A vehicle aggregation detection method and device, electronic equipment and medium

Abstract

The application relates to the technical field of vehicle clustering detection, in particular to a vehicle clustering detection method and device, electronic equipment and a medium, which are used to solve the problem that a vehicle clustering detection method cannot accurately predict the clustering situation at any time in related technologies. The method is as follows: based on a plurality of driving section information in driving path information of a target vehicle, the position and driving direction of the target vehicle at a plurality of detection times in the future are determined, and the target vehicle is any one of a plurality of vehicles to be detected; the following operations are performed for each detection time: based on the position and driving direction of the plurality of vehicles at the target detection time, the clustering area of the plurality of vehicles is abnormally detected, and an abnormal clustering area is obtained, alarm information is generated for the abnormal clustering area, and the abnormal clustering area is output, and the target detection time is any one of the plurality of detection times; in this way, the vehicle detection accuracy is improved.

Description

Technical Field

[0001] This application relates to the field of vehicle clustering detection technology, and in particular to a vehicle clustering detection method, device, electronic equipment and medium. Background Technology

[0002] In recent years, driverless ride-hailing services have gradually emerged, but so have problems, such as instances of driverless ride-hailing vehicles breaking down en masse. Currently, there is a lack of solutions for vehicle cluster detection in driverless ride-hailing scenarios. While related technologies often employ predictive models for vehicle cluster detection, these models are typically trained on historical data, which can deviate from real-time data, leading to inaccurate predictions of the current clustering situation. This results in frequent missed and false detections, rendering the methods ineffective. Therefore, a vehicle cluster detection method is urgently needed. Summary of the Invention

[0003] This application provides a vehicle clustering detection method, device, electronic equipment, and medium to improve the detection accuracy of vehicle clustering and thus output alarms in a timely manner for abnormal clustering areas.

[0004] The specific technical solutions provided in this application are as follows: In a first aspect, embodiments of this application provide a vehicle aggregation detection method, including: Obtain the driving path information of multiple vehicles to be detected. The driving path information of the target vehicle includes multiple driving segment information, and the target vehicle is any one of the multiple vehicles. Based on multiple road segment information of the target vehicle, determine the position and direction of travel of the target vehicle at multiple future detection times; For each detection time, the following operations are performed: based on the position and driving direction of the multiple vehicles at the target detection time, anomaly detection is performed on the clustering area of ​​the multiple vehicles to obtain the abnormal clustering area, and alarm information is generated and output for the abnormal clustering area, wherein the target detection time is any one of the multiple detection times.

[0005] The vehicle clustering detection method in this application embodiment obtains the driving path information of multiple vehicles to be detected within the system platform. Based on the driving path information of any one vehicle, including the driving segment information, it accurately locates the position and driving direction of the vehicle at any future time. Then, based on the position and driving direction of multiple vehicles, it can pre-perceive the area where these vehicles may gather in the future and perform anomaly detection on the gathering area of ​​multiple vehicles. This enables the detection of anomalies in the spatiotemporal gathering state of vehicles at any future time, thereby achieving early warning of possible abnormal gathering areas in the future, prompting relevant personnel to intervene in advance.

[0006] In one possible implementation, any road segment information includes the travel time and start and end latitude and longitude information of the road segment traveled by the target vehicle; The step of determining the position and direction of travel of the target vehicle at multiple future detection times based on multiple road segment information of the target vehicle includes: From the multiple road segment information of the target vehicle, determine the road segment information associated with the multiple detection times; For the target detection time, based on the travel time and start and end latitude and longitude information in the travel segment information associated with the target detection time, the position of the target vehicle at the target detection time is obtained by interpolation; Based on the start and end latitude and longitude information in the driving segment information associated with the target detection time, the driving direction of the target vehicle is determined.

[0007] The above method acquires vehicle travel path information, which includes information on each travel segment during the vehicle's journey. Based on the travel time and origin / end latitude and longitude information in the travel segment information, it accurately determines the vehicle's position at any future time through interpolation processing, and accurately determines the vehicle's travel direction based on the origin / end latitude and longitude information, providing data support for subsequent vehicle aggregation detection.

[0008] In one possible implementation, the information of any driving segment also includes the road name of the segment. Then, based on the positions and driving directions of the multiple vehicles at the target detection time, anomaly detection is performed on the clustering area of ​​the multiple vehicles to obtain an abnormal clustering area, including: Cluster analysis is performed on the positions of the multiple vehicles at the target detection time to obtain multiple sets of clustered vehicles; According to the road name and driving direction, any one of the multiple vehicle sets is subdivided to obtain multiple sub-vehicle sets; The region corresponding to the sub-vehicle set whose number of vehicles exceeds the vehicle threshold among the multiple sub-vehicle sets is determined as the abnormal clustering region.

[0009] The above method, through cluster analysis and further subdividing each vehicle set obtained after clustering according to road name and driving direction, can predict the areas where these vehicles may gather in the future, and identify abnormal gathering areas based on vehicle thresholds. This enables the detection of anomalies in the spatiotemporal gathering state of vehicles at any future time, thereby identifying abnormal gathering areas that may appear in the future.

[0010] In one possible implementation, the step of performing cluster analysis on the positions of the multiple vehicles at the target detection time to obtain multiple clustered vehicle sets includes: Based on the preset search radius and density threshold, and the positions of the multiple vehicles at the target detection time, the multiple vehicles are clustered to obtain the multiple vehicle sets.

[0011] The above method predicts vehicle clustering in the future by using clustering processing. Compared with related technologies that use historical data reinforcement learning to determine the possible areas where vehicles may cluster, it can accurately predict the areas where vehicles may cluster at any future time based on the positions of multiple vehicles to be detected, i.e., the areas corresponding to the vehicle sets.

[0012] In one possible implementation, the process of subdividing any one of the multiple vehicle sets according to road name and direction of travel to obtain multiple sub-vehicle sets includes: Using either the road name or the direction of travel as the first grouping criterion, vehicles with the same grouping criterion within any vehicle set are grouped into one group, resulting in multiple groups. For any group, vehicles in that group that share the same reference to another group are divided into a subgroup, resulting in multiple subgroups; Each subgroup and the undivided group are respectively identified as a sub-vehicle set.

[0013] The above method, through road grouping and driving direction grouping filtering, compared with related technologies that detect the degree of vehicle clustering in each grid by dividing the grid, can achieve vehicle clustering detection at the road level and in any driving direction on the same road, thereby improving the detection accuracy of vehicle clustering.

