Vehicle travel path prediction method and storage medium

Abstract

Embodiments of the present application provide a vehicle driving path prediction method and a storage medium. The method comprises: obtaining, based on the card mouth passing data of a target vehicle that has not entered a warning area, an external driving path of the target vehicle outside the warning area as a first path; finding a second path matching the first path in the external driving path of the target vehicle in historical travel, and obtaining an internal driving path corresponding to the second path as a third path; performing route clustering on the third path to obtain a possible route; obtaining a prediction route of the target vehicle driving in the warning area based on the possible route; and determining the prediction time of the target vehicle reaching each card mouth on the prediction route to obtain a prediction path. Since the electronic device can predict the driving path of the target vehicle according to the card mouth passing data of the target vehicle, the driving path can be predicted even if GPS data cannot be obtained, thereby improving the success rate of predicting the driving path of the vehicle.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a method for predicting vehicle travel paths and a storage medium. Background Technology

[0002] In a comprehensive vehicle management scenario, specific vehicles face restrictions when entering pre-defined warning zones. For example, if the pre-defined warning zone is an urban area, trucks, as specific vehicles, are prohibited from entering the urban area between 06:00 and 22:00. Therefore, when it is determined that a specific vehicle may enter the pre-defined warning zone without lifting the restrictions, it is necessary to predict the vehicle's travel path after entering the warning zone, and then issue a warning based on the travel path, so that relevant personnel can intercept the specific vehicle based on the travel path.

[0003] Current methods for predicting vehicle routes are based on GPS (Global Positioning System) data. If the specific vehicle does not have a GPS device installed or its GPS device is actively turned off, then GPS data for that specific vehicle cannot be obtained, and therefore the above method cannot be used to predict its route. It can be seen that the above method can only be executed if vehicle GPS data can be successfully acquired, resulting in a low success rate in predicting vehicle routes. Summary of the Invention

[0004] The purpose of this application is to provide a vehicle driving path prediction method and storage medium to improve the success rate of predicting vehicle driving paths. The specific technical solution is as follows:

[0005] In a first aspect, embodiments of this application provide a vehicle driving path prediction method, the method comprising:

[0006] Based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, the external driving path of the target vehicle outside the warning area in the latest trip is obtained as the first path;

[0007] In the external driving paths of the target vehicle's historical trips, find the second path that matches the first path, and obtain the internal driving path within the warning area in the historical trips to which the second path belongs, as the third path;

[0008] The third path is clustered to obtain the possible routes for the target vehicle to travel within the warning area;

[0009] Based on the obtained possible routes, the predicted route for the target vehicle to travel within the warning area is obtained;

[0010] Based on the time when the target vehicle arrives at the last checkpoint in the first path, the predicted time when the target vehicle arrives at each checkpoint on the predicted route is determined, and the predicted path of the target vehicle in the warning area is obtained.

[0011] Optionally, determining the predicted arrival time of the target vehicle at each checkpoint on the predicted route based on the time the target vehicle arrives at the last checkpoint in the first path includes:

[0012] For each pair of adjacent checkpoints, based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, a first shortest route between checkpoints in the pair is determined, and the vehicle travel time of the first shortest route is obtained based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship; wherein, the pair of adjacent checkpoints includes: a checkpoint pair consisting of the last checkpoint in the first path and the first checkpoint in the predicted route, and a checkpoint pair consisting of adjacent checkpoints in the predicted route;

[0013] Based on the time it takes for the target vehicle to reach the last checkpoint on the first path and the obtained vehicle travel time, the predicted time for the target vehicle to reach each checkpoint on the predicted route is determined.

[0014] Optionally, obtaining the predicted route for the target vehicle to travel within the warning area based on the obtained possible routes includes:

[0015] Determine the number of cluster paths corresponding to each possible route, where each possible route corresponds to the third path obtained by route clustering.

[0016] The longest common route reflected by the clustering path corresponding to the target route is obtained as the predicted route for the target vehicle to travel within the warning area, wherein the target route is the possible route with the largest number of paths.

[0017] Optionally, the checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint; and the external driving path of the target vehicle outside the warning area during its latest trip, obtained based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, as the first path, including:

[0018] Obtain the vehicle passage data of the target vehicle within the latest preset time period;

[0019] Based on the arrival times at the checkpoints included in the checkpoint vehicle data, the checkpoints reached by the target vehicle within the latest preset time period are sorted to obtain the total path of the target vehicle;

[0020] Based on the checkpoint locations and arrival times at checkpoints in the total route, sub-routes corresponding to different trips are segmented from the total route;

[0021] Based on the location of the warning area, determine whether the sub-path corresponding to the latest trip of the target vehicle is located in the warning area;

[0022] If not, then the sub-path corresponding to the latest trip will be determined as the first path.

[0023] Optionally, the checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint; and the external driving path of the target vehicle outside the warning area during its latest trip, obtained based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, as the first path, including:

[0024] Obtain the vehicle passage data of the target vehicle within the latest preset time period;

[0025] Based on the location of the warning area and the vehicle passage data at the checkpoint, it is determined whether the target vehicle has entered the warning area within the latest preset time period;

[0026] If not, then according to the arrival time of the checkpoint included in the checkpoint vehicle data, the checkpoints reached by the target vehicle within the latest preset time period are sorted to obtain the total path of the target vehicle;

[0027] Based on the checkpoint locations and arrival times at checkpoints in the total route, sub-routes corresponding to different trips are segmented from the total route;

[0028] The sub-path corresponding to the latest trip is determined as the first path.

[0029] Optionally, the step of segmenting sub-paths corresponding to different trips from the total path based on the checkpoint locations and arrival times at checkpoints in the total path includes:

[0030] Based on the checkpoint locations and arrival times at checkpoints along the total path, calculate the distance between adjacent checkpoints and the vehicle travel time along the total path.

[0031] Identify adjacent checkpoints where the distance is greater than a first distance threshold and / or the vehicle travel time is greater than a first time threshold;

[0032] The total path is divided between the identified adjacent checkpoints to obtain sub-paths corresponding to different trips.

[0033] Optionally, the step of segmenting sub-paths corresponding to different trips from the total path based on the checkpoint locations and arrival times at checkpoints in the total path includes:

[0034] Based on the checkpoint locations and arrival times at checkpoints in the total path, the total path is divided to obtain initial sub-paths;

[0035] Determine the number of checkpoints included in the resulting initial sub-path;

[0036] The initial sub-path with a number of checkpoints greater than a preset checkpoint number threshold is determined as the sub-path corresponding to different trips segmented from the total path.

[0037] Optionally, determining the sub-path corresponding to the latest trip as the first path includes:

[0038] Identify adjacent checkpoints in the sub-path corresponding to the latest trip whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold, and designate them as adjacent checkpoints with missing paths;

[0039] Based on the checkpoint topology, the missing paths between adjacent checkpoints are restored to obtain the first path.

[0040] Optionally, the step of restoring the missing paths between adjacent checkpoints based on the pre-established checkpoint topology to obtain the first path includes:

[0041] Based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, the second shortest route between adjacent checkpoints where the path is missing is determined, and the vehicle travel time of the second shortest route is obtained based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship.

[0042] Determine whether the length of the second shortest route is less than a length threshold, and whether the vehicle travel time of the second shortest route is less than a third time threshold;

[0043] If so, the missing path between adjacent checkpoints is restored using the second shortest route to obtain the first path.

[0044] Optionally, the method for establishing the checkpoint topology includes:

[0045] Obtain the historical driving paths of sample vehicles that entered the warning area;

[0046] The total number of times each adjacent checkpoint pair appears in the historical driving path, the vehicle travel time between checkpoints in the sample checkpoint pair, and the distance between checkpoints in the sample checkpoint pair are statistically analyzed.

[0047] Select sample checkpoint pairs whose frequency of occurrence exceeds a preset threshold, whose vehicle travel time does not exceed a second time threshold, and whose distance does not exceed a second distance threshold.

[0048] The checkpoint topology is generated based on the selected sample checkpoint pairs, the frequency of occurrence of the selected sample checkpoint pairs, the vehicle travel time, and the distance.

[0049] Optionally, after obtaining the predicted path of the target vehicle within the warning area, the method further includes:

[0050] Among the internal driving paths in the target vehicle's historical trips, the internal driving path in the historical trips that is closest to the time of the latest trip is determined as the driving path to be compared.

[0051] Based on the similarity between the driving path to be compared and the predicted path, the confidence level corresponding to the predicted path is determined.

[0052] Secondly, embodiments of this application provide a vehicle driving path prediction device, the device comprising:

[0053] The first path determination module is used to obtain the external driving path of the target vehicle outside the warning area in its latest trip within the latest preset time period based on the checkpoint vehicle data of the target vehicle that has not entered the warning area, and use it as the first path;

[0054] The third path determination module is used to find a second path that matches the first path in the external driving paths of the target vehicle's historical trips, and to obtain the internal driving path of the second path in the warning area in the historical trips to which the second path belongs, as the third path;

[0055] The route clustering module is used to perform route clustering on the third path to obtain the possible routes for the target vehicle to travel within the warning area;

[0056] The predicted route determination module is used to obtain the predicted route of the target vehicle within the warning area based on the obtained possible routes;

[0057] The predicted path determination module is used to determine the predicted time of the target vehicle to each checkpoint on the predicted route based on the time when the target vehicle arrives at the last checkpoint in the first path, so as to obtain the predicted path of the target vehicle traveling in the warning area.

[0058] Optionally, the predicted path determination module includes: a time consumption determination unit, configured to determine, for each adjacent checkpoint pair, a first shortest route between checkpoints in the adjacent checkpoint pair based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, and obtain the vehicle travel time of the first shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship; wherein, the adjacent checkpoint pair includes: a checkpoint pair consisting of the last checkpoint in the first path and the first checkpoint in the predicted route, and a checkpoint pair consisting of adjacent checkpoints in the predicted route; and a prediction time determination unit, configured to determine the predicted time for the target vehicle to reach each checkpoint on the predicted route based on the time when the target vehicle arrives at the last checkpoint in the first path and the obtained vehicle travel time.

[0059] Optionally, the predicted route determination module includes: a path number determination unit, used to determine the number of clustered paths corresponding to each possible route, wherein the clustered path corresponding to each possible route is: the third path obtained by route clustering for the possible route; and a predicted route determination unit, used to obtain the longest common route reflected by the clustered path corresponding to the target route, as the predicted route for the target vehicle to travel in the warning area, wherein the target route is: the possible route with the largest number of paths.

[0060] Optionally, the checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint. The first path determination module includes: a data acquisition unit, used to acquire checkpoint vehicle passage data of the target vehicle within the latest preset time period; a total path determination unit, used to sort the checkpoints reached by the target vehicle within the latest preset time period according to the arrival time of the checkpoints included in the checkpoint vehicle passage data, to obtain the total path of the target vehicle; a path segmentation unit, used to segment sub-paths corresponding to different trips from the total path based on the checkpoint location and arrival time of the checkpoints in the total path; a first judgment unit, used to determine whether the sub-path corresponding to the latest trip of the target vehicle is located in the warning area based on the location of the warning area; and a first path determination unit, used to determine the sub-path corresponding to the latest trip as the first path when the judgment result of the first judgment unit is negative.

[0061] Optionally, the checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint; the first path determination module includes: a data acquisition unit, used to acquire checkpoint vehicle passage data of the target vehicle within the latest preset time period; a second judgment unit, used to determine whether the target vehicle has entered the warning area within the latest preset time period based on the location of the warning area and the checkpoint vehicle passage data; a total path determination unit, used to sort the checkpoints reached by the target vehicle within the latest preset time period according to the arrival time of the checkpoints included in the checkpoint vehicle passage data when the judgment result of the second judgment unit is negative, to obtain the total path of the target vehicle; a path segmentation unit, used to segment sub-paths corresponding to different trips from the total path based on the checkpoint location and arrival time of the checkpoints in the total path; and a first path determination unit, used to determine the sub-path corresponding to the latest trip as the first path.

