Method, device and computer equipment for identifying fleet based on longitude and latitude

Abstract

The present application relates to a fleet identification method based on longitude and latitude, comprising: obtaining trajectory parameters of several vehicles; the trajectory parameters include several groups of position data at adjacent moments, the position data are arranged in chronological order of occurrence, and the position data include longitude and latitude; generating the trajectory feature coordinates of the vehicle according to the trajectory parameters; generating clusters of the vehicle using a density clustering algorithm according to the trajectory feature coordinates; obtaining a vehicle combination based on the cluster cluster, calculating the KL divergence of the position data of the vehicle combination, and if the KL divergence of the position data of the vehicle combination is less than a preset KL divergence threshold, determining that the vehicles in the vehicle combination belong to the same fleet. In the fleet identification method based on longitude and latitude described in the above embodiment, a large amount of vehicle trajectory data is incorporated into the calculation program through an algorithm, data processing is convenient and fast, and double clustering is performed by density clustering and comparison of KL divergence, with higher clustering accuracy.

Description

Technical Field

[0001] The present application relates to the field of data mining technology, and in particular to a method, device and computer equipment for identifying a fleet based on longitude and latitude. Background Art

[0002] The vehicle type, cargo type, power system, transmission system and driving route of the same fleet are basically consistent. By identifying the vehicle information of the same fleet, manufacturers can obtain more customer information, which is convenient for serving and maintaining customers. On the other hand, the results of fleet identification can be used as input for other algorithms, such as analyzing the different performances of the same type of vehicles on the same driving route.

[0003] Traditional fleet identification mostly relies on automobile manufacturers' mastery of major customer information, and obtains information about vehicles belonging to the same fleet by consulting relevant information. This method of obtaining fleet information is inefficient and has a small amount of data, and cannot meet the needs of increasingly large and complex fleet systems. Summary of the invention

[0004] Based on this, it is necessary to provide a fleet identification method, device and computer equipment based on longitude and latitude to address the above technical problems, so as to solve the problem of low efficiency in obtaining fleet information.

[0005] The first aspect of the present application provides a method for identifying a fleet based on longitude and latitude, comprising:

[0006] Acquire trajectory parameters of several vehicles; the trajectory parameters include several groups of position data at adjacent moments, the position data are arranged in chronological order of occurrence, and the position data include longitude and latitude;

[0007] Generating trajectory characteristic coordinates of the vehicle according to the trajectory parameters;

[0008] Generating a cluster of the vehicle using a density clustering algorithm according to the trajectory feature coordinates;

[0009] A vehicle combination is obtained based on the clustering cluster, and the KL divergence of the position data of the vehicle combination is calculated. If the KL divergence of the position data of the vehicle combination is less than a preset KL divergence threshold, it is determined that the vehicles in the vehicle combination belong to the same fleet; the vehicle combination includes any two different vehicles in the clustering cluster.

[0010] In the longitude-based fleet identification method described in the above embodiment, a large number of vehicle driving trajectory data, including longitude and latitude information at each adjacent time, are collected to obtain the longitude-based trajectory feature coordinates of different vehicles, and then a density clustering algorithm is used to cluster the trajectory feature coordinates of all vehicles, and vehicles whose distance from the cluster center is within a preset threshold range are selected to form a cluster cluster to achieve preliminary fleet screening. Then, on this basis, any two vehicles in the cluster cluster are combined into a vehicle combination, and the KL divergence of the longitude and latitude between the two vehicles in each vehicle combination is calculated one by one. When the calculation result is lower than the preset KL divergence threshold, it is determined that the two vehicles belong to the same fleet, until all vehicle combinations are compared, and finally the vehicles with the closest trajectory features are obtained to form the same fleet. The above method incorporates a large amount of vehicle trajectory data into the calculation program through the algorithm, and data processing is convenient and fast. Double clustering is performed by density clustering and comparison of KL divergence, and the clustering accuracy is higher.