[0014] In one possible implementation, if the road names of any two vehicles within any set of vehicles are specific names representing unnamed roads, the method further includes: When it is determined that the distance between any two vehicles is less than a distance threshold, it is determined that the two vehicles are located on the same road, and the two vehicles are assigned to the same target group, wherein the target group is the group or the subgroup. When it is determined that the distance between any two vehicles is not less than the distance threshold, the two vehicles are located on different roads, and the two vehicles are assigned to two target groups.

[0015] The above method proposes a distance threshold for cases where there are unnamed roads in the actual road names. By identifying whether the distance between two vehicles is less than the distance threshold, it can identify whether two unnamed roads are the same road, and then refine and correct the grouping to make the vehicle set more consistent with the actual situation, thereby improving the detection accuracy of subsequent vehicle cluster detection.

[0016] In one possible implementation, defining the region corresponding to the sub-vehicle sets whose number of vehicles exceeds a vehicle threshold as the abnormal clustering region includes: Perform the following operations for each sub-vehicle set: The target sub-vehicle set is obtained by statistically analyzing the position of each vehicle within the target sub-vehicle set. The target sub-vehicle set is the number of vehicles within any preset distance range on the corresponding road. Any one of the multiple sub-vehicle sets; Based on a base threshold matching the road type and the area where the road is located. The region coefficient corresponding to the region type, and the time period type of the time period to which the target detection time belongs. The corresponding time period coefficient is used to determine the vehicle threshold corresponding to the target sub-vehicle set; The region corresponding to the sub-vehicle set whose number of vehicles exceeds the corresponding vehicle threshold is defined as the abnormal clustering region.

[0017] The above method dynamically determines the threshold for each vehicle based on the road type, the area type, and the time period of the detection time. Compared with the fixed threshold in related technologies, the dynamically determined vehicle threshold is more accurate, thus making the abnormal clustering areas identified based on this vehicle threshold more consistent with the actual situation and improving the detection accuracy and precision of vehicle clustering.

[0018] In one possible implementation, obtaining the driving path information of the multiple vehicles to be detected includes: The driving path information of the target vehicle is obtained through a preset travel interface; Wherein, if the target vehicle is a vehicle with an assigned order, then the trip interface is an order trip interface, and the driving route information of the target vehicle is obtained by route planning based on the start and end points of the assigned order; If the target vehicle is a vehicle without assigned orders, then the trip interface is a cruise trip interface, and the driving path information of the target vehicle is obtained by route planning based on the start and end points of the cruise route of the target vehicle.

[0019] The above method retrieves vehicle travel path information through different travel interfaces for vehicles in different driving states, thereby enabling the detection of travel paths for vehicles in different driving states, timely detection of vehicle clustering in different driving states, providing data support for the system platform's scheduling decisions, and avoiding the occurrence of abnormal clustering.

[0020] In one possible implementation, the target vehicle is a driverless ride-hailing vehicle.

[0021] The above method is applicable to the scenario of driverless ride-hailing. Based on the special characteristics of driverless ride-hailing, it can obtain all the paths of vehicle dispatch and scheduling, thereby accurately identifying areas where vehicles may gather in the future. This provides data support for the scheduling decision of the driverless ride-hailing system platform and avoids the occurrence of vehicle gathering and traffic congestion caused by malicious orders.

[0022] Secondly, embodiments of this application provide a vehicle aggregation detection device, comprising: The acquisition unit is used to acquire the driving path information of multiple vehicles to be detected. The driving path of the target vehicle includes information on multiple driving segments, and the target vehicle is any one of the multiple vehicles. The prediction unit is used to determine the position and direction of travel of the target vehicle at multiple future detection times based on multiple road segment information of the target vehicle; An anomaly detection unit is used to perform the following operations for each detection time: based on the position and driving direction of the multiple vehicles at the target detection time, perform anomaly detection on the cluster area of ​​the multiple vehicles to obtain the abnormal cluster area, generate alarm information for the abnormal cluster area and output it, wherein the target detection time is any one of the multiple detection times.

[0023] Thirdly, embodiments of this application provide an electronic device, including: Memory is used to store computer programs or instructions; A processor for executing a computer program or instructions in the memory, such that the method described in any of the first aspects is performed.

[0024] Fourthly, embodiments of this application provide a computer-readable storage medium that, when instructions in the storage medium are executed by a processor, enables the processor to perform the method described in any one of the first aspects above.

[0025] Fifthly, embodiments of this application provide a computer program product, the computer program product comprising: computer program code, which, when run on a computer, causes the computer to perform the method described in any one of the first aspects.

[0026] Furthermore, the technical effects of any of the implementation methods in the second to fifth aspects can be found in the technical effects of different implementation methods in the first aspect, and will not be repeated here.

[0027] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0028] Figure 1 This is a schematic diagram illustrating an application scenario of an optional vehicle aggregation detection method in this application embodiment; Figure 2 This is a schematic diagram of a vehicle aggregation detection method according to an embodiment of this application; Figure 3 This is a schematic diagram of a position and driving direction determination process in an embodiment of this application; Figure 4 This is a schematic diagram of a process for detecting anomalies in a vehicle gathering area and outputting alarms in an embodiment of this application; Figure 5 This is a schematic diagram of a process for subdividing roads according to road names and driving directions in an embodiment of this application. Figure 6 This is a schematic diagram of an unnamed road grouping correction process in an embodiment of this application; Figure 7 This is a schematic diagram illustrating a specific process for determining abnormal clustering areas based on road vehicle density and dynamic vehicle thresholds in an embodiment of this application. Figure 8 This is a schematic diagram illustrating the specific processing logic for determining the number of vehicles using two pointers in an embodiment of this application. Figure 9 This is a schematic diagram of the logical architecture of a vehicle clustering detection device according to an embodiment of this application; Figure 10 This is a schematic diagram of the physical architecture of the electronic device in the embodiments of this application. Detailed Implementation

[0029] 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 a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0030] It should be noted that the terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0031] The data collection, dissemination, and use in this application all comply with relevant national laws and regulations.

[0032] In this embodiment of the application, in order to solve the problem that vehicle clustering detection methods in related technologies cannot accurately predict the clustering situation at any given time, this embodiment of the application provides a vehicle clustering detection method. This method involves pre-acquiring the driving path information of multiple vehicles to be detected, wherein the driving path of the target vehicle includes information on multiple driving segments, and the target vehicle is any one of the aforementioned multiple vehicles. Based on the multiple driving segment information of the target vehicle, the position and driving direction of the target vehicle at multiple future detection times are determined. For each detection time, the following operations are performed: based on the position and driving direction of the multiple vehicles at the target detection time, anomaly detection is performed on the clustering area of ​​the multiple vehicles to obtain an abnormal clustering area, and an alarm message is generated and output for the abnormal clustering area. The target detection time is any one of the multiple detection times.