[0062] Optionally, the path segmentation unit includes: a calculation subunit, used to calculate the distance between adjacent checkpoints and the vehicle travel time in the total path based on the checkpoint location and arrival time of the checkpoints in the total path; a checkpoint segmentation determination subunit, used to determine adjacent checkpoints whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold; and a path segmentation subunit, used to segment the total path between the determined adjacent checkpoints to obtain sub-paths corresponding to different trips.

[0063] Optionally, the path segmentation unit includes: a path segmentation sub-unit, used to segment the total path based on the checkpoint location and arrival time of the checkpoints in the total path to obtain initial sub-paths; a checkpoint number determination sub-unit, used to determine the number of checkpoints included in the obtained initial sub-paths; and a sub-path determination sub-unit, used to determine the initial sub-paths with a checkpoint number greater than a preset checkpoint number threshold as sub-paths corresponding to different trips segmented from the total path.

[0064] Optionally, the first path determination unit includes: a checkpoint restoration determination subunit, used to determine adjacent checkpoints in the sub-path corresponding to the latest trip whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold, as path missing adjacent checkpoints; and a restoration subunit, used to restore the missing path between the path missing adjacent checkpoints based on the checkpoint topology relationship, to obtain the first path.

[0065] Optionally, the restoration subunit is specifically used to determine the second shortest route between adjacent checkpoints with missing paths based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship record, and to obtain the vehicle travel time of the second shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship record; to determine whether the length of the second shortest route is less than a length threshold, and whether the vehicle travel time of the second shortest route is less than a third time threshold; if so, to restore the missing path between adjacent checkpoints with missing paths using the second shortest route, thereby obtaining the first path.

[0066] Optionally, the method for establishing the checkpoint topology includes: acquiring the historical driving paths of sample vehicles entering the warning area; counting the total number of times sample checkpoint pairs formed by adjacent checkpoints in the historical driving paths, the vehicle driving time between checkpoints in the sample checkpoint pairs, and the distance between checkpoints in the sample checkpoint pairs; selecting sample checkpoint pairs whose occurrence count is greater than a preset threshold, whose vehicle driving time is not greater than a second time threshold, and whose distance is not greater than a second distance threshold; and generating the checkpoint topology based on the selected sample checkpoint pairs and the corresponding occurrence count, vehicle driving time, and distance.

[0067] Optionally, the device further includes: a driving path determination module for comparison, used to determine, among the internal driving paths in the target vehicle's historical travels, the internal driving path in the historical travels that is closest in time to the latest travel, as the driving path to be compared; and a confidence determination module, used to determine the confidence level corresponding to the predicted path based on the similarity between the driving path to be compared and the predicted path.

[0068] Thirdly, embodiments of this application provide an electronic device, including:

[0069] Memory, used to store computer programs;

[0070] When a processor executes a program stored in memory, it implements any of the methods described in the first aspect above.

[0071] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described in the first aspect above.

[0072] Beneficial effects of the embodiments in this application:

[0073] In the solution provided in this application embodiment, the electronic device can obtain the external driving path of the target vehicle outside the warning area in its latest trip within the latest preset time period based on the checkpoint vehicle data of the target vehicle that has not entered the warning area, as the first path; in the external driving paths in the target vehicle's historical trips, find the second path that matches the first path, and obtain the internal driving path in the warning area in the historical trip to which the second path belongs, as the third path; perform route clustering on the third path to obtain the possible routes of the target vehicle in the warning area; based on the obtained possible routes, obtain the predicted route of the target vehicle in the warning area; based on the time when the target vehicle arrives at the last checkpoint in the first path, determine the predicted time when the target vehicle arrives at each checkpoint on the predicted route, and obtain the predicted route of the target vehicle in the warning area. Since most roads now have checkpoints, when a target vehicle passes through a checkpoint, the equipment installed at the checkpoint can acquire the vehicle's checkpoint passage data. Then, electronic devices can predict the target vehicle's travel path based on this data. Therefore, even without GPS data, it's possible to predict the target vehicle's path, thus improving the success rate of path prediction. Furthermore, because the predicted path for the target vehicle within the warning area is based on its historical travel routes, the resulting predicted path closely matches the vehicle's actual travel situation.

[0074] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0075] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0076] Figure 1 A flowchart of a vehicle driving path prediction method provided in an embodiment of this application;

[0077] Figure 2 Based on Figure 1 A flowchart illustrating a method for determining prediction time in the illustrated embodiment;

[0078] Figure 3 Based on Figure 2 A schematic diagram of a bayonet topology in the embodiment shown;

[0079] Figure 4 Based on Figure 1A flowchart illustrating a method for determining a predicted route in the illustrated embodiment;

[0080] Figure 5 Based on Figure 1 A flowchart illustrating one method for determining a first path in the illustrated embodiment;

[0081] Figure 6 Based on Figure 1 A flowchart illustrating another method for determining the first path in the illustrated embodiment;

[0082] Figure 7 Based on Figure 5 or Figure 6 A flowchart of a path segmentation method in the illustrated embodiment;

[0083] Figure 8 Based on Figure 5 or Figure 6 A flowchart of another path segmentation method in the illustrated embodiment;

[0084] Figure 9 Based on Figure 5 or Figure 6 A flowchart illustrating one method for determining a first path in the illustrated embodiment;

[0085] Figure 10 Based on Figure 9 A flowchart of one restoration method of the illustrated embodiment;

[0086] Figure 11 Based on Figure 1 A flowchart illustrating a confidence level determination method in the illustrated embodiment;

[0087] Figure 12 Based on Figure 2 A flowchart illustrating a method for establishing a checkpoint topology in the illustrated embodiment;

[0088] Figure 13 Based on Figure 1 A flowchart illustrating one method for establishing a correspondence in the illustrated embodiment;

[0089] Figure 14 Based on Figure 13 A schematic diagram of path segmentation in the illustrated embodiment;

[0090] Figure 15 A detailed flowchart of a vehicle driving path prediction method provided in an embodiment of this application;

[0091] Figure 16 This is a schematic diagram of the structure of a vehicle driving path prediction device provided in an embodiment of this application;

[0092] Figure 17This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0093] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0094] To improve the success rate of predicting vehicle travel paths, this application provides a vehicle travel path prediction method, apparatus, electronic device, computer-readable storage medium, and computer program product. The following is a description of a vehicle travel path prediction method provided by an embodiment of this application.

[0095] The vehicle path prediction method provided in this application can be applied to any electronic device that needs to predict vehicle paths, such as a mobile phone, computer, or other terminal device, or a server; no specific limitation is made here. For clarity, it will be referred to as an electronic device below.

[0096] like Figure 1 As shown, a vehicle driving path prediction method includes:

[0097] S101, based on the checkpoint vehicle data of the target vehicle that has not entered the warning area within the latest preset time period, obtain the external driving path of the target vehicle outside the warning area in the latest trip, as the first path.

[0098] S102, in the external driving paths of the target vehicle's historical trips, find a second path that matches the first path, and obtain the internal driving path within the warning area in the historical trips to which the second path belongs, as the third path.

[0099] S103, perform route clustering on the third path to obtain the possible routes for the target vehicle to travel within the warning area.

[0100] S104, Based on the obtained possible routes, obtain the predicted route of the target vehicle traveling within the warning area.

[0101] S105, based on the time when the target vehicle arrives at the last checkpoint in the first path, determine the predicted time when the target vehicle arrives at each checkpoint on the predicted route, and obtain the predicted path of the target vehicle traveling in the warning area.

[0102] In the solution provided in this application embodiment, since most roads currently have checkpoints, when a target vehicle passes through a checkpoint, the device installed at the checkpoint can acquire the target vehicle's checkpoint passage data. Furthermore, the electronic device can predict the target vehicle's travel path based on the checkpoint passage data captured by the devices installed at each checkpoint. Therefore, even if the target vehicle's GPS data cannot be obtained, the travel path can still be predicted, thereby improving the success rate of predicting the vehicle's travel path. Moreover, because the predicted path of the target vehicle within the warning area is based on the target vehicle's historical travel paths, the obtained predicted path closely matches the target vehicle's actual travel situation.

[0103] The electronic device can store vehicle information of vehicles that the user wants to follow; these vehicles are the target vehicles. The user can send commands to the electronic device to trigger it to predict the driving path of the target vehicle. Of course, the electronic device can also trigger the prediction of the target vehicle's driving path automatically, for example, by using a timer installed on it to periodically trigger the prediction of the target vehicle's driving path.

[0104] Since the target vehicle has already entered the pre-set warning area, the actual path of the target vehicle within the warning area can be directly obtained based on the checkpoint vehicle data within the latest preset time period, without the need to predict the target vehicle's path within the warning area. Therefore, after being triggered, the electronic device can only target vehicles that have not entered the warning area, and obtain their latest external travel path outside the warning area based on the checkpoint vehicle data within the latest preset time period, as the first path, i.e., execute the above step S101.

[0105] The aforementioned latest preset duration can be a preset duration prior to the current moment when the electronic device is triggered. The format of the aforementioned checkpoint vehicle passage data can be (data ID, vehicle information, timestamp, checkpoint ID), where data ID is the identification information of the checkpoint vehicle passage data, vehicle information is the representation information of the target vehicle, timestamp is the arrival time of the target vehicle at the checkpoint, and checkpoint ID is the ID of the data collection device installed at the checkpoint that collected the checkpoint vehicle passage data.

[0106] The electronic device generates a path based on checkpoint vehicle passage data in several ways. One method is to sort the checkpoint IDs included in the vehicle passage data according to their arrival times at the checkpoints, and then generate a path by sorting the checkpoint locations or names corresponding to those checkpoint IDs from pre-stored checkpoint information. No specific limitation is made here.

[0107] Taking the method of generating a path by sorting the checkpoint IDs included in the checkpoint vehicle passage data as an example, assuming that the checkpoint IDs and arrival times of the target vehicle in the latest preset time period are (0001, 2023.01.01 10:00:09), (0003, 2023.01.01 10:10:00), (0005, 2023.01.01 09:55:45), (0008, 2023.01.01 09:41:30), (0009, 2023.01.01 10:08:20) and (0010, 2023.01.01 09:48:20). The electronic device can sort the checkpoint IDs included in the checkpoint passage data according to the arrival time of the checkpoint, and obtain the travel path of the target vehicle as {(0008, 2023.01.01 09:41:30), (0010, 2023.01.01 09:48:20), (0005, 2023.01.01 09:55:45), (0001, 2023.01.01 10:00:09), (0009, 2023.01.01 10:08:20), (0003, 2023.01.01 10:10:00)}.

[0108] Therefore, after being triggered, the electronic device can obtain the checkpoint vehicle data of the target vehicle that has not entered the warning area within the latest preset time period. Based on the obtained checkpoint vehicle data, the device generates the external driving path of the target vehicle outside the warning area within the latest preset time period through the above-mentioned path generation method, and then obtains the first path based on the external driving path.

[0109] In one embodiment, after obtaining the aforementioned external driving path, the electronic device can directly determine the external driving path as the external driving path of the target vehicle outside the warning area in its latest trip, thus obtaining the first path.

[0110] In another implementation, after obtaining the aforementioned external travel path, the electronic device can segment sub-paths corresponding to different trips based on the checkpoint locations and arrival times at checkpoints along the external travel path. Finally, the sub-path with the latest arrival time is determined as the external travel path outside the warning area for the target vehicle's latest trip, thus obtaining the first path. The sub-path with the latest arrival time can be the sub-path with the latest arrival time at the first checkpoint in the path, the sub-path with the latest arrival time at the last checkpoint in the path, or the sub-path with the latest average arrival time at all checkpoints in the path; no specific limitation is made here.

[0111] After obtaining the first path, the electronic device can search for the second path that matches the first path in the external driving paths of the target vehicle's historical trips, and then obtain the internal driving path within the warning area in the historical trips to which the second path belongs, as the third path, that is, execute the above step S102. The second path that matches the first path can be a preset number of external driving paths with the highest similarity to the first path.