[0011] In one embodiment, the trajectory parameters include N groups of longitudes and latitudes at adjacent moments, where N is a positive integer; the step of generating the trajectory characteristic coordinates of the plurality of vehicles according to the trajectory parameters includes:

[0012] Calculate the first-order derivative of the trajectory parameters of any vehicle, the first-order derivative μ i Calculated according to the following formula:

[0013]

[0014] Where i∈[2,N], J i is the longitude at the i-th moment, J i-1 is the longitude at the i-1th moment, W i is the latitude at the i-th moment, W i-1 is the latitude at the i-1th moment.

[0015] In the longitude-latitude based fleet identification method described in the above embodiment, on the one hand, preliminary positioning is performed by using discrete longitude-latitude coordinates arranged in chronological order, and on the other hand, by introducing the first-order derivative of longitude with respect to latitude, the trajectory between discrete time points is made closer to the real trajectory, which is beneficial to improving the accuracy of the trajectory.

[0016] In one of the embodiments, if the latitude of the trajectory parameter at the i-th moment is the same as the latitude at the i-1-th moment, the value range of the first-order derivative of the trajectory parameter at the i-th moment is [-0.1, 0.1].

[0017] In one embodiment, the trajectory characteristic coordinates include at least one of the characteristic value of the first-order derivative of the trajectory parameter, the characteristic value of longitude, and the characteristic value of latitude; the characteristic value includes at least one of the mean, sum, variance, quartile, and data quantity.

[0018] In the longitude-latitude based fleet identification method described in the above embodiment, the trajectory of any vehicle is converted into a coordinate point in a five-dimensional coordinate system including at least the first-order derivative of trajectory parameters, the mean, sum, variance, quartile and data quantity of three variables, namely, longitude and latitude. This makes it easier to compare trajectory features between different vehicles and facilitates clustering calculations of trajectories of different vehicles.

[0019] In one of the embodiments, the distance between the trajectory feature coordinates of any vehicle in the cluster and the cluster center coordinates is less than a preset distance value.

[0020] In the longitude-latitude based fleet identification method described in the above embodiment, the trajectory of the vehicle is concretized as coordinate points. For different vehicles, the closer the distance between the coordinate points of the two vehicles is, the more similar their trajectories are, and the possibility that they belong to the same fleet is higher. Therefore, through the density clustering algorithm, vehicles within a preset distance value from the cluster center are classified into the same cluster, which can achieve preliminary screening.

[0021] In one embodiment, it also includes:

[0022] Forming a vehicle combination with any two different vehicles in the cluster;

[0023] Calculating the KL divergence of longitude, the KL divergence of latitude, and the KL divergence of the joint distribution of longitude and latitude between two vehicles in the vehicle combination;

[0024] If the KL divergence of the longitude is less than a preset longitude KL divergence threshold, the KL divergence of the latitude is less than a preset latitude KL divergence threshold, and the KL divergence of the joint distribution of longitude and latitude is less than a preset longitude and latitude KL divergence threshold, it is determined that the vehicles in the vehicle combination belong to the same fleet.

[0025] In the longitude and latitude-based fleet identification method described in the above embodiment, on the basis of the clustering clusters in the above embodiment, the vehicles in the cluster are paired and combined in pairs, and the KL divergence of the longitude, the KL divergence of the latitude and the KL divergence of the joint distribution of the longitude and latitude of each pair of vehicle combinations are calculated respectively, and the above three KL divergence values ​​are compared with the corresponding KL divergence thresholds. When all three are within the corresponding threshold range, it is determined that the vehicles in the vehicle combination belong to the same fleet. This method can further improve the clustering accuracy of the fleet, and the driving trajectories of the finally selected fleet combination have high similarity.

[0026] In one embodiment, it also includes:

[0027] A trajectory scatter plot of vehicles in the same fleet is drawn according to the trajectory parameters, and the trajectory scatter plot and vehicle identification codes of the corresponding vehicles are output.