[0033] The vehicle aggregation detection method described in this application obtains the driving path information of multiple vehicles to be detected. Based on the driving path information of any one vehicle, including the driving segment information, it accurately locates the position and driving direction of the vehicle at any future time. Then, based on the position and driving direction of multiple vehicles, it can detect in advance the area where these vehicles may gather in the future and perform anomaly detection on the gathering area of ​​multiple vehicles. This enables the detection of anomalies in the spatiotemporal aggregation state of vehicles at any future time, thereby providing early warning of possible abnormal aggregation areas in the future and prompting relevant personnel to intervene in advance.

[0034] The preferred embodiments of this application will be further described in detail below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for illustration and explanation of this application and are not intended to limit this application. Furthermore, the embodiments of this application and the features in the embodiments can be combined with each other without conflict.

[0035] See Figure 1 , Figure 1 This is a schematic diagram of an application scenario for an optional vehicle aggregation detection method provided in an embodiment of this application. The application scenario includes multiple driverless ride-hailing vehicles and a vehicle aggregation detection terminal, wherein each driverless ride-hailing vehicle and the vehicle aggregation detection terminal are wirelessly connected.

[0036] In some embodiments, the vehicle aggregation detection device can be an electronic device. For example, the electronic device can be a server, an edge computing node, etc.

[0037] Figure 1 In the illustrated application scenario, the driverless ride-hailing vehicle executes a cruising or order-taking and delivery process according to control instructions issued by a server (not shown) within the system platform. This application scenario also includes a passenger terminal (not shown), such as a mobile phone, iPad, or computer, with ride-hailing software installed. Passengers send order requests to the server through the ride-hailing software, and the order requests include order information such as the origin and destination. After receiving an order request from any passenger, the server selects a driverless ride-hailing vehicle based on the order information in the order request, assigns the passenger's order to the selected driverless ride-hailing vehicle, and sends an order acceptance instruction. The selected driverless ride-hailing vehicle executes the order acceptance and delivery process based on the order acceptance instruction, while driverless ride-hailing vehicles that are not selected or have no orders to fulfill execute the cruising process.

[0038] Of course, the vehicle aggregation detection method provided in the embodiments of this application is not limited to... Figure 1 The application scenarios shown can also be used in other possible application scenarios, such as aggregation detection in food delivery and other application scenarios; or centralized fleet operation scenarios, where the scheduling and routes of all vehicles are uniformly issued by the back-end main control platform and the vehicles cannot plan their own routes, etc. This application does not limit these applications.

[0039] After introducing the application scenarios of the embodiments of this application, the preferred embodiments of this application will be further described in detail below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for illustration and explanation of this application and are not intended to limit this application. Furthermore, the embodiments of this application and the features in the embodiments can be combined with each other without conflict.

[0040] See Figure 2 As shown in the figure, this application provides a vehicle aggregation detection method, the specific process of which is as follows: Step 200: Obtain the driving path information of the multiple vehicles to be detected. The driving path of the target vehicle includes information on multiple driving segments. The target vehicle is any one of the multiple vehicles.

[0041] In this embodiment of the application, when performing step 200, the driving path information of the target vehicle is obtained through a preset travel interface, thereby obtaining the driving path information of multiple vehicles to be detected. The driving path information of the target vehicle includes multiple driving segment information, and any driving segment information includes a segment of the road in which the target vehicle is continuously driving.

[0042] In this embodiment of the application, the information of a driving segment of the target vehicle also includes the name of the road segment in which the target vehicle travels, as well as the driving time and the starting and ending latitude and longitude information of the driving segment.

[0043] by Figure 1 Taking the application scenario shown as an example, the target vehicle is an autonomous ride-hailing vehicle.

[0044] In some possible implementations, if the target vehicle is a vehicle with an assigned order, such as the aforementioned selected driverless ride-hailing vehicle, then the trip interface is an order trip interface, and the driving route information of the target vehicle is obtained by route planning based on the start and end points of the assigned order.

[0045] In some other possible implementations, if the target vehicle is a vehicle without assigned orders, such as the aforementioned driverless ride-hailing vehicle that has not been selected or has no order fulfillment, then the trip interface is a cruising trip interface, and the driving route information of the target vehicle is obtained by route planning based on the start and end points of the target vehicle's cruising route.

[0046] It should be noted that the above-mentioned route planning relies on the route planning (navigation) function of map applications in related technologies. This application does not limit the map application, nor does it limit it to a map application. Any related application or system that can realize the route planning function is acceptable.

[0047] In one possible implementation, the aforementioned starting and ending points are represented using latitude and longitude coordinates, such as latitude and longitude coordinates in JavaScript Object Notation (JSON) format. Then, the data returned by the interface is in JSON format, with each driving segment representing detailed data for each driving step, containing the following core fields: (1) road: road name, which is the core basis for subsequent road grouping and filtering; a specific name, such as an empty string, indicates an unnamed road; (2) duration: indicates the travel time (unit: seconds), used to determine the time range of the vehicle on that road segment; (3) polyline: The latitude and longitude information of each step, used to calculate the driving distance (unit: meters), and combined with the duration field to complete the latitude and longitude interpolation calculation.

[0048] In this embodiment, the field is formatted as “lng1, lat1; lng2, lat2, ...”, where “lng1, lat1” represents the latitude and longitude of the starting point, and “lng2, lat2” represents the latitude and longitude of the second point. In specific implementation, the starting and ending points are extracted as the starting and ending latitude and longitude information of the road segment, and then interpolation is used to calculate the vehicle position at any time.

[0049] It is important to note that the vehicle's travel status may change, such as from "cruising" to "fulfilling a contract," or the navigation map may update the travel status based on road conditions. Each change in status requires updating the vehicle's travel path information to ensure the accuracy of subsequent vehicle aggregation detection.

[0050] Step 210: Based on the target vehicle's multiple road segment information, determine the target vehicle's position and driving direction at multiple future detection times.

[0051] In this embodiment of the application, the above-mentioned multiple detection times are obtained by dividing the detection period based on a preset time interval.

[0052] In practice, multiple detection times over a period of time are considered to determine the position and direction of travel of multiple vehicles at these multiple detection times, so as to accurately predict the vehicle aggregation situation at each detection time.

[0053] In one possible implementation, the aforementioned future time period, such as 30 minutes, is preset to a time interval of 5 seconds. The reason for setting the time interval to 5 seconds is that the average driving speed on typical urban roads is about 36 km / h (10 m / s), and the vehicle travels about 50 meters within 5 seconds, with minimal positional change, allowing for accurate location of the vehicle's road and position. At the same time, the level of road congestion remains relatively stable within 5 seconds. Therefore, choosing 5 seconds as the time interval achieves a balance between positioning accuracy and computational efficiency.