[0112] In one implementation, after obtaining a first path, the electronic device can calculate the similarity between the first path and the stored external driving paths in the target vehicle's historical travel history. Then, it identifies a preset number of external driving paths with the highest similarity to the first path as the matching external driving paths, serving as the second path. Furthermore, for each historical trip of the target vehicle entering the warning area, the electronic device can establish a correspondence between external driving paths and internal driving paths in that historical trip. Thus, after obtaining the second path, the electronic device can obtain the corresponding internal driving path based on the aforementioned correspondence, i.e., obtain the internal driving path within the warning area in the historical trip to which the second path belongs, serving as the third path.

[0113] For example, for the first to nth historical trips of a target vehicle entering a warning area, the electronic device can establish correspondences between external driving paths a1 and internal driving paths b1 in the first historical trip, external driving paths a2 and internal driving paths b2 in the second historical trip, and so on, between external driving paths an and internal driving paths bn in the nth historical trip. In this way, after obtaining the first path, the electronic device can calculate the similarity between the first path and the external driving paths a1, a2, ..., an in the aforementioned correspondences.

[0114] Assume the calculated similarity between the first path and external driving paths a1, a2, ..., an is 0.5, 0.7, ..., 0.3 respectively, with a preset number of 10. Since these 10 external driving paths a1, a2, a5, ..., am are the preset number of external driving paths with the highest similarity to the first path, they can be identified as the external driving paths matching the first path, thus obtaining the second path. Furthermore, the electronic device can, based on the established correspondence, obtain the internal driving paths b1, b2, b5, ..., bm corresponding to each of the second paths, as the third path.

[0115] Since the second path matches the first path, the third path is very likely to match the internal driving path corresponding to the first path. Therefore, after obtaining the third path, the electronic device can perform route clustering on the third path to obtain the possible routes for the target vehicle to travel within the warning area, that is, to execute the above step S103.

[0116] In one implementation, the electronic device can remove the time information included in the third path to obtain the route reflected by the third path. That is, the route includes the various checkpoints the vehicle passes through and the order of these checkpoints, while the path is the route plus the time it takes for the vehicle to reach each checkpoint along the route. After obtaining the routes reflected by each third path, the electronic device can cluster the routes reflected by each third path based on the distance between them using a path clustering algorithm such as LCS (Longest Common Subsequence). After clustering, the electronic device can use the resulting clusters as possible routes. The method of clustering the routes reflected by each third path based on the distance between them can also use clustering algorithms such as K-Means; no specific limitation is made here.

[0117] For example, after an electronic device uses internal driving paths b1, b2, b5, ... bm as the third path, it can remove the time information included in each internal driving path to obtain routes b1', b2', b5', ... bm'. Then, based on the distance between each route, it can cluster routes b1', b2', b5', ... bm'.

[0118] Assuming that after route clustering, routes b1', b2', and bm' are clustered into one class, routes b5', b8', and b10' are clustered into another class, and routes b7', b11', b12', and b13' are clustered into yet another class, then these three classes represent the three possible routes that the target vehicle might take within the warning area.

[0119] After obtaining the possible routes, the electronic device can obtain the predicted route of the target vehicle within the warning area based on the obtained possible routes, that is, execute the above step S104.

[0120] In one implementation, the electronic device can determine, for each possible route, the third path among the third paths obtained by route clustering as the cluster center, and then determine the third path among the third paths that is closest to the latest travel time. The route reflected by the third path is determined as the predicted route for the target vehicle to travel within the warning area. Since each third path is a travel path in the target vehicle's historical travels, the third path that is closest to the latest travel time is actually the third path with the latest time information. The method for determining the path with the latest time information has been explained in the foregoing and will not be repeated here.

[0121] In another implementation, the electronic device can perform route clustering to obtain the number of third paths of possible routes, determine the possible route with the largest number of third paths as the target route, and then determine the third path that is the cluster center among the various third paths of the target route, or the third path that is closest to the latest travel time among the various third paths of the target route, as the predicted route for the target vehicle to travel within the warning area.

[0122] Because the predicted route only includes each checkpoint and the order in which the target vehicle arrives at each checkpoint, but does not include the arrival time of the target vehicle at each checkpoint, after obtaining the predicted route of the target vehicle traveling within the warning area, the electronic equipment can further determine the predicted time of the target vehicle arriving at each checkpoint on the predicted route based on the time of the target vehicle arriving at the last checkpoint in the first path, so as to obtain the predicted route of the target vehicle traveling within the warning area.

[0123] In one implementation, the electronic device can calculate the distance between adjacent checkpoints in the predicted route, and the distance between the last checkpoint in the first path and the first checkpoint in the predicted route, based on the checkpoint locations of each checkpoint in the predicted route. Then, based on the distance between adjacent checkpoints and the target speed, it can calculate the vehicle travel time between adjacent checkpoints in the predicted route, and based on the distance between the last checkpoint in the first path and the first checkpoint in the predicted route and the target speed, it can calculate the vehicle travel time between the last checkpoint in the first path and the first checkpoint in the predicted route. Finally, based on the time the target vehicle arrives at the last checkpoint in the first path, the vehicle travel time between the last checkpoint in the first path and the first checkpoint in the predicted route, and the vehicle travel time between adjacent checkpoints in the predicted route, the predicted time for the target vehicle to arrive at each checkpoint on the predicted route is determined, thus obtaining the predicted path for the target vehicle to travel within the warning area. The target speed can be a pre-selected speed or the average speed of the target vehicle on the first path; no specific limitation is made here.

[0124] In the solution provided in this application embodiment, since most roads currently have checkpoints, when a target vehicle passes through a checkpoint, the device installed at the checkpoint can acquire the target vehicle's checkpoint passage data. Furthermore, the electronic device can predict the target vehicle's travel path based on the checkpoint passage data captured by the devices installed at each checkpoint. Therefore, even if the target vehicle's GPS data cannot be obtained, the travel path can still be predicted, thereby improving the success rate of predicting the vehicle's travel path. Moreover, because the predicted path of the target vehicle within the warning area is based on the target vehicle's historical travel paths, the obtained predicted path closely matches the target vehicle's actual travel situation.

[0125] As one implementation method of this application, such as Figure 2 As shown, the above-mentioned determination of the predicted arrival time of the target vehicle at each checkpoint on the predicted route based on the time when the target vehicle arrives at the last checkpoint in the first path may include:

[0126] S201, for each pair of adjacent checkpoints, based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, determine the first shortest route between checkpoints in the pair of adjacent checkpoints, and obtain the vehicle travel time of the first shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship.

[0127] To more accurately determine the predicted time for the target vehicle to arrive at each checkpoint on the predicted route, the electronic device can determine the vehicle travel time between adjacent checkpoints on the predicted route based on the pre-established checkpoint topology. The aforementioned checkpoint topology can be established based on the vehicle travel time of the sample vehicle when traveling between adjacent checkpoints and the distance between each adjacent checkpoint. The specific establishment method will be described in detail in subsequent embodiments and will not be repeated here.

[0128] In one implementation, the above-mentioned checkpoint topology can be... Figure 3 The directed weighted graph shown in the figure represents each checkpoint in the topology. The arrows between the black dots represent the driving directions between adjacent checkpoints. Each arrow can have corresponding weight information, which records the distance between adjacent checkpoints and the average vehicle travel time when the sample vehicle travels between adjacent checkpoints. In other words, the weight information records the distance between adjacent checkpoints and the vehicle travel time.

[0129] For the predicted route, adjacent checkpoints in the predicted route can form adjacent checkpoint pairs. At the same time, the last checkpoint in the first path and the first checkpoint in the predicted route can also form adjacent checkpoint pairs. For these adjacent checkpoint pairs, the electronic device can determine the shortest route between checkpoints in the adjacent checkpoint pair based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, and obtain the vehicle travel time of the shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship.

[0130] For example, if the last checkpoint in the first path is checkpoint A, and the checkpoints included in the predicted route are checkpoints B, C, and D in sequence, then checkpoints A and B can form an adjacent checkpoint pair, checkpoints B and C can form an adjacent checkpoint pair, and checkpoints C and D can form an adjacent checkpoint pair.

[0131] The electronic device determines the drivable routes between checkpoints A and B based on the drivable directions between adjacent checkpoints recorded in the checkpoint topology relationship. These routes include: drivable route 1 (checkpoint A → checkpoint E → checkpoint F → checkpoint B) and drivable route 2 (checkpoint A → checkpoint G → checkpoint B). Furthermore, based on the distances between adjacent checkpoints recorded in the checkpoint topology relationship, the electronic device determines that the length of drivable route 1 is 10km and the length of drivable route 2 is 4km. Since the length of drivable route 1 is greater than the length of drivable route 2, the electronic device can use drivable route 2 as the first shortest route between checkpoints A and B.

[0132] In one implementation, the electronic device can determine the first shortest route between adjacent checkpoints based on the distance between them recorded in the checkpoint topology relationship, using shortest path algorithms such as Dijkstra's algorithm.

[0133] Then, based on the vehicle travel time records between adjacent checkpoints in the checkpoint topology, the electronic device finds that the vehicle travel time between checkpoint A and checkpoint G is 10 minutes, and the vehicle travel time between checkpoint G and checkpoint B is 6 minutes, thus determining that the vehicle travel time for the first shortest route is 16 minutes. Similarly, the electronic device uses the above steps to determine that the vehicle travel time for the first shortest route between checkpoint B and checkpoint C is 10 minutes, and the vehicle travel time for the first shortest route between checkpoint C and checkpoint D is 12 minutes.

[0134] S202, based on the time when the target vehicle arrives at the last checkpoint in the first path and the obtained vehicle travel time, determine the predicted time when the target vehicle arrives at each checkpoint on the predicted route.

[0135] After obtaining the vehicle travel time of the shortest route between checkpoints in each adjacent checkpoint pair, the electronic device can, based on the time it takes for the target vehicle to reach the last checkpoint in the first path, sequentially accumulate the vehicle travel time corresponding to each adjacent checkpoint pair according to the order of the adjacent checkpoint pairs in the predicted route, thereby determining the predicted time for the target vehicle to reach each checkpoint on the predicted route.

[0136] For example, if the target vehicle arrives at the last checkpoint on the first route at 12:00:00, the electronic device can calculate the predicted arrival time of the target vehicle at checkpoint B as 12:16:00, based on the order of adjacent checkpoint pairs on the predicted route and adding the travel time of the vehicles corresponding to the adjacent checkpoint pair formed by checkpoints A and B (16 minutes). Then, it can calculate the predicted arrival time of the target vehicle at checkpoint C as 12:26:00, by adding the travel time of the vehicles corresponding to the adjacent checkpoint pair formed by checkpoints A and B (16 minutes) and the travel time of the vehicles corresponding to the adjacent checkpoint pair formed by checkpoints B and C (10 minutes). Based on this time, the travel time of vehicles corresponding to the adjacent checkpoints formed by checkpoints A and B (16 minutes), the travel time of vehicles corresponding to the adjacent checkpoints formed by checkpoints B and C (10 minutes), and the travel time of vehicles corresponding to the adjacent checkpoints formed by checkpoints C and D (12 minutes) are added to determine the predicted time for the target vehicle to arrive at checkpoint D on the predicted route as 12:38:00.

[0137] In the solution provided in this application embodiment, because the electronic device determines the vehicle travel time of the target vehicle between adjacent checkpoints by using the average vehicle travel time of the sample vehicle when traveling between adjacent checkpoints as reflected in the checkpoint topology, the predicted time of the target vehicle to each checkpoint on the predicted route can be well matched with the actual driving situation of the vehicle on the predicted route, and can more accurately determine the predicted time of the target vehicle to each checkpoint on the predicted route based on the time of the target vehicle to reach the last checkpoint in the first path and the obtained vehicle travel time.