[0028] The second aspect of the present application provides a fleet identification device based on longitude and latitude, including a data acquisition module, a calculation module, a judgment module and a display module, wherein the data acquisition module is used to obtain trajectory parameters of a plurality of vehicles; the calculation module is used to calculate the trajectory characteristic coordinates of the plurality of vehicles according to the trajectory parameters of the plurality of vehicles, and calculate and obtain the clustering clusters of the vehicles using a density clustering algorithm according to the trajectory characteristic coordinates, obtain a vehicle combination based on the clustering cluster, and calculate the KL divergence of the position data of the vehicle combination; the judgment module is used to judge whether the vehicle belongs to the clustering cluster according to the distance between the trajectory characteristic coordinates of the vehicle and the cluster center coordinates of the clustering cluster, and / or judge whether the vehicles in the vehicle combination belong to the same fleet according to the KL divergence of the position data of the vehicle combination; the display module is used to output and display the trajectory scatter plot of the vehicles in the same fleet and the vehicle identification codes of the corresponding vehicles.

[0029] In the longitude-latitude based vehicle fleet identification device described in the above embodiment, a data acquisition module is set to collect trajectory parameters of different vehicles at different times on a large scale, so that the collected data is more comprehensive. On this basis, the collected data is processed by the calculation module to obtain the trajectory characteristic coordinates of each vehicle, and then the clustering clusters of all vehicles are calculated based on the density clustering algorithm as the preliminary selected vehicle fleet, and then the vehicles in the clusters are paired in pairs, and the KL divergence of the mutual position data is calculated. The judgment module is used to judge whether the paired vehicle combination meets the characteristics of the same vehicle fleet, so as to further refine the vehicle fleet clustering results.

[0030] In one embodiment, it also includes:

[0031] The data processing module is used to perform data normalization processing on the track feature coordinates according to preset rules so that the data of different vehicles corresponding to the same coordinates adopt a unified standard.

[0032] A third aspect of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any of the aforementioned methods for identifying fleets based on longitude and latitude are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0034] Figure 1A schematic diagram of a flow chart of a method for identifying a fleet based on longitude and latitude in an embodiment provided in the present application;

[0035] Figure 2 A schematic diagram of a density-based clustering method with noise in two-dimensional space in one embodiment provided in the present application;

[0036] Figure 3 A schematic diagram of a flow chart of a method for identifying a fleet based on longitude and latitude in another embodiment provided by the present application;

[0037] Figure 4 A schematic diagram of the structure of a vehicle fleet identification device based on longitude and latitude in one embodiment provided in the present application. DETAILED DESCRIPTION

[0038] In order to facilitate understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. Embodiments of the present application are provided in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0040] It can be understood that the terms "first", "second", etc. used in the present application can be used in this article to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0041] It should be noted that when an element is considered to be "connected" to another element, it can be directly connected to the other element, or connected to the other element through an intermediate element. In addition, the "connection" in the following embodiments should be understood as "electrical connection", "communication connection", etc. if there is transmission of electrical signals or data between the connected objects.

[0042] When used herein, the singular forms "a", "an", and "said / the" may also include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" etc. specify the presence of stated features, wholes, steps, operations, components, parts or combinations thereof, but do not exclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts or combinations thereof.

[0043] With the popularization of automobiles, the traffic volume of highway networks is increasing. In order to facilitate vehicle management, vehicles on the same route need to be managed in teams to facilitate customer dispatch and maintenance of vehicles. Therefore, how to identify vehicles on the same route has become an urgent problem to be solved. Traditional fleet identification mostly relies on automobile manufacturers' mastery of key customer information and obtaining information about vehicles belonging to the same fleet by consulting relevant information. This method of obtaining fleet information is inefficient and has a small amount of data, which cannot meet the needs of increasingly large and complex fleet systems.

[0044] Based on this, it is necessary to provide a fleet identification method, device and computer equipment based on longitude and latitude to address the above technical problems, so as to solve the problem of low efficiency of fleet information acquisition methods.

[0045] The following will be described through specific examples.