[0054] For example, assuming the start time of the detection period is T0, such as 00:00:00 of the current day, and a detection time is obtained by taking a time point every 5 seconds, then the total number of detection times N within 30 minutes is... total =60 / 5 30; Based on this data, areas where vehicles may gather in the future can be accurately identified, thus providing strong data support for subsequent identification of abnormal gatherings.

[0055] In specific implementation, when performing step 210, refer to... Figure 3 As shown, the specific process is as follows: Step 2101: Determine the driving segment information associated with multiple detection times from the multiple driving segment information of the target vehicle.

[0056] In this embodiment of the application, when performing step 2101, the driving segment information corresponding to the corresponding detection time is selected from multiple driving segment information of the target vehicle according to each detection time, thereby forming a spatiotemporal data set of the target vehicle-each detection time-driving segment information.

[0057] Then, by performing step 2101 for each vehicle, we obtain a spatiotemporal data set of information on each vehicle to be detected, each detection time, and the road segment being traveled.

[0058] Step 2102: For the target detection time, based on the travel time and start and end latitude and longitude information in the travel segment information associated with the target detection time, interpolate to obtain the position of the target vehicle at the target detection time.

[0059] In this embodiment of the application, the travel time in each road segment information is usually based on the time it takes for the vehicle to travel from the starting point to the end point of the road segment. That is, the travel time is not given according to the detection time. Therefore, by treating the vehicle's travel trajectory on each road segment as uniform linear motion, the position of the vehicle at any time can be estimated by linear interpolation calculation.

[0060] In specific implementation, when performing step 2102, for the target detection time (i.e., any one of multiple detection times), based on the travel time in the travel segment information associated with the target detection time, the travel time range of the target vehicle on this road segment is determined; then, based on the travel time range and the start and end latitude and longitude information in the travel segment information associated with the target detection time, the position of the target vehicle at the target detection time is calculated by interpolation. In this way, the position of each vehicle at the target detection time can be obtained.

[0061] For example, suppose the above travel time range is denoted as [t] start , t end ), where t end =t start +duration step , t start It can be calculated based on the current system time and the travel time in the previous travel segments; and the start and end latitude and longitude information in the travel segment information associated with the target detection time is denoted as (Lng). start Lat start ) and (Lng end Lat end Then, the position of the target vehicle (Lng(t), Lat(t)) at the target detection time (denoted as t) can be obtained by linear interpolation using the following formula: Lng(t) = Lng start + (t - t start ) / duration step × (Lng end - Lng start ); Lat(t) = Lat start + (t - t start ) / duration step × (Lat end - Lat start ).

[0062] Step 2103: Determine the driving direction of the target vehicle based on the start and end latitude and longitude information in the driving segment information associated with the target detection time.

[0063] In this embodiment of the application, when performing step 2103, for the target detection time, the travel direction of the target vehicle is obtained by subtracting the start and end latitude and longitude information from the travel segment information associated with the target detection time. For example, it can be divided into 8 directions: north, south, east, west, northeast, southeast, northwest, and southwest. In this way, the travel direction of each vehicle at the target detection time can be obtained.

[0064] Step 220: Perform the following operations for each detection time: Based on the position and driving direction of the above-mentioned multiple vehicles at the target detection time, perform anomaly detection on the cluster area of ​​the multiple vehicles to obtain the abnormal cluster area, generate alarm information for the abnormal cluster area and output it, wherein the target detection time is any one of the above-mentioned multiple detection times.

[0065] In this embodiment of the application, when performing step 220, based on the positions and driving directions of the multiple vehicles at the target detection time, anomaly detection is performed on the clustered area of ​​the multiple vehicles. When an abnormal clustered area is obtained, refer to... Figure 4 As shown, the specific process is as follows: Step 401: Perform cluster analysis on the positions of multiple vehicles at the target detection time to obtain multiple sets of clustered vehicles.

[0066] In this embodiment, the clustering analysis in step 401 is specifically implemented using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. This algorithm does not require a pre-defined number of clusters and can identify clusters of arbitrary shapes. By pre-determining the search radius and density threshold, the algorithm can perform clustering based on these preset parameters, thereby obtaining multiple clustered vehicle sets.

[0067] The following explanation addresses several concepts involved in this algorithm: (1) Search radius (eps), denoted as ε, such as 100 meters; of course, it can be adjusted according to the actual operation situation, such as the search radius can be appropriately reduced in the core congested road sections of the city. (2) Density threshold, i.e. the minimum number of vehicles in a cluster, denoted as MinPts, such as 10 vehicles, indicates that 10 or more vehicles within a 100-meter range can be preliminarily determined to be clustered; for congested road sections, it can be appropriately increased, such as 15 vehicles, etc. (3) Core point, in this embodiment of the application, is defined as: if a vehicle is in the ε neighborhood and the number of surrounding vehicles is not less than MinPts, then the vehicle is called the core point; (4) Boundary point, in this embodiment of the application, is defined as: if a vehicle is in the ε neighborhood, the number of surrounding vehicles is less than MinPts, and it is in the ε neighborhood of another core point, then the vehicle is called a boundary point; (5) Abnormal point, which is defined in this application embodiment as: a vehicle that is neither a core point nor a boundary point.

[0068] In specific implementation, when executing step 401, based on the preset search radius and density threshold, and the positions of the aforementioned multiple vehicles at the target detection time, the multiple vehicles are clustered to obtain multiple vehicle sets; the specific execution logic of the above clustering process is as follows: 1. Traverse all the untraversed cars among the above multiple cars, denoted as p; 2. For each iteration to p, count the number of vehicles in the ε neighborhood of p, where the ε neighborhood refers to all vehicles whose distance from p does not exceed ε. 3. If the number of vehicles is greater than the aforementioned density threshold (MinPts), then p is used as the core point to generate a new cluster C, and the vehicles in the ε neighborhood are added to cluster C; 4. Count all other cars in cluster C, denoted as q. If q is a core point, add the cars in its ε neighborhood to cluster C, and continue to repeat this process for these newly added cars until cluster C can no longer be expanded. 5. After traversing all the cars, multiple clusters are formed. The cars in each cluster are recorded as a vehicle set, thus obtaining the above multiple vehicle sets.

[0069] In this embodiment of the application, the distance between any two vehicles can be calculated using Euclidean distance, and its formula is as follows: d(i,j)=sqrt((Lng i -Lng j )² + (Lat i -Lat j )²) Where d(i,j) represents the Euclidean distance between car i and car j; sqrt() represents the square root operation; (Lng i Lat i (Lng) represents the latitude and longitude information of the location of vehicle i; j Lat j ) represents the latitude and longitude information of the location of vehicle j.

