time matches each road segment, the prediction condition probability that the positioning point corresponding to the second positioning time matches each road segment;

[0137] Correspondingly, the prediction module 703 is further configured to select a trajectory consisting of a road segment with the largest prediction condition probability matched by the positioning point corresponding to the second positioning time and road segments matched by all the positioning points corresponding to the first positioning time as the predicted trajectory.

[0138] In some embodiments, the user positioning data further includes the speed of the user positioning point, the road network road data further includes the direction of a preset road segment and a speed threshold of a preset road segment, and the device 700 for determining the road trajectory may further include a calculation module for determining the distance, average speed and first change angle of adjacent user positioning points corresponding to adjacent positioning times according to the coordinates, positioning time and speed of the user positioning point;

[0139] Correspondingly, the calculation module is also used to determine the average speed difference between the user positioning point and the corresponding road segment and the second change angle according to the average speed of the user positioning point, the first change angle, the direction of the road segment and the speed threshold of the road segment;

[0140] The correction module 704 is also used to use the distance between the adjacent user positioning points, the average speed, the first change angle, the average speed difference between the positioning point and the corresponding road segment, and the second change angle as input parameters of the feature function, and based on the input parameters of the feature function according to the pre-trained feature weights in the trajectory correction model, the relationship between the input parameters of the feature function and the target probability, determine in sequence according to the positioning time sequence of the positioning points that, when the positioning point corresponding to the first positioning time matches each target road segment, the target probability that the positioning point corresponding to the second positioning time matches each target road segment is obtained, so as to obtain a target trajectory consisting of road segments with the largest cumulative target probability.

[0141] In some embodiments, the road trajectory determination device 700 may further include:

[0142] The determination module 702 is further used to determine the road segment matched by each positioning point according to the target trajectory, and determine the starting and ending mileages of the target trajectory in the road segment according to the coordinates of the positioning point and the coordinates of the road segment;

[0143] The determination module 702 is further used to determine the topological relationship of the road segments matched by the positioning points corresponding to the adjacent positioning times;

[0144] A module is established to establish a linear reference system event sequence for representing a target trajectory. The linear reference system event sequence includes multiple linear reference system events. Each linear reference system event includes a preset identification ID of a road segment matched by a corresponding positioning point of the event, the start and end mileage of the target trajectory segment on the road segment, and a topological relationship of the road segment.

[0145] Figure 7 The various modules in the device shown can be implemented Figure 1 Each step in the above steps can achieve the corresponding technical effect, which will not be repeated here for the sake of brevity.

[0146] Figure 8 A schematic diagram of the hardware structure of a terminal device provided in an embodiment of the present application is shown.

[0147] The terminal device may include a processor 801 and a memory 802 storing computer program instructions.

[0148] Specifically, the processor 801 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0149] The memory 802 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 802 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In an example, the memory 802 may include a removable or non-removable (or fixed) medium, or the memory 802 is a non-volatile solid-state memory. The memory 802 may be inside or outside the integrated gateway disaster recovery device.

[0150] In one example, the memory 802 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method for determining a road trajectory according to the present disclosure.

[0151] The processor 801 reads and executes the computer program instructions stored in the memory 802 to implement Figure 1 The method for determining the road trajectory in the illustrated embodiment.

[0152] In one example, the terminal device may further include a communication interface 803 and a bus 804. Figure 8 As shown, the processor 801, the memory 802, and the communication interface 803 are connected via a bus 804 and communicate with each other.

[0153] The communication interface 803 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0154] Bus 804 includes hardware, software or both, and the components of terminal equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (Accelerated Graphics Port, AGP) or other graphics bus, enhanced industry standard architecture (Extended Industry Standard Architecture, EISA) bus, front side bus (Front Side Bus, FSB), Hyper Transport (Hyper Transport, HT) interconnection, industry standard architecture (Industry Standard Architecture, ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 804 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0155] In addition, in combination with the method for determining a road trajectory in the above embodiment, the embodiment of the present application may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the methods for determining a road trajectory in the above embodiment is implemented.

[0156] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, implements any one of the road trajectory determination methods in the above embodiments.