[0046] In one embodiment of the present application, Figure 1 A method for identifying a fleet based on longitude and latitude is provided, comprising:

[0047] Step 22, obtaining trajectory parameters of several vehicles;

[0048] The trajectory parameters include several groups of position data at adjacent moments, and the position data are arranged in chronological order of occurrence; the position data include longitude and latitude.

[0049] Specifically, the longitude and latitude coordinates of the vehicle arranged in chronological order are obtained through the GPS navigation system or the Beidou navigation system. The vehicle's driving trajectory can be reflected through a series of discrete longitude and latitude coordinate points. The shorter the time interval between adjacent sampling points, the higher the accuracy, and the more realistic the vehicle's driving trajectory can be restored. The driving trajectory data of any vehicle must be complete, without missing items and without obvious abnormalities.

[0050] Step 24, generating the trajectory characteristic coordinates of the vehicle according to the trajectory parameters;

[0051] Step 26, generating clusters of the vehicles using a density clustering algorithm according to the trajectory feature coordinates;

[0052] Specifically, density-based clustering algorithms assume that the clustering structure can be determined by the density of sample distribution, and clustering is based on the density of the spatial distribution of the data set, that is, as long as the sample density in an area is greater than a certain threshold, it will be classified into a cluster similar to it. Among them, the density-based spatial clustering of applications with noise (DBSCAN) is a typical density clustering algorithm, which is based on a set of neighborhood parameters (ε, MinPts) to describe the density of sample distribution. Compared with partition-based clustering methods and hierarchical clustering methods, the DBSCAN algorithm defines a cluster as the maximum set of density-connected samples, can divide areas with high enough density into clusters, does not require a given number of clusters, and can find clusters of arbitrary shapes in noisy spatial data sets. The DBSCAN algorithm first selects a core object in the data set as a seed, creates a cluster and finds all its core objects, and finds objects that can be reached by merging the core objects until all core objects have been visited. A DBSCAN cluster contains at least one core object: if there is only one core object, then all other non-core objects fall within the ε-neighborhood of the core object; if there are multiple core objects, then there is at least one other core object within the ε-neighborhood of any core object, otherwise the two core objects cannot be density-reachable, and a cluster containing too few objects can be considered noise.

[0053] To further illustrate the principle of density clustering algorithm, Figure 2As shown in FIG. 1 , a principle diagram of a density-based clustering method with noise in a two-dimensional space is provided. The figure contains 9 points, among which, in (a), points 1 to 8 are close to each other and have a high local density. Point 5 is selected as the core object. The distances between points 1, 2, 3, 4, 6, 7 and 8 and point 5 are all within the radius of the dotted circle, that is, points 1, 2, 3, 4, 6, 7 and 8 are all within the ε-neighborhood of point 5. Therefore, points 1 to 8 can be delineated as a cluster with point 5 as the cluster center. The distance between point 9 and the cluster center point 5 is greater than the radius of the dotted circle, and point 9 is an isolated point. Point 9 can be identified as noise. Therefore, points 1 to 8 constitute the cluster of the entire data set, with the cluster center being Point 5, and in Figure (b), at this time, the cluster center includes point 5, point 3 and point 7, among which point 1 and point 6 are within the ε-neighborhood range of cluster center point 5, point 4 and point 8 are within the ε-neighborhood range of cluster center point 3, point 2 is within the ε-neighborhood range of cluster center point 7, and at the same time, point 3 is within the ε-neighborhood range of point 5, and point 7 is within the ε-neighborhood range of point 3, so point 5 to point 7 can be connected by density. Therefore, points 1 to 8 form a density-connected cluster, and the cluster center includes point 3, point 5 and point 7. The distance between point 9 and cluster center point 3, cluster center point 5 and cluster center point 7 exceeds the radius of the dotted circle, and there are no other points around to form a cluster, so point 9 can be classified as noise and excluded. It should be noted here that the radius of the cluster is not fixed and can be set according to actual needs. In this embodiment, any vehicle is positioned as a single point in the coordinate system through the trajectory feature coordinates. Therefore, based on the above principle, a clustering cluster of all vehicle sets can be obtained through a density clustering algorithm. The distance between the trajectory feature coordinates of any vehicle in the cluster and the cluster center coordinates is less than the preset distance value. The vehicles in the cluster are preliminarily screened vehicles with similar driving trajectories.