[0070] In the embodiments of this application, Figure 1 Taking the application scenarios shown as an example, driverless ride-hailing vehicles mostly undertake same-city fulfillment or cruising activities. The latitude and longitude span within the same city is small, and the deviation of the plane Euclidean distance is negligible, so the calculation efficiency and accuracy are appropriate.

[0071] It should be noted that the DBSCAN algorithm described above is merely an example, and other clustering algorithms can be used instead. For instance, the density-based clustering analysis algorithm (Ordering Points To Identify the Clustering Structure, OPTICS) and the density-based hierarchical DBSCAN algorithm (HDBSCAN) can be used. Similarly, the Euclidean distance algorithm described above is also just an example; Haversine distance, Manhattan distance, and other methods can also be used to calculate distances.

[0072] In this embodiment, the clustering operation will group vehicles located on different roads but in close proximity into the same cluster (vehicle set), for example, two vehicles located on parallel roads or at an intersection. However, the vehicles are actually located on different roads and do not actually form a cluster. In addition, on a two-way road, the clustering operation will also group vehicles traveling in different directions into the same cluster, which is obviously unreasonable, as vehicles traveling in one direction do not affect the road conditions for vehicles traveling in the other direction, and they do not actually form a cluster. Based on this, in this embodiment, after obtaining multiple clustered vehicle sets, step 402 is executed to further subdivide them according to road name and direction of travel. For details, please refer to the description of step 402.

[0073] Step 402: Subdivide any one of the multiple vehicle sets according to the road name and driving direction to obtain multiple sub-vehicle sets.

[0074] In this embodiment of the application, when performing step 402, refer to... Figure 5 As shown, the specific process is as follows: Step 501: Using either the road name or the direction of travel as the first grouping reference, divide vehicles with the same grouping reference within the vehicle set into one group, resulting in multiple groups.

[0075] Step 502: For any group, divide the cars in that group that have the same reference to another group into a subgroup, resulting in multiple subgroups.

[0076] In this embodiment, when performing step 501, either the road name or the direction of travel can be selected as the first grouping reference. Based on this grouping reference, the vehicles in the vehicle set are grouped to obtain multiple groups. Then, step 502 is performed to further subdivide each of the above groups according to another grouping reference. The following explanation uses the road name as the first grouping reference as an example. The processing flow when the first grouping reference is the direction of travel can be performed in the same way.

[0077] In this embodiment, since the travel segment information of each vehicle includes the road name of the travel segment, when executing step 501, the road name of the road where the vehicle is located can be obtained from the travel segment information of each vehicle in the vehicle set, such as XX Road, YY Road, etc. Then, the vehicles in the vehicle set are subdivided according to the road name, that is, vehicles located on roads with the same road name (the same road) are grouped together, thereby subdividing into multiple groups.

[0078] In practical applications, some roads lack names. This may be related to the map version used in the route planning application, or it may simply be that the road genuinely has no name; these are collectively referred to as unnamed roads. For such roads, for example, if they appear as empty strings in the driving segment information, corresponding to a specific name, then during step 501, if the road names of any two vehicles within the vehicle set are the specific names representing unnamed roads, then refer to... Figure 6 As shown, the following process also needs to be performed: Step 601: Determine the distance between the two vehicles.

[0079] In this embodiment of the application, when performing step 601, the distance between the two vehicles is calculated based on the aforementioned Euclidean distance.

[0080] Step 602: Determine whether the distance is less than the distance threshold. If yes, proceed to step 603; otherwise, proceed to step 604.

[0081] In this embodiment of the application, a distance threshold is preset, for example, denoted as Φ. When executing step 602, the distance determined above is compared with the distance threshold Φ to determine whether the distance is less than the distance threshold Φ. If it is less, then step 603 is executed; if it is not less, then step 604 is executed.

[0082] Step 603: Determine that the two vehicles are on the same road and group them into the same group.

[0083] Step 604: Determine that the two vehicles are located on different roads and assign them to two groups.

[0084] Thus, through steps 601 to 604, the distance between vehicles on two unnamed roads can be calculated using latitude and longitude, and then compared with a pre-set distance threshold to determine whether the two vehicles are on the same unnamed road. If they are, they are grouped into the same group; otherwise, they are grouped into two different groups. This avoids group confusion caused by the lack of road names, thereby improving the accuracy of vehicle cluster detection. It should be noted that if the first grouping reference is the direction of travel, then the above groupings are subgroups.

[0085] In this embodiment of the application, after subdividing according to road name, step 502 is executed to further subdivide the subdivided groups, that is, to subdivide vehicles located on the same road based on the vehicle's driving direction. In specific implementation, when executing step 502, for each group obtained in step 501, vehicles with the same driving direction are divided into a subgroup according to the vehicle's driving direction.

[0086] For example, assuming the road name is "XX Road" and it is a two-way road, when performing step 502, vehicles traveling in the same direction are grouped into a subgroup. The resulting group can be denoted as "XX Road - Forward" and "XX Road - Backward". Of course, it is not important whether the northward direction is forward or backward. The focus here is on vehicles traveling in different directions being assigned to different groups.

[0087] Step 503: Determine each obtained subgroup and the undivided group as a sub-vehicle set.

[0088] In this embodiment of the application, when performing step 503, each subgroup obtained after subdivision in step 502 and the subgroup that does not require subdivision in step 502 obtained after subdivision in step 501 are respectively determined as a sub-vehicle set, wherein the subgroup that does not require subdivision in step 502 represents one-way roads, etc.

[0089] Step 403: Determine the region corresponding to the sub-vehicle set whose number of vehicles exceeds the vehicle threshold from among multiple sub-vehicle sets as the abnormal clustering region.

[0090] In this embodiment of the application, when performing step 403, refer to Figure 7 As shown, the specific process is as follows: Step 701: Perform the following operations for each sub-vehicle set: Operation 1: Based on the position of each vehicle in the target sub-vehicle set, count the number of vehicles within any preset distance range on the road corresponding to the target sub-vehicle set, where the target sub-vehicle set is any one of the aforementioned multiple sub-vehicle sets.

[0091] Operation 2: Based on the basic threshold that matches the road type of the road, the regional coefficient corresponding to the regional type of the area where the road is located, and the time period coefficient corresponding to the time period type of the time period to which the target detection time belongs, determine the vehicle threshold corresponding to the target sub-vehicle set.

[0092] In this embodiment, the vehicle threshold for a road is the core basis for the final determination of aggregation. This vehicle threshold refers to the maximum number of vehicles allowed within the system platform at any preset distance range on a given road at a given detection time. If the number of vehicles on the road within the preset distance threshold exceeds this threshold, it is determined that spatiotemporal aggregation has occurred, and the area corresponding to this sub-vehicle set is an abnormal aggregation area.