[0138] As one implementation method of this application, such as Figure 4 As shown, the predicted route for the target vehicle to travel within the warning area, based on the obtained possible routes, may include:

[0139] S401, determine the number of clustering paths corresponding to each possible route.

[0140] Since the possible routes are obtained by clustering the third path, each of which represents a class in the clustering result, the more third paths the possible route has, that is, the more clustered paths the possible route corresponds to, the more times the target vehicle has traveled along the possible route in its historical trips.

[0141] Therefore, electronic devices can determine the number of clustered paths corresponding to each possible route, thereby determining the possible route with the largest number of paths. The possible route with the largest number of paths is the route that the target vehicle travels most frequently in its historical trips.

[0142] S402, obtain the longest common route reflected by the clustering path corresponding to the target route, and use it as the predicted route for the target vehicle to travel in the warning area.

[0143] After determining the most likely route with the largest number of paths, i.e., after determining the target route, the electronic device can determine the longest common route reflected by the clustered paths corresponding to the target route by calculating the longest common subsequence of each route, and then use the longest common route as the predicted route for the target vehicle to travel within the warning area.

[0144] For example, after determining the number of clustered paths corresponding to each possible route, the electronic device determines the target route as possible route 1. The clustered paths corresponding to possible route 1 reflect routes b1', b2', and bm', respectively. Among them, the checkpoint sequence corresponding to route b1' is (checkpoint A, checkpoint B, checkpoint C, checkpoint E), the checkpoint sequence corresponding to route b2' is (checkpoint B, checkpoint C, checkpoint E), and the checkpoint sequence corresponding to route bm' is (checkpoint B, checkpoint C, checkpoint E, checkpoint G, checkpoint H).

[0145] The electronic device obtains the longest common route corresponding to the clustered path of the target route (checkpoint B, checkpoint C, checkpoint E) by calculating the longest common subsequence of each route based on the checkpoint sequence corresponding to route b1', route b2', and route bm'. Then, the longest common route can be used as the predicted route for the target vehicle to travel within the warning area.

[0146] The solution provided in this application embodiment allows the electronic device to determine the predicted route of a target vehicle by using the longest common route reflected by the clustered paths corresponding to the most likely route as the predicted route for the target vehicle to travel within the warning area. This is because the most likely route is the route the target vehicle has traveled most frequently in its historical travels, and the longest common route reflected by the clustered paths corresponding to the likely route is the route the target vehicle has traveled most frequently when traveling along that likely route. Therefore, the determined predicted route may closely match the target vehicle's historical travel patterns.

[0147] As one implementation method of this application, such as Figure 5 As shown, the aforementioned checkpoint vehicle passage data may include: the arrival time of the target vehicle at the checkpoint; based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, the external driving path of the target vehicle outside the warning area during its latest trip is obtained, which, as the first path, may include:

[0148] S501, obtain the vehicle passage data of the target vehicle within the latest preset time period at the checkpoint.

[0149] Because checkpoint vehicle passage data can include the arrival time of the target vehicle at the checkpoint, electronic devices can obtain the checkpoint passage data of the target vehicle within the latest preset time period by acquiring the checkpoint data including the arrival time at the checkpoint within the latest preset time period.

[0150] S502, according to the arrival time of the checkpoint included in the checkpoint vehicle data, sort the checkpoints that the target vehicle arrives at within the latest preset time period to obtain the total path of the target vehicle.

[0151] In addition to the arrival time of the target vehicle at the checkpoint, the vehicle passage data at the checkpoint may also include the checkpoint ID, which is the ID of the data collection device installed at the checkpoint that collected the vehicle passage data. Furthermore, the electronic device may pre-record the checkpoint information for each checkpoint, in the format of (checkpoint ID, checkpoint name, checkpoint location).

[0152] In one implementation, since the checkpoint IDs included in the checkpoint passage data of the target vehicle within the latest preset time period can reflect the checkpoints reached by the target vehicle within the latest preset time period, the electronic device can sort the checkpoint IDs included in the checkpoint passage data according to the arrival time of the checkpoints included in the checkpoint passage data to obtain the total path of the target vehicle.

[0153] In another implementation, after obtaining vehicle passage data from each checkpoint, the electronic device can query the checkpoint information to obtain the checkpoint location or checkpoint name corresponding to each checkpoint passage data based on the checkpoint ID included in the checkpoint passage data. Then, the electronic device can sort the checkpoint locations or checkpoint names corresponding to the checkpoint passage data according to the arrival time of the checkpoints included in the checkpoint passage data to obtain the total path of the target vehicle.

[0154] In another embodiment, after obtaining the checkpoint process data, the electronic device can, for each checkpoint vehicle passage data, query the corresponding checkpoint information in each checkpoint information based on the checkpoint ID included in the checkpoint vehicle passage data. Then, it associates the checkpoint vehicle passage data with the checkpoint information to obtain the passage point. The format of the passage point can be (data ID, vehicle information, timestamp, checkpoint ID, checkpoint name, checkpoint location). Furthermore, the electronic device can sort the passage points according to their arrival times at the checkpoints to obtain the total path of the target vehicle.

[0155] S503, based on the checkpoint location and arrival time of the checkpoints in the total path, sub-paths corresponding to different trips are segmented from the total path.

[0156] The target vehicle may make multiple trips within the latest preset time period, but only the latest trip can more accurately reflect the target vehicle's subsequent driving plan. For example, if a truck first loads cargo at a first location outside the warning area and then travels to a second location outside the warning area within the latest preset time period, then travels from the second location outside the warning area to a third location outside the warning area to load cargo, and finally loads cargo again at the third location outside the warning area before heading towards the warning area, only the last trip can accurately reflect the target vehicle's driving plan as it is about to enter the warning area.

[0157] Furthermore, because there may be a certain time interval between different trips, there may be a large time difference between the arrival times of two adjacent checkpoints. In addition, because there may be changes in the routes between different trips, there may be a large distance between two adjacent checkpoints.

[0158] Therefore, after obtaining the total path of the target vehicle within the latest preset time period based on the vehicle passage data at the checkpoints, the electronic device can divide the total path into sub-paths corresponding to different trips based on the checkpoint locations and arrival times of each checkpoint included in the total path. This is done according to the principle that if the time difference between the arrival times of adjacent checkpoints is large or the distance between the checkpoint locations of adjacent checkpoints is large, then the device is considered to be two separate trips. This process determines the sub-path corresponding to the latest trip.

[0159] S504, Based on the location of the warning area, determine whether the sub-path corresponding to the latest trip of the target vehicle is located in the warning area.

[0160] Because if a target vehicle has already entered a pre-defined warning zone, the actual path the vehicle traveled within the warning zone can be directly obtained based on the vehicle's checkpoint passage data within the latest preset time period, eliminating the need for further path prediction. Therefore, after obtaining the sub-path corresponding to the target vehicle's latest trip, the electronic device can determine whether the sub-path is located within the warning zone based on its position, thus confirming whether the target vehicle has entered the warning zone.

[0161] In one implementation, the electronic device can determine the location of the last checkpoint in the sub-path corresponding to the latest trip, and then, based on the location of the last checkpoint and the location of the warning area, determine whether the last checkpoint is located in the warning area. If so, it is determined that the sub-path corresponding to the latest trip of the target vehicle is located in the warning area.

[0162] In another implementation, the electronic device can determine the location of each checkpoint in the sub-path corresponding to the latest trip, and then, based on the location of each checkpoint and the location of the warning area, determine whether there is a checkpoint located in the warning area. If so, it is determined that the sub-path corresponding to the latest trip of the target vehicle is located in the warning area.

[0163] If it is determined that the sub-path corresponding to the latest trip of the target vehicle is not located in the warning area, the electronic device may execute step S505. If it is determined that the sub-path corresponding to the latest trip of the target vehicle is located in the warning area, the electronic device may output the sub-path corresponding to the latest trip.

[0164] S505, the sub-path corresponding to the latest trip is determined as the first path.

[0165] When an electronic device determines that the sub-path corresponding to the latest trip of a target vehicle is not located in the warning area, it can identify the sub-path corresponding to the latest trip as the first path, and then predict the driving path of the target vehicle within the warning area based on the first path.

[0166] In the solution provided in this application embodiment, the electronic device can determine the sub-path corresponding to the latest trip of the target vehicle as the external driving path of the target vehicle outside the warning area. This latest trip sub-path can more accurately reflect the driving plan of the target vehicle. Therefore, when predicting the driving path of the target vehicle within the warning area based on this latest trip sub-path, a more accurate predicted path can be obtained. Furthermore, after obtaining the sub-path corresponding to the latest trip, the electronic device can also determine whether the target vehicle has entered the warning area based on this sub-path. When it is determined that the target vehicle has entered the warning area, the prediction step can be omitted, thereby saving computing resources. In this case, the electronic device can also output the sub-path corresponding to the latest trip as the actual path representing the target vehicle's driving within the warning area, thereby outputting the driving path of the target vehicle within the warning area without prediction.

[0167] As one implementation method of this application, such as Figure 6 As shown, the aforementioned checkpoint vehicle passage data may include: the arrival time of the target vehicle at the checkpoint; based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, the external driving path of the target vehicle outside the warning area during its latest trip is obtained, which, as the first path, may include:

[0168] S601, obtain the vehicle passage data of the target vehicle within the latest preset time period at the checkpoint.

[0169] Step S601 is the same as step S501 above, and will not be repeated here.

[0170] S602, based on the location of the warning area and the vehicle passage data at the checkpoint, determine whether the target vehicle has entered the warning area within the latest preset time period.

[0171] Because if the target vehicle has already entered the pre-set warning area, the actual path the target vehicle traveled within the warning area can be directly obtained based on the checkpoint vehicle passage data within the latest preset time period, without the need to predict the target vehicle's path within the warning area. Therefore, after acquiring the checkpoint vehicle passage data, the electronic equipment can determine whether the target vehicle has entered the warning area within the latest preset time period based on the location of the pre-set warning area and the acquired checkpoint vehicle passage data.

[0172] In one implementation, the electronic device can determine the latest arrival time of the vehicle passing data at each checkpoint among the acquired vehicle passing data at each checkpoint, and determine the checkpoint location corresponding to the vehicle passing data at that checkpoint. Then, it can determine whether the checkpoint location corresponding to the vehicle passing data at that checkpoint is within the warning area. If so, it can determine that the target vehicle has entered the warning area within the latest preset time period.

[0173] In another implementation, the electronic device can determine the location of the checkpoint corresponding to the acquired vehicle passage data of each checkpoint, and then determine whether there is a checkpoint location in the warning area among the checkpoint locations corresponding to the vehicle passage data of each checkpoint. If so, it is determined that the target vehicle enters the warning area within the latest preset time period.

[0174] If it is determined that the sub-path corresponding to the latest trip of the target vehicle is not located in the warning area, the electronic device may execute step S603. If it is determined that the sub-path corresponding to the latest trip of the target vehicle is located in the warning area, the electronic device may stop processing the checkpoint passage data of the target vehicle.

[0175] S603, according to the arrival time of the checkpoint included in the checkpoint vehicle data, sort the checkpoints that the target vehicle arrives at within the latest preset time period to obtain the total path of the target vehicle.

[0176] S604, based on the checkpoint location and arrival time at the checkpoint in the total path, sub-paths corresponding to different trips are segmented from the total path.

[0177] S605, the sub-path corresponding to the latest trip is determined as the first path.

[0178] The steps S603-S605 are the same as steps S502, S503 and S505, respectively, and will not be repeated here.

[0179] In the solution provided in this application embodiment, the electronic device can determine the sub-path corresponding to the latest trip of the target vehicle as the external driving path of the target vehicle outside the warning area. This latest trip sub-path can more accurately reflect the driving plan of the target vehicle. Therefore, when predicting the driving path of the target vehicle within the warning area based on this latest trip sub-path, a more accurate predicted path can be obtained. Furthermore, after obtaining the checkpoint vehicle passage data of the target vehicle within the latest preset time period, the electronic device can also determine whether the target vehicle has entered the warning area based on the checkpoint vehicle passage data. When it is determined that the target vehicle has entered the warning area, the steps of determining the total path, dividing the path, and prediction can be omitted, thereby saving computing resources.