[0157] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0158] The functional blocks shown in the structural block diagram described above can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), appropriate firmware, plug-in, function card, etc. When implemented in software, the elements of the present application are programs or text segments used to perform the required tasks. The program or text segment can be stored in a machine-readable medium, or transmitted on a transmission medium or communication link via a data signal carried in a carrier. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (Read-Only Memory, ROM), flash memory, erasable read-only memory (Erasable ReadOnly Memory, EROM), floppy disks, compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical discs, hard disks, optical fiber media, radio frequency (Radio Frequency, RF) links, etc. The text segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0159] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0160] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0161] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A method for determining a road trajectory, characterized in that: include: Acquire user positioning data and road network data, wherein the user positioning data includes coordinates of multiple user positioning points and positioning times of the user positioning points, and the road network data includes coordinates of multiple preset road segments and preset road segment topological relationships; Based on the coordinates of the user positioning point, the coordinates of the preset road segment and the preset road segment topological relationship, determine the target matching probability of each user positioning point and the road segment according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability; Based on the target matching probability and the preset road segment topological relationship, according to the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, in accordance with the positioning time sequence of the positioning point, the predicted conditional probability that the positioning point corresponding to the second positioning time matches each road segment when the positioning point corresponding to the first positioning time matches each road segment is determined, and a predicted trajectory consisting of road segments with the largest predicted conditional probability is obtained, wherein the first positioning time is the positioning time of any user positioning point, and the second positioning time is the positioning time next to the first positioning time; The coordinates of the road segments of the predicted trajectory and the coordinates of the user positioning points are input into the trajectory correction model as input parameters of the feature function. According to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, the target probabilities of the positioning points corresponding to the second positioning time matching the target road segments are determined in sequence according to the positioning time of the positioning points, when the positioning points corresponding to the first positioning time match the target road segments, and the target trajectory consisting of the road segments with the largest cumulative target probability is obtained. The target road segments are the road segments in the predicted trajectory.

2. The method for determining a road trajectory according to claim 1, characterized in that: Before determining the target matching probability of each positioning point and the road segment according to the relationship between the coordinates of the positioning point, the coordinates of the road segment and the matching probability, the method further includes: Clean the user location data to remove missing and duplicate data in the fields of the user location data; The road network data is cleaned to remove duplicate coordinates of preset road segments and topological errors in the topological relationships of preset road segments.

3. The method for determining a road trajectory according to any one of claims 1-2, characterized in that: The method of determining the target matching probability of each positioning point and the road segment based on the coordinates of the user positioning point, the coordinates of the preset road segment and the preset road segment topological relationship according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability includes: Based on the coordinates of the user positioning point and the coordinates of the candidate road segment, according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, and the vertical distance between the positioning point and the road segment, determining the vertical distance between the user positioning point and the candidate road segment, wherein the candidate road segment is a road segment whose coordinates are within a preset range of the coordinates of the user positioning point; Based on the vertical distance between the user's positioning point and the candidate road segments, the preset road segment topological relationship, and the relationship between the vertical distance between the positioning point and the road segment, the road segment topological relationship, and the matching probability, the target matching probability of each user's positioning point and the road segment is determined.

4. The method for determining a road trajectory according to claim 1, characterized in that: The method is based on the target matching probability and the preset road segment topological relationship, according to the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, and in accordance with the positioning time sequence of the positioning point, sequentially determining the predicted conditional probability that the positioning point corresponding to the second positioning time matches each road segment when the positioning point corresponding to the first positioning time matches each road segment, and obtaining a predicted trajectory consisting of the road segment with the maximum predicted conditional probability, including: The target matching probability between the initial user positioning point and the road segment corresponding to the initial positioning time is used as the initial prediction conditional probability of matching the initial user positioning point to each road segment; Based on the target matching probability between each user's positioning point and the road segment, the preset road segment topological relationship and the initial prediction condition probability, according to the matching probability between the positioning point and the road segment, the relationship between the road segment topological relationship and the prediction probability, in accordance with the positioning time sequence of the positioning points, the prediction condition probability that the positioning point corresponding to the second positioning time matches each road segment is determined in turn when the positioning point corresponding to the first positioning time matches each road segment; A trajectory consisting of a road segment with the largest prediction condition probability matched by the positioning point corresponding to the second positioning time and road segments matched by all the positioning points corresponding to the first positioning time is selected as the predicted trajectory.