[0093] In this embodiment, the vehicle threshold is set as the maximum number of vehicles allowed within the system platform at any preset distance range on a certain road at a certain detection time. The preset distance range is proposed because for long roads, such as a road with a length of 3km and a road with a length of 300m, the former has a significantly greater capacity to carry vehicles than the latter. It is reasonable or in line with reality that the number of vehicles in a cluster should be greater for the former than for the latter. Therefore, a preset distance range is set to consider the vehicle density within any preset distance range on the road. Based on the vehicle density, it is determined whether there is abnormal clustering on the road. In other words, if the road is long and there are many vehicles, the determination of abnormal clustering is not based on the number of vehicles on the road, but on whether the number of vehicles within any preset distance range on the road exceeds the vehicle threshold.

[0094] In step 701, operation one involves using two pointers to determine the number of vehicles within any preset distance range on the road corresponding to the target sub-vehicle set when counting the number of vehicles within any preset distance range on the road corresponding to the target sub-vehicle set.

[0095] For example, assuming the preset distance range is 500 meters, and taking the target sub-vehicle set as the m-th sub-vehicle set among multiple sub-vehicle sets, and the target detection time as the k-th detection time among multiple detection times, see the example below. Figure 8 As shown, the specific processing procedure for performing operation one is as follows: Operation 801: Calculate the maximum longitude distance and the maximum latitude distance of vehicles in the m-th sub-vehicle set.

[0096] Among them, longitude distance represents the difference in longitude values ​​between the positions of the two vehicles, and latitude distance represents the difference in latitude values ​​between the positions of the two vehicles.

[0097] Operation 802: If the maximum longitude distance is greater than the maximum latitude distance, sort the positions of the vehicles in the m-th sub-vehicle set according to the longitude values ​​from smallest to largest, and obtain a one-dimensional array pos[].

[0098] If the maximum longitude distance is greater than the maximum latitude distance, it means that the road corresponding to the m-th sub-vehicle set is an east-west road.

[0099] Operation 803: Initialize two pointers and set the maximum number of vehicles within 500 meters: left=0, right=0, maxCount(m,k)=0.

[0100] At this point, both pointers left and right are located at the position of the first car, which is the first element in the array pos[].

[0101] Operation 804: Iterate through the array pos[] starting from 0 for right. For each car added, calculate pos[right] - pos[left] until a new car is added and pos[right] - pos[left] > 500. Then record the total number of cars in the window [left, right] when the car was not added. The total number of cars in the window [left, right] is right - left + 1. Compare this with maxCount(m,k) and take the maximum value, which is then assigned to maxCount(m,k).

[0102] In step 804, only the right side is moved while the left side remains stationary. By executing step 804, the number of vehicles within the first 500 meters can be found. However, it should be noted that if the total length of the road is less than 500 meters, then after including the last car in step 804, if pos[right] - pos[left] ≤ 500, then the total number of vehicles within the window [left, right] is recorded. This number of vehicles is the number of vehicles on the road corresponding to the m-th sub-vehicle set, and the subsequent process ends.

[0103] Operation 805: Repeat the following steps until right has traversed the array pos[]: Shift left to the right, and right continues to traverse the array pos[] until a new car is included and pos[right] - pos[left] > 500. Then record the number of all cars in the window [left, right] when the car was not included, compare it with maxCount(m,k) and take the maximum value, and assign it to maxCount(m,k); determine whether right has traversed the array pos[].

[0104] When operation 805 is executed for the first time, left is moved to the position of the second car. When it is executed repeatedly in subsequent runs, left is moved to the position of the next car.

[0105] In this embodiment of the application, by executing operations 804 and 805, it can be ensured that the number of vehicles recorded each time does not exceed 500 meters, thereby ensuring that the final maxCount(m,k) is the number of vehicles with the highest vehicle density on the road.

[0106] Operation 806: Assign maxCount(m,k) to Count(m,k) to obtain the number of vehicles corresponding to the m-th sub-vehicle set.

[0107] In the second operation of step 701, in this embodiment of the application, multiple first correspondences, multiple second correspondences, and multiple third correspondences are pre-set. The first correspondence represents the relationship between road type and basic threshold, the second correspondence represents the relationship between area type and area coefficient, and the third correspondence represents the relationship between time period type and time period coefficient.

[0108] In a specific embodiment, based on a plurality of preset first correspondences, a basic threshold matching the road type of the road corresponding to the aforementioned target sub-vehicle set is determined; based on a plurality of preset second correspondences, a regional coefficient corresponding to the regional type of the area where the road is located is determined; based on a plurality of preset third correspondences, a time period coefficient corresponding to the time period to which the target detection time belongs is determined; then, based on the basic threshold, regional coefficient, and time period coefficient, i.e., the product of the basic threshold, regional coefficient, and time period coefficient, the vehicle threshold corresponding to the target sub-vehicle set is determined.

[0109] Still assuming that the target sub-vehicle set is the m-th sub-vehicle set among multiple sub-vehicle sets, and that the target detection time is the k-th detection time among multiple detection times, then the formula for the vehicle threshold is as follows: Th(m,k) = Th_base,m × α_time,k × β_area,m; Where Th(m,k) represents the vehicle threshold of the m-th sub-vehicle set at the k-th detection time, Th_base,m is the base threshold for matching the road type of the road corresponding to the m-th sub-vehicle set; α_time,k represents the time period coefficient corresponding to the time period type of the time period to which the k-th detection time belongs; β_area,m represents the area coefficient corresponding to the area type of the area where the road corresponding to the m-th sub-vehicle set is located.

[0110] In this embodiment of the application, the values ​​and explanations of the parameters for determining the vehicle threshold are detailed in the table below: Table 1. Summary of Values ​​and Explanations

[0111] As shown in the table above, the vehicle threshold is dynamically changing, varying with road type, area type, and time period. For example, assuming a main road (Th_base = 8 vehicles) is located near a school (β_area = 0.7), during the morning rush hour (α_time = 0.6), the dynamic vehicle threshold Th = 8 × 0.6 × 0.7 = 3.36, rounded down to 3 vehicles. This means that if the number of vehicles on this road within 500 meters at the detection time exceeds 3, it is considered a clustering.

[0112] Step 702: The region corresponding to the sub-vehicle set whose number of vehicles is greater than the corresponding vehicle threshold among the above multiple sub-vehicle sets is determined as the aforementioned abnormal clustering region.

[0113] In this embodiment of the application, by executing step 701, the number of vehicles corresponding to each sub-vehicle set and their respective vehicle thresholds are obtained. The number of vehicles corresponding to each sub-vehicle set is compared with the corresponding vehicle threshold. If the number of vehicles corresponding to any sub-vehicle set is greater than its vehicle threshold, then step 702 is executed. If the number of vehicles corresponding to each sub-vehicle set is not greater than its corresponding vehicle threshold, then there is no aggregation, and monitoring can continue.