[0180] As one implementation method of this application, such as Figure 7As shown, the above-mentioned sub-paths for different trips, based on the checkpoint locations and arrival times at checkpoints in the total path, can include:

[0181] S701, based on the checkpoint location and arrival time of the checkpoints in the total path, calculate the distance between adjacent checkpoints and the vehicle travel time in the total path.

[0182] Because there may be a certain time interval between different trips, there may be a large time difference between the arrival times of two adjacent checkpoints. In addition, because there may be changes in the routes between different trips, there may be a large distance between two adjacent checkpoints.

[0183] Therefore, electronic devices can calculate the distance between adjacent checkpoints and the vehicle's travel time in the total path based on the checkpoint locations and arrival times of each checkpoint included in the total path.

[0184] S702, identify adjacent checkpoints where the distance is greater than a first distance threshold and / or the vehicle travel time is greater than a first time threshold.

[0185] If the distance between adjacent checkpoints is greater than a first distance threshold, and / or the vehicle travel time between adjacent checkpoints is greater than a first time threshold, it indicates that the adjacent checkpoints may be checkpoints passed through in different trips. Therefore, after calculating the distance between adjacent checkpoints and the vehicle travel time in the total path, the electronic device can determine the adjacent checkpoints in the total path whose distance is greater than the first distance threshold and / or whose vehicle travel time is greater than the first time threshold.

[0186] S703, the total path is divided between the determined adjacent checkpoints to obtain sub-paths corresponding to different trips.

[0187] Because it is determined that adjacent checkpoints may be checkpoints passed through in different trips, electronic devices can divide the total path between the determined adjacent checkpoints, thereby obtaining sub-paths corresponding to different trips.

[0188] For example, if the total path is: checkpoint A → checkpoint E → checkpoint D → checkpoint H → checkpoint B, the electronic device can calculate the distance between adjacent checkpoints and the vehicle travel time in the total path. If the vehicle travel time between checkpoint D and checkpoint H is greater than the first time threshold, then the total path can be divided between checkpoint D and checkpoint H to obtain sub-path 1: checkpoint A → checkpoint E → checkpoint D and sub-path 2: checkpoint H → checkpoint B.

[0189] In the solution provided in this application embodiment, if the distance between adjacent checkpoints is greater than a first distance threshold, and / or the vehicle travel time between adjacent checkpoints is greater than a first time threshold, it indicates that the adjacent checkpoints may be checkpoints passed through in different trips. Therefore, the electronic device can divide the total path between adjacent checkpoints where the distance is greater than the first distance threshold and / or the vehicle travel time is greater than the first time threshold to obtain more accurate sub-paths corresponding to different trips.

[0190] As one implementation method of this application, such as Figure 8 As shown, the above-mentioned sub-paths for different trips, based on the checkpoint locations and arrival times at checkpoints in the total path, can include:

[0191] S801, based on the checkpoint location and arrival time of the checkpoint in the total path, the total path is divided to obtain initial sub-paths.

[0192] Based on the checkpoint locations and arrival times at checkpoints in the total path, the total path is divided to obtain initial sub-paths. The method for dividing the total path can be found in steps S701-S703 above, and will not be repeated here.

[0193] S802, determine the number of checkpoints included in the obtained initial sub-path.

[0194] S803, the initial sub-path with a number of checkpoints greater than a preset checkpoint number threshold is determined as the sub-path corresponding to different trips segmented from the total path.

[0195] Because when the number of checkpoints included in a path is too small, the path cannot reflect the driving pattern of the vehicle. Therefore, after the electronic equipment divides the total path into initial sub-paths, it can determine the number of checkpoints included in the initial sub-paths, and then filter out the initial sub-paths with a number of checkpoints not greater than a preset checkpoint number threshold. Only the initial sub-paths with a number of checkpoints greater than the preset checkpoint number threshold are identified as the sub-paths corresponding to different trips divided from the total path.

[0196] In the solution provided in this application embodiment, after the electronic device divides the total path to obtain initial sub-paths, it can filter out initial sub-paths with a number of checkpoints not greater than a preset checkpoint number threshold by the number of checkpoints included in the sub-paths. Only initial sub-paths with a number of checkpoints greater than the preset checkpoint number threshold are determined as sub-paths corresponding to different trips divided from the total path. This avoids the problem of large prediction errors when predicting the driving path of the target vehicle in the warning area because the number of checkpoints included in the first path is too small when using an initial sub-path with too few checkpoints as the first path.

[0197] As one implementation method of this application, such as Figure 9 As shown, the above-mentioned determination of the sub-path corresponding to the latest trip as the first path includes:

[0198] S901, determine the adjacent checkpoints in the sub-path corresponding to the latest trip whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold, and designate them as adjacent checkpoints with missing paths.

[0199] Because checkpoints are generally sparsely distributed in space, and there may be cases where a vehicle passes through a checkpoint but the data collection equipment installed at the checkpoint fails to capture the vehicle, the first path generated based on the obtained checkpoint vehicle data may contain missing adjacent checkpoints. Such a first path is not conducive to the subsequent prediction process. Here, missing adjacent checkpoints refers to checkpoints between adjacent checkpoints where there are checkpoints that have not captured the target vehicle's vehicle data.

[0200] Therefore, after obtaining the latest sub-path corresponding to the target vehicle's journey, the electronic device can calculate the distance between adjacent checkpoints and the vehicle's travel time within the latest sub-path based on the checkpoint locations and arrival times of each checkpoint included in the latest sub-path. Furthermore, the electronic device can identify adjacent checkpoints in the latest sub-path whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold, and these are designated as adjacent checkpoints with missing path information.

[0201] S902, based on the checkpoint topology, restore the missing path between adjacent checkpoints to obtain the first path.

[0202] Because the missing adjacent checkpoints in the path essentially mean the absence of checkpoints located between those adjacent checkpoints, and the checkpoint topology is established based on the distance between adjacent checkpoints and the travel time of sample vehicles traveling between adjacent checkpoints, it can reflect the drivable directions between adjacent checkpoints. Therefore, after identifying the missing adjacent checkpoints, the electronic device can determine the checkpoints located between the missing adjacent checkpoints based on the pre-established checkpoint topology, and then reconstruct the missing path between the missing adjacent checkpoints based on the determined checkpoints, thus obtaining the first path.

[0203] In the solution provided in this application embodiment, the electronic device can restore the missing path between adjacent checkpoints based on the checkpoint topology, and then use the sub-path corresponding to the latest travel after restoration as the first path. This improves the accuracy of the first path, thereby improving the accuracy of the subsequent prediction process. Furthermore, because the electronic device can restore the path before making predictions, it can also predict paths when faced with checkpoint vehicle passage data of poor quality.

[0204] As one implementation method of this application, such as Figure 10 As shown, based on the pre-established checkpoint topology, the missing paths between adjacent checkpoints are restored to obtain the first path, which may include:

[0205] S1001, based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, determine the second shortest route between adjacent checkpoints where the path is missing, and obtain the vehicle travel time of the second shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship.

[0206] Step S1001 is similar to step 201, except that it targets different adjacent checkpoints. Therefore, it will not be described again here.

[0207] S1002, determine whether the length of the second shortest route is less than the length threshold, and whether the vehicle travel time of the second shortest route is less than the third time threshold.

[0208] When there are no checkpoints with missing paths and no checkpoints with vehicle data that have not been captured between adjacent checkpoints, that is, when there are no missing paths between adjacent checkpoints, if a second shortest route between adjacent checkpoints can be obtained based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, then the difference between the length of the second shortest route and the distance between adjacent checkpoints with missing paths, as well as the difference between the vehicle travel time of the second shortest route and the vehicle travel time between adjacent checkpoints with missing paths, are generally quite large.

[0209] For example, because checkpoints A and B were relatively far apart when they were first built, when determining the missing adjacent checkpoints based on distance and vehicle travel time, checkpoints A and B would be identified as missing adjacent checkpoints. However, the path from checkpoint A to checkpoint B is actually the complete path between checkpoints A and B, and there is no missing path between checkpoints A and B.

[0210] Based on this, if a second shortest route between adjacent checkpoints with missing paths can be obtained based on the distances between them recorded in the checkpoint topology, then this second shortest route is very likely a path from checkpoint A to other checkpoints, then through multiple other checkpoints, and finally back to checkpoint B. The difference between the length of this second shortest route and the distance between adjacent checkpoints with missing paths, as well as the difference between the vehicle travel time of this second shortest route and the vehicle travel time between adjacent checkpoints with missing paths, is quite large. In this case, it would be completely unreasonable to use the second shortest route to reconstruct the missing paths between adjacent checkpoints with missing paths.

[0211] Correspondingly, when there are indeed checkpoints between adjacent checkpoints where the target vehicle was not captured, i.e. when it is determined that there is a missing path between adjacent checkpoints, the length of the shortest route between adjacent checkpoints determined based on the checkpoint topology should be relatively close to the distance between adjacent checkpoints, and the vehicle travel time of the shortest route should also be relatively close to the vehicle travel time between adjacent checkpoints.

[0212] Therefore, after determining the second shortest route, the electronic device can determine whether the length of the second shortest route is less than the length threshold and whether the vehicle travel time of the second shortest route is less than the third time threshold, so as to determine whether the determined second shortest route is reasonable. The length threshold and the third time threshold can be set according to the actual usage scenario and experience, or they can be obtained by adding the distance increment and time increment preset according to the actual usage scenario and experience to the distance between adjacent checkpoints where the path is missing and the vehicle travel time.

[0213] If the length of the second shortest route is less than the length threshold, and the vehicle travel time of the second shortest route is also less than the third time threshold, then the second shortest route is considered reasonable. At this time, the electronic device can execute step S1003 to restore the missing path between adjacent checkpoints using the determined second shortest route.

[0214] If the length of the second shortest route is not less than the length threshold, or the vehicle travel time of the second shortest route is not less than the third time threshold, then the second shortest route is considered unreasonable. In this case, the electronic device may not restore the missing adjacent checkpoints of the route.

[0215] S1003, the missing path between adjacent checkpoints of the missing path is restored using the second shortest route to obtain the first path.

[0216] If the obtained second shortest route is deemed reasonable, the electronic device can use the second shortest route to restore the missing path between adjacent checkpoints and obtain the first path.

[0217] In one implementation, the electronic device can add other checkpoints in the second shortest route, excluding the first and last checkpoints, to the gaps between adjacent checkpoints in the second shortest route, in the order of the other checkpoints in the second shortest route, so as to restore the missing path between adjacent checkpoints.

[0218] For example, if checkpoints A and B are adjacent checkpoints with missing paths, and the determined second shortest route is checkpoint A → checkpoint E → checkpoint D → checkpoint H → checkpoint B, then if the electronic device determines that this second shortest route is reasonable, it can add the other checkpoints (excluding the first and last checkpoints) in this second shortest route, namely checkpoints E, D, and H, to between checkpoints A and B, according to their order in the second shortest route, i.e., checkpoint E → checkpoint D → checkpoint H, to restore the missing path between checkpoints A and B.

[0219] In another implementation, the electronic device may use the checkpoints included in the second shortest route to directly replace the adjacent checkpoints with missing paths in the order of the checkpoints in the second shortest route, so as to restore the missing path between the adjacent checkpoints with missing paths.

[0220] For example, if checkpoints A and B are adjacent checkpoints with missing paths, and the determined second shortest route is checkpoint A → checkpoint E → checkpoint D → checkpoint H → checkpoint B, then if the electronic device determines that the second shortest route is reasonable, it can replace checkpoints A and B with the checkpoints included in the second shortest route, namely checkpoints A, E, D, H, and B, according to their order in the second shortest route, i.e., checkpoint A → E → D → H → B, to restore the missing path between checkpoints A and B.