[0054] Step 28, obtaining a vehicle combination based on the clustering cluster, calculating the KL divergence of the position data of the vehicle combination, and if the KL divergence of the position data of the vehicle combination is less than a preset KL divergence threshold, determining that the vehicles in the vehicle combination belong to the same fleet.

[0055] The vehicle combination includes any two different vehicles in the cluster.

[0056] Specifically, information entropy can express the amount of data information. For a discrete random variable x i , the information entropy calculation formula is as follows: Among them, p(x i ) is a discrete random variable x iThe relative entropy is also called the KL divergence (Kullback-Leibler divergence) or information divergence (information divergence), which is an asymmetric measure of the difference between two probability distributions. In information theory, the relative entropy is equivalent to the difference between the information entropies of the two probability distributions. If one of the probability distributions is the true distribution and the other is the fitted distribution, then the relative entropy is equal to the difference between the cross entropy and the information entropy of the true distribution, indicating the information loss when the theoretical distribution is used to fit the true distribution. The relative entropy calculation formula is as follows: Among them, p(x i ) is the probability distribution of real events, q(x i ) is the probability distribution of the event fitted theoretically. The relative entropy is always non-negative. If the relative entropy is zero, the probability distributions of the two are the same, indicating that the real distribution is exactly the same as the event fitted theoretically. If the relative entropy is greater than zero, it means that there is a difference between the two, and the larger the relative entropy, the more obvious the difference. Therefore, the relative entropy can be used to measure the distance between two random distributions. In this embodiment, any two cars are paired in the cluster described in the above embodiment. For the two cars in any vehicle combination, it is assumed that they are vehicle A and vehicle B respectively. The information entropy of the longitude of the driving trajectory of vehicle A is: Among them, p(J Ai ) is the longitude J of the driving trajectory of vehicle A, a discrete random variable Ai The probability distribution function of , similarly, the information entropy of the longitude of vehicle B’s driving trajectory is: Among them, p(J Bi ) is the longitude J of vehicle B’s driving trajectory, a discrete random variable Bi The probability distribution function of vehicle A is obtained from this, and the KL divergence of the longitude of vehicle A’s driving trajectory to the longitude of vehicle B’s driving trajectory is: If D KL (J A ||J B )=0, indicating that the longitude of the driving track of vehicle A is exactly the same as the longitude of the driving track of vehicle B. KL (J A ||J B )>0, indicating that the longitude of the driving track of vehicle A is different from the longitude of the driving track of vehicle B, and D KL (J A ||J B), the larger the value of , the greater the difference in longitude of the driving trajectories of the two. Similarly, the above method can be used to solve the KL divergence of the latitude of vehicle A's driving trajectory with respect to the longitude of vehicle B's driving trajectory and the KL divergence of the joint distribution of longitude and latitude of vehicle A's driving trajectory with respect to the joint distribution of longitude and latitude of vehicle B's driving trajectory. If the calculated values ​​of the above three data items are all within the corresponding preset KL divergence threshold ranges, it means that the driving trajectories of vehicle A and vehicle B are highly similar, and the two can be classified as the same fleet.

[0057] In the longitude-based fleet identification method described in the above embodiment, a large number of vehicle driving trajectory data, including longitude and latitude information at each adjacent time, are collected to obtain the longitude-based trajectory feature coordinates of different vehicles, and then a density clustering algorithm is used to cluster the trajectory feature coordinates of all vehicles, and vehicles whose distance from the cluster center is within a preset threshold range are selected to form a cluster cluster to achieve preliminary fleet screening. Then, on this basis, any two vehicles in the cluster cluster are combined into a vehicle combination, and the KL divergence of the longitude and latitude between the two vehicles in each vehicle combination is calculated one by one. When the calculation result is lower than the preset KL divergence threshold, it is determined that the two vehicles belong to the same fleet, until all vehicle combinations are compared, and finally the vehicles with the closest trajectory features are obtained to form the same fleet. The above method incorporates a large amount of vehicle trajectory data into the calculation program through the algorithm, and data processing is convenient and fast. Double clustering is performed by density clustering and comparison of KL divergence, and the clustering accuracy is higher.