5. The method for determining a road trajectory according to claim 1, characterized in that: The user positioning data also includes the speed of the user positioning point, and the road network road data also includes the direction of the preset road segment and the speed threshold of the preset road segment. After obtaining the predicted trajectory consisting of the road segment with the maximum prediction condition probability, the method further includes: Determine the distance, average speed and first change angle of adjacent user positioning points corresponding to adjacent positioning times according to the coordinates, positioning time and speed of the user positioning points; Determine the average speed difference between the user positioning point and the corresponding road segment and the second change angle according to the average speed of the user positioning point, the first change angle, the direction of the road segment and the speed threshold of the road segment; The distance between the adjacent user positioning points, the average speed, the first change angle, the average speed difference between the positioning point and the corresponding road segment, and the second change angle are used as input parameters of the feature function. Based on the input parameters of the feature function and the relationship between the feature weights pre-trained in the trajectory correction model, the input parameters of the feature function and the target probability, the target probability of the positioning point corresponding to the second positioning time matching each target road segment when the positioning point corresponding to the first positioning time matches each target road segment is determined in sequence according to the positioning time of the positioning points, so as to obtain a target trajectory consisting of road segments with the largest cumulative target probability.

6. The method for determining a road trajectory according to claim 1, characterized in that: After obtaining the target trajectory consisting of the road segments with the largest cumulative target probability, the method further includes: Determine the road segment that matches each positioning point according to the target trajectory, and determine the starting and ending mileages of the target trajectory in the road segment according to the coordinates of the positioning point and the coordinates of the road segment; Determine the topological relationship of the road segments that match the positioning points corresponding to adjacent positioning times; A linear reference system event sequence for representing a target trajectory is established, wherein the linear reference system event sequence includes a plurality of linear reference system events, each linear reference system event includes a preset identification ID of a road segment matched by a positioning point corresponding to the event, the start and end mileages of the target trajectory segment on the road segment, and a topological relationship of the road segment.

7. A device for determining a road trajectory, characterized in that: The device comprises: An acquisition module, used to acquire user positioning data and road network data, wherein the user positioning data includes coordinates of multiple user positioning points and positioning times of the user positioning points, and the road network data includes coordinates of multiple preset road segments and preset road segment topological relationships; A determination module, for determining a target matching probability between each user positioning point and a road segment based on the coordinates of the user positioning point, the coordinates of the preset road segment and the preset road segment topological relationship, according to the relationship between the coordinates of the positioning point, the coordinates of the road segment, the road segment topological relationship and the matching probability; A prediction module is used to determine, based on the target matching probability and the preset road segment topological relationship, according to the matching probability of the positioning point and the road segment, the relationship between the road segment topological relationship and the conditional probability, in accordance with the positioning time sequence of the positioning point, in the case where the positioning point corresponding to the first positioning time matches the road segment, the predicted conditional probability that the positioning point corresponding to the second positioning time matches the road segment, and obtain a predicted trajectory consisting of the road segment with the maximum predicted conditional probability, wherein the first positioning time is the positioning time of any user positioning point, and the second positioning time is the positioning time next to the first positioning time; A correction module is used to input the coordinates of the road segments of the predicted trajectory and the coordinates of the user positioning points into a trajectory correction model as input parameters of the feature function, and determine, according to the relationship between the pre-trained feature weights in the trajectory correction model, the input parameters of the feature function and the target probability, in accordance with the positioning time sequence of the positioning points, the target probability that the positioning points corresponding to the second positioning time match the respective target road segments when the positioning points corresponding to the first positioning time match the respective target road segments, and obtain a target trajectory consisting of road segments with the largest cumulative target probability, wherein the target road segments are the road segments in the predicted trajectory.

8. A terminal device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the method for determining a road trajectory as claimed in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method for determining a road trajectory as described in any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method for determining a road trajectory as described in any one of claims 1-6.

Images