[0221] In the solution provided in this application embodiment, the electronic device can reconstruct the missing path between adjacent checkpoints based on the checkpoint topology, and then use the sub-path corresponding to the latest travel after reconstruction as the first path. This improves the accuracy of the first path, thereby improving the accuracy of the subsequent prediction process. Furthermore, because the electronic device can reconstruct the path before prediction, it can still predict the path even when faced with checkpoint vehicle data of poor quality. In addition, before using the second shortest route to reconstruct the missing path between adjacent checkpoints, the electronic device can first determine whether the second shortest route is reasonable based on its length and vehicle travel time. Only if it is reasonable will the reconstruction operation be performed, thereby improving the accuracy of the reconstruction.

[0222] As one implementation method of this application, such as Figure 11 As shown, after obtaining the predicted path of the target vehicle within the warning area, the method may further include:

[0223] S1101, among the internal driving paths in the target vehicle's historical trips, determine the internal driving path in the historical trips that is closest to the time of the latest trip, and use it as the driving path to be compared.

[0224] Since the possible route with the largest number of paths is the route most frequently traveled by the target vehicle in historical trips, and the longest common route reflected by the clustered route corresponding to the possible route is also the most frequently traveled route when the target vehicle travels according to the possible route. Therefore, using the longest common route reflected by the clustered route corresponding to the possible route with the largest number of paths as the predicted route can well reflect the most frequently traveled route of the target vehicle after entering the early warning area in history.

[0225] The internal travel route in the historical trip that is closest to the latest travel time, that is, the travel path to be compared, reflects the latest travel route of the target vehicle after entering the early warning area in history. When the matching degree between the predicted route and the travel path to be compared is high, it indicates that the predicted route is both the most frequently traveled route of the target vehicle after entering the early warning area in history and the latest travel route of the target vehicle after entering the early warning area in history, so the accuracy of the predicted route is relatively high.

[0226] Therefore, the electronic device can determine, among the internal travel routes in the historical trips of the target vehicle, the internal travel route in the historical trip that is closest to the latest travel time as the travel path to be compared, and calculate the similarity between the travel path to be compared and the predicted path.

[0227] S1102, determining the confidence corresponding to the predicted path based on the similarity between the travel path to be compared and the predicted path.

[0228] After calculating the similarity between the travel path to be compared and the predicted path, the electronic device can determine the confidence corresponding to the predicted path based on the similarity between the travel path to be compared and the predicted path. For example, the electronic device may directly use the similarity as the confidence corresponding to the predicted path, or may determine the confidence corresponding to the predicted path according to the similarity and a preset corresponding relationship between similarity and confidence.

[0229] For example, a similarity threshold M may be preset in the electronic device, where M>0. If the trajectory similarity is greater than or equal to M, Y is assigned a high confidence; if 0<trajectory similarity<M, Y is assigned a medium confidence; if the trajectory similarity is 0, Y is assigned a low confidence.

[0230] In the solution provided by the embodiments of the present application, after obtaining the predicted path of the target vehicle traveling in the early warning area, the electronic device can also calculate the similarity between the internal travel route in the historical trip closest to the latest travel time and the predicted path, and then determine the confidence according to the similarity, so as to output the similarity between the predicted path and the path of the target vehicle after the target vehicle entered the early warning area last time.

[0231] As one implementation of this application, the above-mentioned obtaining the external driving path of the target vehicle outside the warning area in its latest travel, as a first path, may include:

[0232] Based on the fused checkpoint information, checkpoint fusion is performed on the external driving path of the target vehicle outside the warning area during its latest trip to obtain the first path.

[0233] Multiple data collection devices may be installed at the same checkpoint, and these multiple devices may have different checkpoint IDs. When a vehicle passes through the checkpoint, it may be captured by different devices, resulting in multiple checkpoint vehicle passage data being obtained at one checkpoint. However, these multiple checkpoint vehicle passage data actually represent the same data. Therefore, after generating a driving route based on the checkpoint vehicle passage data, checkpoint fusion can be performed on the checkpoints in the driving route to reduce the number of checkpoints included in the driving route while ensuring the integrity of the driving route.

[0234] Because the format of checkpoint information can be (checkpoint ID, checkpoint name, checkpoint location), and the locations and names of various data acquisition devices installed at the same checkpoint are relatively close, electronic devices can pre-cluster each checkpoint ID according to the checkpoint location and / or checkpoint name corresponding to the checkpoint ID. That is, determine the checkpoint ID of the data acquisition devices installed at the same checkpoint, then map multiple checkpoint IDs clustered into the same category to a new checkpoint ID, and assign the checkpoint location corresponding to the checkpoint ID that serves as the cluster center to this new checkpoint ID, thereby completing checkpoint fusion and obtaining fused checkpoint information. When clustering each checkpoint ID according to the checkpoint location and / or checkpoint name corresponding to the checkpoint ID, clustering algorithms such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise) can be used.

[0235] For example, after clustering the checkpoint IDs according to their corresponding checkpoint locations, checkpoint IDs 0001, 0005, and 0006 are clustered into the same class, with checkpoint ID 0001 serving as the cluster center. The electronic device can then map these multiple checkpoint IDs to a new checkpoint ID 1000 and assign the checkpoint location corresponding to checkpoint ID 0001 to checkpoint ID 1000.

[0236] In one implementation, the electronic device can first cluster each checkpoint ID according to the checkpoint location, and then modify the clustering results by using the similarity of checkpoint names, thereby obtaining a more accurate clustering result.

[0237] Furthermore, after obtaining the target vehicle's latest travel route outside the warning area, the electronic device can determine whether the checkpoint ID in the external travel route belongs to the checkpoint ID to be merged. If so, it can update the checkpoint ID and checkpoint position based on the checkpoint information of the merged checkpoint. Then, it can determine whether there are adjacent checkpoint IDs or checkpoint positions that are the same in the updated external travel route. If so, it can retain only one of the adjacent checkpoints, thereby completing the checkpoint fusion in the external travel route.

[0238] For example, if the IDs of the checkpoints included in the obtained external driving path are 0001, 0005, 0008, 0009, and 0010 respectively, then since checkpoint IDs 0001 and 0005 belong to the checkpoint IDs being merged, the checkpoint IDs of these checkpoints can be updated based on the checkpoint information of the merged checkpoints, resulting in an updated external driving path of 1000, 1000, 0008, 0009, and 0010. Furthermore, since there are adjacent checkpoints with the same checkpoint ID in the updated external driving path, the path can include only one checkpoint, thus obtaining a merged external driving path of 1000, 0008, 0009, and 0010.

[0239] After the electronic device merges the checkpoints in the external driving path, it can use the merged external driving path as the first path.

[0240] In the solution provided in this application embodiment, after the electronic device generates a driving path based on the vehicle passage data at the checkpoint, it can perform checkpoint fusion on the checkpoints in the driving path, thereby reducing the number of checkpoints included in the driving path while ensuring the integrity of the driving path.

[0241] As one implementation method of this application, such as Figure 12 As shown, the method for establishing the checkpoint topology includes:

[0242] S1201, Obtain the historical driving path of the sample vehicles that entered the warning area.

[0243] Electronic devices can acquire the historical driving paths of sample vehicles entering the warning area, and perform data mining based on these historical driving paths to establish checkpoint topology relationships. These historical driving paths include: the external driving paths of sample vehicles outside the warning area and the internal driving paths of target vehicles within the warning area.

[0244] In one approach, after acquiring the historical travel paths of sample vehicles entering the warning area, the electronic device can first perform checkpoint fusion on these historical travel paths, then segment them into sub-paths corresponding to each trip, and finally perform subsequent steps based on the sub-path corresponding to the latest trip. The specific methods of checkpoint fusion and segmentation to obtain the sub-paths corresponding to each trip have been described in detail in the preceding embodiments and will not be repeated here.

[0245] S1202, count the total number of times sample checkpoint pairs formed by each adjacent checkpoint in the historical driving path, the vehicle travel time between checkpoints in the sample checkpoint pairs, and the distance between checkpoints in the sample checkpoint pairs.

[0246] After acquiring the historical driving path of a sample vehicle entering the warning area, the electronic device can split the historical driving path into adjacent checkpoint pairs. The format of the adjacent checkpoint pairs is (upstream checkpoint, downstream checkpoint). The electronic device can then use the identical adjacent checkpoint pairs obtained from the splitting as a sample checkpoint pair. For each sample checkpoint pair, the electronic device counts the number of adjacent checkpoint pairs corresponding to the sample checkpoint pair split from the historical driving path, which is taken as the total occurrence of the sample checkpoint pair. The distance between the checkpoints in the sample checkpoint pair is determined as the distance between the checkpoints in the sample checkpoint pair. The average value of the arrival time difference between the checkpoints in the sample checkpoint pair is determined as the vehicle travel time between the checkpoints in the sample checkpoint pair.

[0247] For example, suppose historical driving route 1 includes checkpoints A → E → D, historical driving route 2 includes checkpoints A → E → G, and historical driving route 3 includes checkpoints B → A → E → D. The electronic device can use the adjacent checkpoint pairs (checkpoint A, checkpoint E) obtained from splitting the historical driving routes as sample checkpoint pairs.

[0248] Based on the various checkpoints included in the historical driving route, the electronic device can count the total number of times the sample checkpoint pair appears (3 times). The electronic device can also determine the distance between checkpoints A and B based on their positions, which is used as the distance between the checkpoints in the sample checkpoint pair. Furthermore, the electronic device can determine the arrival time difference between checkpoints A and B in historical driving route 1 based on the arrival time of checkpoint A and checkpoint B, as the first difference. Based on the arrival time of checkpoint A and checkpoint B in historical driving route 2, the electronic device can determine the arrival time difference between checkpoints A and B in historical driving route 3, as the second difference. Finally, based on the arrival time of checkpoint A and checkpoint B in historical driving route 3, the electronic device can determine the arrival time difference between checkpoints A and B in historical driving route 3, as the third difference. The average of the first, second, and third differences is then used as the vehicle travel time between the sample checkpoints and the central checkpoints.

[0249] S1203, Select sample checkpoint pairs whose occurrence frequency is greater than a preset threshold, whose vehicle travel time is not greater than a second time threshold, and whose distance is not greater than a second distance threshold.

[0250] If the number of occurrences of a sample checkpoint pair is not greater than a preset threshold, it indicates that vehicles rarely pass through the checkpoints in that sample checkpoint pair. Therefore, establishing a topological relationship based on the checkpoints in that sample checkpoint pair is not very meaningful. Thus, it is possible to choose not to establish a topological relationship between the checkpoints in that sample checkpoint pair.

[0251] If the vehicle travel time of a sample checkpoint pair exceeds the second time threshold or the distance exceeds the second distance threshold, it indicates that there may be missing paths between the checkpoints in the sample checkpoint pair. Establishing a topological relationship based on the checkpoints in the sample checkpoint pair cannot reflect the true topological relationship between the checkpoints in the sample checkpoint pair. Therefore, it is also possible to choose not to establish a topological relationship between the checkpoints in the sample checkpoint pair.

[0252] Therefore, after statistically obtaining the total number of times the sample checkpoint pairs appear, the vehicle travel time between checkpoints in the sample checkpoint pairs, and the distance between checkpoints in the sample checkpoint pairs, the electronic device can select sample checkpoints from each sample checkpoint pair whose number of appearances is greater than a preset threshold, whose vehicle travel time is not greater than a second time threshold, and whose distance is not greater than a second distance threshold to generate checkpoint topology relationships.

[0253] S1204, Based on the selected sample checkpoint pairs and the corresponding occurrence count, vehicle travel time, and distance, the checkpoint topology is generated.

[0254] After selecting a pair of checkpoints that meet the criteria, the electronic device can determine the drivable relationship between the checkpoints in the selected checkpoint pair. Then, based on the frequency of occurrence, vehicle travel time, and distance of the selected checkpoint pair, the drivable relationship between the checkpoints is weighted to generate a checkpoint topology.