[0058] As an example, the trajectory parameters include longitudes and latitudes of N groups of adjacent moments, where N is a positive integer; step 22, generating trajectory feature coordinates of the plurality of vehicles according to the trajectory parameters, includes:

[0059] Calculate the first-order derivative of the trajectory parameters of any vehicle, the first-order derivative μ i Calculated according to the following formula:

[0060]

[0061] Where i∈[2,N], J i is the longitude at the i-th moment, J i-1 is the longitude at the i-1th moment, W i is the latitude at the i-th moment, W i-1 is the latitude at the i-1th moment.

[0062] Specifically, longitude and latitude positioning is the main method for automobile positioning. However, longitude and latitude data are discrete, while the vehicle's driving trajectory is continuous. Sampling the longitude and latitude data at different times during the vehicle's driving process cannot fully reflect the vehicle's true trajectory. The lack of longitude and latitude data between adjacent sampling points will directly lead to a deviation between the trajectory data and the true trajectory. Therefore, this embodiment introduces a method of the first-order derivative of longitude to latitude between adjacent moments to make the trajectory between discrete time points closer to the true trajectory, which is beneficial to improving the accuracy of the trajectory.

[0063] As an example, if the latitude at the i-th moment in the trajectory parameter is the same as the latitude at the i-1-th moment, the value range of the first-order derivative of the trajectory parameter at the i-th moment is [-0.1, 0.1].

[0064] Specifically, from the calculation formula of the first-order derivative of the trajectory parameter in the aforementioned embodiment, it can be seen that at two adjacent moments, if the vehicles are at the same latitude, the calculated value of the first-order derivative of the trajectory parameter will tend to infinity, causing the system program to report an error. When this happens, the value of the first-order derivative of the trajectory parameter at that moment can be set within the range of [-0.1, 0.1]. For example, for convenience of calculation, the value can be set to 0.

[0065] As an example, the trajectory characteristic coordinates include at least one of the characteristic value of the first-order derivative of the trajectory parameter, the characteristic value of longitude, and the characteristic value of latitude; the characteristic value includes at least one of the mean, sum, variance, quartile, and data quantity.

[0066] Specifically, the trajectory of any vehicle is converted into a coordinate point in a five-dimensional coordinate system including at least the first-order derivative of the trajectory parameters, the mean, sum, variance, quartile and data quantity of three variables, longitude and latitude. This makes it easier to compare the trajectory characteristics of different vehicles and facilitates the clustering calculation of the trajectories of different vehicles.

[0067] In one embodiment provided in the present application, Figure 3 As shown, step 28, based on the clustering cluster, a vehicle combination is obtained, and the KL divergence of the position data of the vehicle combination is calculated. If the KL divergence of the position data of the vehicle combination is less than a preset KL divergence threshold, it is determined that the vehicles in the vehicle combination belong to the same fleet, and further includes:

[0068] Step 282, any two different vehicles in the cluster are combined into a vehicle group;

[0069] Specifically, it can be seen from the description of the density clustering algorithm in the above embodiment that when the cluster generated by density clustering has only one cluster center, the distance between the trajectory feature coordinates of the vehicles in the cluster is small, and the vehicles in the cluster can be directly grouped into the same fleet. However, when there are multiple cluster centers, the trajectory feature coordinates of the vehicles in the ε-neighborhood corresponding to different cluster centers may be far apart. Please refer to Figure 2 As shown in the middle (b) figure, for example, point 6 and point 2 belong to the same cluster, but the actual distance between them is far, and the driving trajectories of the corresponding two vehicles are also quite different. Therefore, in this case, the vehicles in the same cluster cannot be completely classified as the same fleet, and the accuracy needs to be further improved. In this embodiment, the vehicles in the formed clusters are grouped into vehicle combinations in pairs, and the clustering accuracy is further improved by judging the similarity of the driving trajectories of the vehicles in each vehicle combination.