[0114] In this embodiment of the application, after determining the abnormal clustering area in step 702, the step of generating and outputting alarm information for the abnormal clustering area in step 220 is executed. The output of the alarm information is used to remind relevant personnel that there is an abnormal clustering area at the time of target detection.

[0115] For example, still using the aforementioned preset distance range of 500 meters, and taking the target sub-vehicle set as the m-th sub-vehicle set and the target detection time as the k-th detection time, in the execution... Figure 8 After operation 806, the number of vehicles (Count(m,k)) corresponding to the m-th sub-vehicle set is compared with the vehicle threshold (Th(m,k)). If Count(m,k)>Th(m,k), it is determined that "the road has spatiotemporal aggregation at detection time k". Step 702 is executed to determine the abnormal aggregation area at detection time k for the area corresponding to the m-th sub-vehicle set. Then, the step of generating alarm information for the abnormal aggregation area and outputting it in step 220 is executed.

[0116] In other embodiments of this application, if the number of vehicles in a certain sub-vehicle set exceeds the vehicle threshold at consecutive preset detection times, the abnormal clustering area is determined to be a continuous clustering area, and the alarm level is upgraded to prompt relevant personnel to take timely intervention, such as adjusting vehicle routes or suspending vehicle dispatching in the area.

[0117] Based on the same inventive concept, see [reference] Figure 9As shown in the figure, this application provides a vehicle aggregation detection device, including: The acquisition unit 901 is used to acquire the driving path information of multiple vehicles to be detected. The driving path of the target vehicle includes multiple driving segment information, and the target vehicle is any one of the multiple vehicles. The prediction unit 902 is used to determine the position and driving direction of the target vehicle at multiple future detection times based on multiple road segment information of the target vehicle; The anomaly detection unit 903 is used to perform the following operations for each detection time: based on the position and driving direction of the multiple vehicles at the target detection time, perform anomaly detection on the cluster area of ​​the multiple vehicles to obtain the abnormal cluster area, generate alarm information for the abnormal cluster area and output it, wherein the target detection time is any one of the multiple detection times.

[0118] In one possible implementation, any road segment information includes the travel time and origin / end latitude and longitude information of the road segment traveled by the target vehicle; then the prediction unit 902 is specifically used for: From the multiple road segment information of the target vehicle, determine the road segment information associated with the multiple detection times; For the target detection time, based on the travel time and start and end latitude and longitude information in the travel segment information associated with the target detection time, the position of the target vehicle at the target detection time is obtained by interpolation; Based on the start and end latitude and longitude information in the driving segment information associated with the target detection time, the driving direction of the target vehicle is determined.

[0119] In one possible implementation, the information on any driving segment also includes the road name of the segment, then the anomaly detection unit 903 is specifically used for: Cluster analysis is performed on the positions of the multiple vehicles at the target detection time to obtain multiple sets of clustered vehicles; According to the road name and driving direction, any one of the multiple vehicle sets is subdivided to obtain multiple sub-vehicle sets; The region corresponding to the sub-vehicle set whose number of vehicles exceeds the vehicle threshold among the multiple sub-vehicle sets is determined as the abnormal clustering region.

[0120] In one possible implementation, the anomaly detection unit 903 is specifically used for: Based on the preset search radius and density threshold, and the positions of the multiple vehicles at the target detection time, the multiple vehicles are clustered to obtain the multiple vehicle sets.

[0121] In one possible implementation, the anomaly detection unit 903 is specifically used for: Using either the road name or the direction of travel as the first grouping criterion, vehicles with the same grouping criterion within any vehicle set are grouped into one group, resulting in multiple groups. For any group, vehicles in that group that share the same reference to another group are divided into a subgroup, resulting in multiple subgroups; Each subgroup and the undivided group are respectively identified as a sub-vehicle set.

[0122] In one possible implementation, if the road names of any two vehicles within any set of vehicles are specific names representing unnamed roads, the anomaly detection unit 903 is further configured to: When it is determined that the distance between any two vehicles is less than a distance threshold, it is determined that the two vehicles are located on the same road, and the two vehicles are assigned to the same target group, wherein the target group is the group or the subgroup. When it is determined that the distance between any two vehicles is not less than the distance threshold, the two vehicles are located on different roads, and the two vehicles are assigned to two target groups.

[0123] In one possible implementation, the anomaly detection unit 903 is specifically used for: Perform the following operations for each sub-vehicle set: The target sub-vehicle set is obtained by statistically analyzing the position of each vehicle within the target sub-vehicle set. The target sub-vehicle set is the number of vehicles within any preset distance range on the corresponding road. Any one of the multiple sub-vehicle sets; Based on a base threshold matching the road type and the area where the road is located. The region coefficient corresponding to the region type, and the time period type of the time period to which the target detection time belongs. The corresponding time period coefficient is used to determine the vehicle threshold corresponding to the target sub-vehicle set; The region corresponding to the sub-vehicle set whose number of vehicles exceeds the corresponding vehicle threshold is defined as the abnormal clustering region.

[0124] In one possible implementation, the acquisition unit 901 is specifically used for: The driving path information of the target vehicle is obtained through a preset travel interface; Wherein, if the target vehicle is a vehicle with an assigned order, then the trip interface is an order trip interface, and the driving route information of the target vehicle is obtained by route planning based on the start and end points of the assigned order; If the target vehicle is a vehicle without assigned orders, then the trip interface is a cruise trip interface, and the driving path information of the target vehicle is obtained by route planning based on the start and end points of the cruise route of the target vehicle.

[0125] In one possible implementation, the target vehicle is a driverless ride-hailing vehicle.

[0126] Based on the same inventive concept, this application provides an electronic device, which is described below with reference to... Figure 10 This application describes an electronic device according to such an embodiment. Figure 10 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments described in this application. Figure 10 As shown, electronic devices may include, but are not limited to: Memory 1001 is used to store computer programs or instructions; Processor 1002 is configured to execute computer programs or instructions in memory 1001 such that any of the methods described in the above embodiments is performed.

[0127] In some embodiments, memory 1001 may include a readable medium in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read-only memory (ROM).

[0128] In some embodiments, processor 1002 may include one or more central processing units (CPUs) or digital processing units, etc., for executing computer programs or instructions in memory 1001, such that any of the methods described in the above embodiments is executed.