[0255] In one implementation, the electronic device can use a set of elements such as {upstream checkpoint, downstream checkpoint, distance, vehicle travel time, number of occurrences} to store the selected sample checkpoint pairs and the corresponding number of occurrences, vehicle travel time, and distance for the selected sample checkpoint pairs, thereby obtaining the checkpoint topology.

[0256] In the solution provided in this application embodiment, the electronic device can acquire the historical driving paths of sample vehicles entering the warning area, perform data mining based on these historical driving paths, and thus establish checkpoint topology relationships without needing to acquire road network data. Furthermore, when establishing checkpoint topology relationships, sample checkpoint pairs are also filtered based on the frequency of occurrence, vehicle travel time, and distance, thereby improving the accuracy of the generated checkpoint topology relationships.

[0257] As one implementation method of this application, such as Figure 13 As shown, the method for establishing the correspondence between the external driving path and the internal driving path can include:

[0258] S1301, Obtain the historical driving path of the sample vehicle that entered the warning area.

[0259] Step S1301 is the same as step S1201 described above, and will not be repeated here. When the checkpoint topology relationship and the correspondence between the external driving path and the internal driving path are established by the same device, the device can execute only one of steps S1301 and S1201. For example, when the checkpoint topology relationship and the correspondence between the external driving path and the internal driving path are established by an electronic device, the electronic device can execute only one of steps S1301 and S1201.

[0260] In one embodiment, after the electronic device completes step S1301 or step S1201 and obtains the historical driving path of the sample vehicle entering the warning area, it can first execute steps S1202-S1203 to establish the checkpoint topology relationship. Then, based on the checkpoint topology relationship, it restores the missing paths between adjacent checkpoints that are missing from the historical driving path. Finally, based on the restored historical driving path, it executes steps S1302-S1303 to establish the correspondence between the external driving path and the internal driving path. The method of restoring the missing paths between adjacent checkpoints based on the checkpoint topology relationship has been described in detail in the foregoing embodiments and will not be repeated here.

[0261] S1302, based on the location of the warning area, the historical driving path is divided into historical external path and historical internal path.

[0262] like Figure 14 As shown, after obtaining the historical driving path of the sample vehicle, if the electronic device determines that there is a pair of adjacent checkpoints in the historical driving path based on the location of the warning area, with one checkpoint located outside the warning area and the other checkpoint located inside the warning area, then the historical driving path can be divided between the adjacent checkpoints to obtain the historical external path and the historical internal path.

[0263] S1303, establish the correspondence between the historical external path and the historical internal path, and establish the correspondence between the correspondence and the sample vehicle.

[0264] After the electronic device obtains the historical driving path divided into historical external path and historical internal path, it can first establish the correspondence between the historical external path and the historical internal path, and then establish the correspondence between the correspondence and the sample vehicle, thereby obtaining the correspondence between the external driving path and the internal driving path.

[0265] In the solution provided in this application embodiment, after the electronic device obtains the historical driving path of the sample vehicle that has entered the warning area, it can segment the historical driving path to obtain the historical external path and the historical internal path. Then, based on the historical external path and the historical internal path obtained after segmentation, the electronic device establishes a correspondence between the external driving path and the internal driving path. In this way, after the electronic device obtains the first path of the target vehicle, it can quickly find the second path and the third path according to the correspondence.

[0266] As one embodiment of this application, after obtaining the checkpoint vehicle passage data of the target vehicle within the latest preset time period, the above method may further include:

[0267] Based on the vehicle passage data at the checkpoint, it is determined whether the target vehicle passed through the target checkpoint outside the warning area within a preset time period before the current moment.

[0268] Because the goal is to predict the target vehicle's path within the warning area, path prediction is only necessary if the target vehicle is likely to enter the pre-defined warning area. Therefore, after acquiring checkpoint data on the target vehicle within the latest preset time period, the electronic device can first determine whether the target vehicle is likely to enter the pre-defined warning area based on this data. If so, it then predicts the target vehicle's path within the warning area based on the acquired checkpoint data. Specifically:

[0269] For the warning area, target checkpoints that can be used to characterize a vehicle's impending entry into the warning area can be predetermined. For example, checkpoints outside the warning area that are close to it can be used as target checkpoints, or checkpoints outside the warning area that are typically passed through to enter the warning area can be used as target checkpoints. After the electronic device acquires the vehicle's checkpoint passage data within the latest preset time period, it can determine whether the target vehicle passed through a target checkpoint outside the warning area within that time period. If yes, it is determined that the target vehicle is likely to enter the warning area, and then a predicted path for the target vehicle within the warning area is determined. If no, it is determined that the target vehicle is unlikely to enter the warning area, and subsequent steps can be omitted to save computational resources.

[0270] In the solution provided in this application embodiment, before determining the predicted path of the target vehicle traveling in the area, the electronic device can determine whether the target vehicle is likely to enter the warning area based on the vehicle passage data of the target vehicle at the checkpoint. Then, when the target vehicle is likely to enter the warning area, the prediction step is executed, thereby saving computing resources.

[0271] The following is based on Figure 15 Taking an example, this application provides a method for predicting vehicle travel paths.

[0272] like Figure 15 As shown, during the training process, the electronic device can first obtain the historical driving paths of each sample vehicle, that is, obtain the historical driving paths of vehicle 1 to vehicle N, and then perform checkpoint fusion and driving path segmentation on the checkpoints in the historical driving paths to obtain the sub-path corresponding to the latest trip of the target vehicle.

[0273] Furthermore, the electronic device can establish checkpoint topology relationships based on historical driving routes, and then perform sub-path reconstruction on the latest travel route corresponding to the target vehicle based on these topology relationships. This involves reconstructing the missing paths between adjacent checkpoints within the sub-path. The electronic device can then identify warning areas for the reconstructed sub-paths, dividing them into historical external paths and historical internal paths, thereby constructing corresponding external and internal path libraries, establishing a correspondence between external and internal driving paths. Simultaneously, the checkpoint topology relationships established in the above steps are also stored, thus completing the training process.

[0274] During the prediction process, the electronic device first determines the target vehicle's external driving path, i.e., the first path. Then, it performs a similarity calculation. After the similarity calculation, the electronic device determines the internal driving path based on the checkpoint trajectory prediction model. That is, it identifies the K external paths with the highest similarity from the external driving paths stored during training, i.e., it searches for the second path that matches the first path in the target vehicle's historical external driving paths. Then, based on the correspondence between the external and internal driving paths, it determines the corresponding K internal driving paths.

[0275] After obtaining K internal driving paths, the electronic device can perform route clustering on these K internal driving paths, that is, perform route clustering on the third path. After the clustering is completed, the electronic device can calculate the longest common subsequence of the routes reflected by the clustered paths corresponding to the target route for the class with the most paths, that is, the target route, and obtain the predicted route.

[0276] Finally, the electronic device can predict the arrival time based on the arrival time prediction model, i.e. the checkpoint topology relationship, to determine the predicted time of the target vehicle's arrival at each checkpoint on the predicted route, obtain the predicted path of the target vehicle in the warning area, and output the confidence level. That is, based on the similarity between the driving path to be compared and the predicted path, the confidence level corresponding to the predicted path is determined, thereby completing the prediction process.

[0277] In the solution provided in this application, the electronic device can predict the driving path of the target vehicle based on the vehicle passage data captured by the devices installed at each checkpoint. Therefore, even if the GPS data of the target vehicle cannot be obtained, the driving path of the target vehicle can still be predicted, thereby improving the success rate of predicting the vehicle's driving path. Furthermore, because the predicted path of the target vehicle within the warning area is based on the target vehicle's historical travel paths, the obtained predicted path can closely match the actual driving situation of the target vehicle. Moreover, checkpoint fusion and path reconstruction can be used to enhance the data quality of checkpoint vehicle passage data, making the prediction method of this application still applicable in scenarios with poor checkpoint vehicle passage data quality.

[0278] Corresponding to the above-described vehicle path prediction method, this application also provides a vehicle path prediction device. The following is a description of the vehicle path prediction device provided in this application.

[0279] like Figure 16 As shown, a vehicle driving path prediction device includes:

[0280] The first path determination module 1610 is used to obtain the external driving path of the target vehicle outside the warning area in its latest trip within the latest preset time period based on the checkpoint vehicle data of the target vehicle that has not entered the warning area, and use it as the first path.

[0281] The third path determination module 1620 is used to find a second path that matches the first path in the external driving paths of the target vehicle's historical trips, and to obtain the internal driving path of the second path in the warning area in the historical trips to which the second path belongs, as the third path;

[0282] The route clustering module 1630 is used to perform route clustering on the third path to obtain the possible routes for the target vehicle to travel within the warning area.

[0283] The predicted route determination module 1640 is used to obtain the predicted route of the target vehicle within the warning area based on the obtained possible routes.

[0284] The predicted path determination module 1650 is used to determine the predicted time of the target vehicle to each checkpoint on the predicted route based on the time when the target vehicle arrives at the last checkpoint in the first path, so as to obtain the predicted path of the target vehicle traveling in the warning area.

[0285] In the solution provided in this application embodiment, since most roads currently have checkpoints, when a target vehicle passes through a checkpoint, the device installed at the checkpoint can acquire the target vehicle's checkpoint passage data. Furthermore, the electronic device can predict the target vehicle's travel path based on the checkpoint passage data captured by the devices installed at each checkpoint. Therefore, even if the target vehicle's GPS data cannot be obtained, the travel path can still be predicted, thereby improving the success rate of predicting the vehicle's travel path. Moreover, because the predicted path of the target vehicle within the warning area is based on the target vehicle's historical travel paths, the obtained predicted path closely matches the target vehicle's actual travel situation.

[0286] As one embodiment of this application, the predicted path determination module 1650 may include:

[0287] The time consumption determination unit is used to determine the first shortest route between checkpoints in each adjacent checkpoint pair based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, and to obtain the vehicle travel time of the first shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship; wherein, the adjacent checkpoint pair includes: a checkpoint pair consisting of the last checkpoint in the first path and the first checkpoint in the predicted route, and a checkpoint pair consisting of adjacent checkpoints in the predicted route;

[0288] The prediction time determination unit is used to determine the predicted time of the target vehicle to each checkpoint on the predicted route based on the time when the target vehicle arrives at the last checkpoint in the first path and the obtained vehicle travel time.

[0289] As one embodiment of this application, the predicted route determination module 1640 may include:

[0290] The path number determination unit is used to determine the number of clustered paths corresponding to each possible route, wherein the clustered path corresponding to each possible route is: the third path of the possible route obtained by route clustering.

[0291] The predicted route determination unit is used to obtain the longest common route reflected by the clustered paths corresponding to the target route, as the predicted route for the target vehicle to travel within the warning area, wherein the target route is the possible route with the largest number of paths.

[0292] As one embodiment of this application, the checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint, and the first path determination module 1610 may include:

[0293] The data acquisition unit is used to obtain the vehicle passage data of the target vehicle within the latest preset time period at the checkpoint;

[0294] The total path determination unit is used to sort the checkpoints that the target vehicle arrives at within the latest preset time period according to the arrival time of the checkpoints included in the checkpoint vehicle data, so as to obtain the total path of the target vehicle.

[0295] The path segmentation unit is used to segment sub-paths corresponding to different trips from the total path based on the checkpoint location and arrival time of the checkpoints in the total path.

[0296] The first judgment unit is used to determine, based on the location of the warning area, whether the sub-path corresponding to the latest trip of the target vehicle is located in the warning area;

[0297] The first path determination unit is used to determine the sub-path corresponding to the latest trip as the first path when the judgment result of the first judgment unit is negative.

[0298] As one embodiment of this application, the checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint, and the first path determination module 1610 may include:

[0299] The data acquisition unit is used to obtain the vehicle passage data of the target vehicle within the latest preset time period at the checkpoint;

[0300] The second judgment unit is used to determine, based on the location of the warning area and the vehicle passage data at the checkpoint, whether the target vehicle has entered the warning area within the latest preset time period.