[0070] Step 284, calculating the KL divergence of longitude, the KL divergence of latitude and the KL divergence of the joint distribution of longitude and latitude between two vehicles in the vehicle combination;

[0071] Step 286, if the KL divergence of the longitude is less than the preset longitude KL divergence threshold, the KL divergence of the latitude is less than the preset latitude KL divergence threshold, and the KL divergence of the joint distribution of longitude and latitude is less than the preset longitude and latitude KL divergence threshold, it is determined that the vehicles in the vehicle combination belong to the same fleet.

[0072] In the longitude and latitude-based fleet identification method described in the above embodiment, on the basis of the clustering clusters in the above embodiment, the vehicles in the cluster are paired and combined in pairs, and the KL divergence of the longitude, the KL divergence of the latitude and the KL divergence of the joint distribution of the longitude and latitude of each pair of vehicle combinations are calculated respectively, and the above three KL divergence values ​​are compared with the corresponding KL divergence thresholds. When all three are within the corresponding threshold range, it is determined that the vehicles in the vehicle combination belong to the same fleet. This method can further improve the clustering accuracy of the fleet, and the driving trajectories of the finally selected fleet combination have high similarity.

[0073] As an example, the fleet identification method based on latitude and longitude also includes:

[0074] A trajectory scatter plot of vehicles in the same fleet is drawn according to the trajectory parameters, and the trajectory scatter plot and vehicle identification codes of the corresponding vehicles are output.

[0075] Specifically, after the system completes the determination of the same fleet, it plots the trajectory parameters of each vehicle in the same fleet into a trajectory scatter plot, and displays it visually with the vehicle identification code of each vehicle one by one, so that the staff can intuitively see the vehicle members of the same fleet and their corresponding driving trajectories, so as to manually correct and improve the results.

[0076] In one embodiment of the present application, Figure 4 As shown, a vehicle fleet identification device based on longitude and latitude is also provided, including a data acquisition module 100, a calculation module 300, a judgment module 400 and a display module 500, wherein the data acquisition module 100 is used to obtain trajectory parameters of a plurality of vehicles; the calculation module 300 is used to calculate the trajectory characteristic coordinates of a plurality of vehicles according to the trajectory parameters of the plurality of vehicles, and to calculate and obtain the clustering clusters of the vehicles using a density clustering algorithm according to the trajectory characteristic coordinates, obtain the vehicle combination based on the clustering cluster, and calculate the KL divergence of the position data of the vehicle combination; the judgment module 400 is used to judge whether the vehicle belongs to the clustering cluster according to the distance between the trajectory characteristic coordinates of the vehicle and the cluster center coordinates of the clustering cluster, and / or to judge whether the vehicles in the vehicle combination belong to the same vehicle fleet according to the KL divergence of the position data of the vehicle combination; the display module 500 is used to output and display the trajectory scatter plot of the vehicles in the same vehicle fleet and the vehicle identification codes of the corresponding vehicles.

[0077] As an example, please refer to Figure 4 , and also includes a data processing module 200, which is used to perform data normalization processing on the trajectory feature coordinates according to preset rules, so that the data of different vehicles corresponding to the same coordinates adopt a unified standard.

[0078] Specifically, to reduce the amount of calculation, all longitude and latitude data can be retained to two decimal places, and the units of variables with the same attributes can be unified to make the clustering results more accurate.

[0079] In one embodiment of the present application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any of the above-mentioned methods for identifying fleets based on longitude and latitude are implemented.

[0080] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0081] In the description of this specification, the description with reference to the terms "some embodiments", "other embodiments", "ideal embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example.