[0129] It should be noted that the specific connection medium between the memory 1001 and the processor 1002 is not limited in the embodiments of this application. The embodiments of this application... Figure 10 In this diagram, the memory 1001 and processor 1002 are connected via bus 1003. The connections between other components are merely illustrative and not intended to be limiting. Bus 1003 can be categorized as an address bus, data bus, control bus, etc. For ease of illustration, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0130] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium that, when executed by a processor, enables the processor to perform the method described in any of the preceding claims. Since the principle by which the computer-readable storage medium solves the problem is similar to that of the vehicle aggregation detection method, the implementation of the computer-readable storage medium can be found in the implementation of the method, and repeated details will not be elaborated further.

[0131] Based on the same inventive concept, this application also provides a computer program product, which includes computer program code that, when run on a computer, causes the computer to execute any of the methods described above. Since the principle by which the above computer program product solves the problem is similar to that of the vehicle aggregation detection method, the implementation of the above computer program product can be referred to the implementation of the method, and repeated details will not be elaborated further.

[0132] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0134] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flows in a flowchart and / or one or more blocks in a block diagram.

[0135] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more blocks in the block diagram.

[0136] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for detecting vehicle aggregation, characterized in that, include: Obtain the driving path information of multiple vehicles to be detected. The driving path information of the target vehicle includes multiple driving segment information, and the target vehicle is any one of the multiple vehicles. Based on multiple road segment information of the target vehicle, determine the position and direction of travel of the target vehicle at multiple future detection times; For each detection time, the following operations are performed: based on the position and driving direction of the multiple vehicles at the target detection time, anomaly detection is performed on the clustering area of ​​the multiple vehicles to obtain the abnormal clustering area, and alarm information is generated and output for the abnormal clustering area, wherein the target detection time is any one of the multiple detection times.

2. The method as described in claim 1, characterized in that, Information on any travel segment includes the travel time and origin and destination latitude and longitude information of the segment traveled by the target vehicle; The step of determining the position and direction of travel of the target vehicle at multiple future detection times based on multiple road segment information of the target vehicle includes: From the multiple road segment information of the target vehicle, determine the road segment information associated with the multiple detection times; For the target detection time, based on the travel time and start and end latitude and longitude information in the travel segment information associated with the target detection time, the position of the target vehicle at the target detection time is obtained by interpolation; Based on the start and end latitude and longitude information in the driving segment information associated with the target detection time, the driving direction of the target vehicle is determined.

3. The method as described in claim 2, characterized in that, The information for any driving segment also includes the road name of the segment. Based on the positions and driving directions of the multiple vehicles at the target detection time, anomaly detection is performed on the clustering area of ​​the multiple vehicles to obtain an abnormal clustering area, including: Cluster analysis is performed on the positions of the multiple vehicles at the target detection time to obtain multiple sets of clustered vehicles; According to the road name and driving direction, any one of the multiple vehicle sets is subdivided to obtain multiple sub-vehicle sets; The region corresponding to the sub-vehicle set whose number of vehicles exceeds the vehicle threshold among the multiple sub-vehicle sets is determined as the abnormal clustering region.

4. The method as described in claim 3, characterized in that, The clustering analysis of the positions of the multiple vehicles at the target detection time yields multiple clustered vehicle sets, including: Based on the preset search radius and density threshold, and the positions of the multiple vehicles at the target detection time, the multiple vehicles are clustered to obtain the multiple vehicle sets.

5. The method as described in claim 3, characterized in that, The process involves further subdividing any one of the multiple vehicle sets according to road name and direction of travel to obtain multiple sub-vehicle sets, including: Using either the road name or the direction of travel as the first grouping criterion, vehicles with the same grouping criterion within any vehicle set are grouped into one group, resulting in multiple groups. For any group, vehicles in that group that share the same reference to another group are divided into a subgroup, resulting in multiple subgroups; Each subgroup and the undivided group are respectively identified as a sub-vehicle set.

6. The method as described in claim 5, characterized in that, If the road names of any two vehicles within any set of vehicles are specific names representing unnamed roads, the method further includes: When it is determined that the distance between any two vehicles is less than a distance threshold, it is determined that the two vehicles are located on the same road, and the two vehicles are assigned to the same target group, wherein the target group is the group or the subgroup. When it is determined that the distance between any two vehicles is not less than the distance threshold, the two vehicles are located on different roads, and the two vehicles are assigned to two target groups.

7. The method as described in claim 6, characterized in that, The step of defining the region corresponding to the sub-vehicle sets in which the number of vehicles exceeds a vehicle threshold as the abnormal clustering region includes: Perform the following operations for each sub-vehicle set: The target sub-vehicle set is obtained by statistically analyzing the position of each vehicle within the target sub-vehicle set. The target sub-vehicle set is the number of vehicles within any preset distance range on the corresponding road. Any one of the multiple sub-vehicle sets; Based on a base threshold matching the road type and the area where the road is located. The region coefficient corresponding to the region type, and the time period type of the time period to which the target detection time belongs. The corresponding time period coefficient is used to determine the vehicle threshold corresponding to the target sub-vehicle set; The region corresponding to the sub-vehicle set whose number of vehicles exceeds the corresponding vehicle threshold is defined as the abnormal clustering region.

8. The method according to any one of claims 1-7, characterized in that, The process of obtaining the driving path information of the multiple vehicles to be detected includes: The driving path information of the target vehicle is obtained through a preset travel interface; Wherein, if the target vehicle is a vehicle with an assigned order, then the trip interface is an order trip interface, and the driving route information of the target vehicle is obtained by route planning based on the start and end points of the assigned order; If the target vehicle is a vehicle without assigned orders, then the trip interface is a cruise trip interface, and the driving path information of the target vehicle is obtained by route planning based on the start and end points of the cruise route of the target vehicle.

9. A vehicle aggregation detection device, characterized in that, include: The acquisition unit is used to acquire the driving path information of multiple vehicles to be detected. The driving path of the target vehicle includes information on multiple driving segments, and the target vehicle is any one of the multiple vehicles. The prediction unit is used to determine the position and direction of travel of the target vehicle at multiple future detection times based on multiple road segment information of the target vehicle; An anomaly detection unit is used to perform the following operations for each detection time: based on the position and driving direction of the multiple vehicles at the target detection time, perform anomaly detection on the cluster area of ​​the multiple vehicles to obtain the abnormal cluster area, generate alarm information for the abnormal cluster area and output it, wherein the target detection time is any one of the multiple detection times.

10. An electronic device, characterized in that, include: Memory is used to store computer programs or instructions; A processor for executing a computer program or instructions in the memory such that the method described in any one of claims 1-8 is performed.

11. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor, the processor is able to perform the method as described in any one of claims 1-8.

12. A computer program product, characterized in that, The computer program product includes: computer program code, which, when run on a computer, causes the computer to perform the method described in any one of claims 1-8.

Images