[0301] The total path determination unit is used to sort the checkpoints that the target vehicle arrives at within the latest preset time period according to the arrival time of the checkpoints included in the checkpoint vehicle data when the judgment result of the second judgment unit is negative, so as to obtain the total path of the target vehicle.

[0302] The path segmentation unit is used to segment sub-paths corresponding to different trips from the total path based on the checkpoint location and arrival time of the checkpoints in the total path.

[0303] The first path determination unit is used to determine the sub-path corresponding to the latest trip as the first path.

[0304] As one embodiment of this application, the path segmentation unit may include:

[0305] The calculation subunit is used to calculate the distance between adjacent checkpoints and the vehicle travel time in the total path based on the checkpoint location and arrival time of the checkpoints in the total path.

[0306] The segmentation checkpoint determination subunit is used to determine adjacent checkpoints where the distance is greater than a first distance threshold and / or the vehicle travel time is greater than a first time threshold;

[0307] The path segmentation unit is used to divide the total path between determined adjacent checkpoints to obtain sub-paths corresponding to different trips.

[0308] As one embodiment of this application, the path segmentation unit may include:

[0309] The path segmentation unit is used to segment the total path based on the checkpoint position and arrival time of the checkpoint in the total path to obtain initial sub-paths;

[0310] The checkpoint number determination subunit is used to determine the number of checkpoints included in the obtained initial sub-path;

[0311] The sub-path determination sub-unit is used to determine the initial sub-paths with a number of checkpoints greater than a preset checkpoint number threshold as sub-paths corresponding to different trips segmented from the total path.

[0312] As one embodiment of this application, the first path determination unit may include:

[0313] The checkpoint determination subunit is used to determine the adjacent checkpoints in the sub-path corresponding to the latest trip whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold, as the adjacent checkpoints with missing paths;

[0314] The restoration subunit is used to restore the missing path between adjacent checkpoints based on the checkpoint topology relationship, so as to obtain the first path.

[0315] As one embodiment of this application, the restoration subunit can be specifically used to determine the second shortest route between adjacent checkpoints with missing paths based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship record, and to obtain the vehicle travel time of the second shortest route based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship record; to determine whether the length of the second shortest route is less than a length threshold, and whether the vehicle travel time of the second shortest route is less than a third time threshold; if so, to restore the missing path between adjacent checkpoints with missing paths using the second shortest route, thereby obtaining the first path.

[0316] As one embodiment of this application, the method for establishing the checkpoint topology may include:

[0317] Obtain the historical driving paths of sample vehicles that entered the warning area;

[0318] The total number of times each adjacent checkpoint pair appears in the historical driving path, the vehicle travel time between checkpoints in the sample checkpoint pair, and the distance between checkpoints in the sample checkpoint pair are statistically analyzed.

[0319] Select sample checkpoint pairs whose frequency of occurrence exceeds a preset threshold, whose vehicle travel time does not exceed a second time threshold, and whose distance does not exceed a second distance threshold.

[0320] The checkpoint topology is generated based on the selected sample checkpoint pairs, the frequency of occurrence of the selected sample checkpoint pairs, the vehicle travel time, and the distance.

[0321] As one embodiment of this application, the apparatus may further include:

[0322] The driving path determination module is used to determine, among the internal driving paths in the historical travel history of the target vehicle, the internal driving path in the historical travel history that is closest to the time of the latest travel, as the driving path to be compared.

[0323] The confidence determination module is used to determine the confidence level of the predicted path based on the similarity between the driving path to be compared and the predicted path.

[0324] This application also provides an electronic device, such as... Figure 17 As shown, it includes:

[0325] Memory 1701 is used to store computer programs;

[0326] The processor 1702, when executing the program stored in the memory 1701, implements the vehicle driving path prediction method described in any of the above embodiments.

[0327] Furthermore, the aforementioned electronic device may also include a communication bus and / or a communication interface, with the processor 1702, the communication interface, and the memory 1701 communicating with each other via the communication bus.

[0328] In the solution provided in this application embodiment, since most roads currently have checkpoints, when a target vehicle passes through a checkpoint, the device installed at the checkpoint can acquire the target vehicle's checkpoint passage data. Furthermore, the electronic device can predict the target vehicle's travel path based on the checkpoint passage data captured by the devices installed at each checkpoint. Therefore, even if the target vehicle's GPS data cannot be obtained, the travel path can still be predicted, thereby improving the success rate of predicting the vehicle's travel path. Moreover, because the predicted path of the target vehicle within the warning area is based on the target vehicle's historical travel paths, the obtained predicted path closely matches the target vehicle's actual travel situation.

[0329] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0330] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0331] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0332] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0333] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the vehicle driving path prediction method described in any of the above embodiments.

[0334] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the vehicle driving path prediction method described in any of the above embodiments.

[0335] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0336] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0337] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, electronic devices, computer-readable storage media, and computer program products are basically similar to the method embodiments, and therefore the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0338] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A method for predicting vehicle travel paths, characterized in that, The method includes: Based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, the external driving path of the target vehicle outside the warning area in its latest trip is obtained as the first path. The external driving path of the target vehicle outside the warning area in its latest trip is: a driving path generated based on the checkpoint vehicle passage data of the target vehicle within the latest preset time period, or the latest sub-path with the latest time information among the sub-paths included in the generated driving path. Each sub-path is obtained by dividing the driving path at adjacent checkpoints where the distance is greater than a first distance threshold and / or the vehicle travel time is greater than a first time threshold. In the external driving paths of the target vehicle's historical trips, find the second path that matches the first path, and obtain the internal driving path within the warning area in the historical trips to which the second path belongs, as the third path; The third path is clustered to obtain the possible routes for the target vehicle to travel within the warning area; Based on the obtained possible routes, the predicted route for the target vehicle to travel within the warning area is obtained; Based on the time when the target vehicle arrives at the last checkpoint in the first path, the predicted time when the target vehicle arrives at each checkpoint on the predicted route is determined, and the predicted path of the target vehicle in the warning area is obtained.

2. The method according to claim 1, characterized in that, The step of determining the predicted arrival time of the target vehicle at each checkpoint on the predicted route based on the time the target vehicle arrives at the last checkpoint in the first path includes: For each pair of adjacent checkpoints, based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, a first shortest route between checkpoints in the pair is determined, and the vehicle travel time of the first shortest route is obtained based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship; wherein, the pair of adjacent checkpoints includes: a checkpoint pair consisting of the last checkpoint in the first path and the first checkpoint in the predicted route, and a checkpoint pair consisting of adjacent checkpoints in the predicted route; Based on the time it takes for the target vehicle to reach the last checkpoint on the first path and the obtained vehicle travel time, the predicted time for the target vehicle to reach each checkpoint on the predicted route is determined.

3. The method according to claim 1, characterized in that, The step of obtaining the predicted route for the target vehicle within the warning area based on the obtained possible routes includes: Determine the number of cluster paths corresponding to each possible route, where each possible route corresponds to the third path obtained by route clustering. The longest common route reflected by the clustering path corresponding to the target route is obtained as the predicted route for the target vehicle to travel within the warning area, wherein the target route is the possible route with the largest number of paths.

4. The method according to claim 1, characterized in that, The checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint; and the external driving path of the target vehicle outside the warning area during its latest trip, obtained based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, as the first path, including: Obtain the vehicle passage data of the target vehicle within the latest preset time period; Based on the arrival times at the checkpoints included in the checkpoint vehicle data, the checkpoints reached by the target vehicle within the latest preset time period are sorted to obtain the total path of the target vehicle; Based on the checkpoint locations and arrival times at checkpoints in the total route, sub-routes corresponding to different trips are segmented from the total route; Based on the location of the warning area, determine whether the sub-path corresponding to the latest trip of the target vehicle is located in the warning area; If not, then the sub-path corresponding to the latest trip will be determined as the first path.

5. The method according to claim 1, characterized in that, The checkpoint vehicle passage data includes: the arrival time of the target vehicle at the checkpoint; and the external driving path of the target vehicle outside the warning area during its latest trip, obtained based on the checkpoint vehicle passage data of the target vehicle that has not entered the warning area within the latest preset time period, as the first path, including: Obtain the vehicle passage data of the target vehicle within the latest preset time period; Based on the location of the warning area and the vehicle passage data at the checkpoint, it is determined whether the target vehicle has entered the warning area within the latest preset time period; If not, then according to the arrival time of the checkpoint included in the checkpoint vehicle data, the checkpoints reached by the target vehicle within the latest preset time period are sorted to obtain the total path of the target vehicle; Based on the checkpoint locations and arrival times at checkpoints in the total route, sub-routes corresponding to different trips are segmented from the total route; The sub-path corresponding to the latest trip is determined as the first path.

6. The method according to claim 4 or 5, characterized in that, The process of segmenting sub-paths corresponding to different trips from the total path based on the checkpoint locations and arrival times at checkpoints in the total path includes: Based on the checkpoint locations and arrival times at checkpoints along the total path, calculate the distance between adjacent checkpoints and the vehicle travel time along the total path. Identify adjacent checkpoints where the distance is greater than a first distance threshold and / or the vehicle travel time is greater than a first time threshold; The total path is divided between the identified adjacent checkpoints to obtain sub-paths corresponding to different trips.

7. The method according to claim 4 or 5, characterized in that, The process of segmenting sub-paths corresponding to different trips from the total path based on the checkpoint locations and arrival times at checkpoints in the total path includes: Based on the checkpoint locations and arrival times at checkpoints in the total path, the total path is divided to obtain initial sub-paths; Determine the number of checkpoints included in the resulting initial sub-path; The initial sub-path with a number of checkpoints greater than a preset checkpoint number threshold is determined as the sub-path corresponding to different trips segmented from the total path.

8. The method according to claim 4 or 5, characterized in that, The step of determining the sub-path corresponding to the latest trip as the first path includes: Identify adjacent checkpoints in the sub-path corresponding to the latest trip whose distance is greater than a first distance threshold and / or whose vehicle travel time is greater than a first time threshold, and designate them as adjacent checkpoints with missing paths; Based on the checkpoint topology, the missing paths between adjacent checkpoints are restored to obtain the first path.

9. The method according to claim 8, characterized in that, The method of reconstructing the missing paths between adjacent checkpoints based on the checkpoint topology to obtain the first path includes: Based on the distance between adjacent checkpoints recorded in the checkpoint topology relationship, the second shortest route between adjacent checkpoints where the path is missing is determined, and the vehicle travel time of the second shortest route is obtained based on the vehicle travel time between adjacent checkpoints recorded in the checkpoint topology relationship. Determine whether the length of the second shortest route is less than a length threshold, and whether the vehicle travel time of the second shortest route is less than a third time threshold; If so, the missing path between adjacent checkpoints is restored using the second shortest route to obtain the first path.

10. The method according to claim 2, characterized in that, The methods for establishing the checkpoint topology include: Obtain the historical driving paths of sample vehicles that entered the warning area; The total number of times each adjacent checkpoint pair appears in the historical driving path, the vehicle travel time between checkpoints in the sample checkpoint pair, and the distance between checkpoints in the sample checkpoint pair are statistically analyzed. Select sample checkpoint pairs whose frequency of occurrence exceeds a preset threshold, whose vehicle travel time does not exceed a second time threshold, and whose distance does not exceed a second distance threshold. The checkpoint topology is generated based on the selected sample checkpoint pairs, the frequency of occurrence of the selected sample checkpoint pairs, the vehicle travel time, and the distance.

11. The method according to any one of claims 1-5, characterized in that, After obtaining the predicted path of the target vehicle within the warning area, the method further includes: Among the internal driving paths in the target vehicle's historical trips, the internal driving path in the historical trips that is closest to the time of the latest trip is determined as the driving path to be compared. Based on the similarity between the driving path to be compared and the predicted path, the confidence level corresponding to the predicted path is determined.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-11.

Images