[0082] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0083] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A fleet identification method based on longitude and latitude, characterized in that: include: Acquire trajectory parameters of several vehicles; the trajectory parameters include several groups of position data at adjacent moments, the position data are arranged in chronological order of occurrence, and the position data include longitude and latitude; Generating trajectory characteristic coordinates of the vehicle according to the trajectory parameters; Generating a cluster of the vehicle using a density clustering algorithm according to the trajectory feature coordinates; A vehicle combination is obtained based on the clustering cluster, and the KL divergence of the position data of the vehicle combination is calculated. If the KL divergence of the position data of the vehicle combination is less than a preset KL divergence threshold, it is determined that the vehicles in the vehicle combination belong to the same fleet; the vehicle combination includes any two different vehicles in the clustering cluster.

2. The method according to claim 1, characterized in that The trajectory parameters include N groups of longitudes and latitudes at adjacent moments, where N is a positive integer; the step of generating the trajectory characteristic coordinates of the plurality of vehicles according to the trajectory parameters includes: Calculate the first-order derivative of the trajectory parameters of any vehicle, the first-order derivative μ i Calculated according to the following formula: Where i∈[2,N], J i is the longitude at the i-th moment, J i-1 is the longitude at the i-1th moment, W i is the latitude at the i-th moment, W i-1 is the latitude at the i-1th moment.

3. The method according to claim 2, characterized in that If the latitude at the i-th moment in the trajectory parameter is the same as the latitude at the i-1-th moment, the value range of the first-order derivative of the trajectory parameter at the i-th moment is [-0.1, 0.1].

4. The method according to claim 2 or 3, characterized in that: The trajectory characteristic coordinates include at least one of the characteristic value of the first-order derivative of the trajectory parameter, the characteristic value of the longitude, and the characteristic value of the latitude; the characteristic value includes at least one of the mean value, the sum, the variance, the quartile, and the number of data.

5. The method according to any one of claims 1 to 3, characterized in that: The distance between the trajectory feature coordinates of any vehicle in the cluster and the cluster center coordinates is less than a preset distance value.

6. The method according to claim 5, characterized in that Also includes: Forming a vehicle combination with any two different vehicles in the cluster; Calculating the KL divergence of longitude, the KL divergence of latitude, and the KL divergence of the joint distribution of longitude and latitude between two vehicles in the vehicle combination; If the KL divergence of the longitude is less than a preset longitude KL divergence threshold, the KL divergence of the latitude is less than a preset latitude KL divergence threshold, and the KL divergence of the joint distribution of longitude and latitude is less than a preset longitude and latitude KL divergence threshold, it is determined that the vehicles in the vehicle combination belong to the same fleet.

7. The method according to claim 6, characterized in that Also includes: A trajectory scatter plot of vehicles in the same fleet is drawn according to the trajectory parameters, and the trajectory scatter plot and vehicle identification codes of the corresponding vehicles are output.

8. A vehicle fleet identification device based on longitude and latitude, characterized in that: include: A data acquisition module, used to acquire trajectory parameters of several vehicles; the trajectory parameters include position data of several groups of adjacent moments, the position data are arranged in chronological order of occurrence, and the position data include longitude and latitude; A calculation module, used to calculate the trajectory characteristic coordinates of the plurality of vehicles according to the trajectory parameters of the plurality of vehicles, and to calculate and obtain the clustering clusters of the vehicles using a density clustering algorithm according to the trajectory characteristic coordinates, to obtain a vehicle combination based on the clustering cluster, and to calculate the KL divergence of the position data of the vehicle combination; the vehicle combination includes any two different vehicles in the clustering cluster; A judgment module, used for judging whether the vehicles in the vehicle combination belong to the same fleet according to the KL divergence of the position data of the vehicle combination; wherein if the KL divergence of the position data of the vehicle combination is less than a preset KL divergence threshold, it is judged that the vehicles in the vehicle combination belong to the same fleet; The display module is used to output and display the trajectory scatter diagram of the vehicles in the same fleet and the vehicle identification codes of the corresponding vehicles.

9. The device according to claim 8, characterized in that Also includes: The data processing module is used to perform data normalization processing on the track feature coordinates according to preset rules so that the data of different vehicles corresponding to the same coordinates adopt a unified standard.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images