Mileage estimation method and device based on base station data and storage medium

Abstract

The invention relates to a mileage estimation method and device based on base station data and a storage medium, and is applied to the technical field of road driving, and the method comprises the steps: obtaining the base station data of a to-be-predicted target in a period of time, carrying out the preprocessing, obtaining the standard base station position data, splitting a track into a plurality of sections of travel tracks, and calculating the mileage of the to-be-predicted target; obtaining the time consumption of each section of travel track, obtaining the gravity center point of each section of travel track, obtaining the distance of each section of travel track and the road density distribution of the path through the gravity center point track, and inputting the time consumption, the distance and the road density distribution of the path of each section of travel track as a pre-trained model; the model outputs a mileage prediction value; according to the scheme, the artificial intelligence model is used for learning the features of the track shape, the track is corrected into the real track based on the base station track, and the accuracy of predicting the traveled distance based on the base station data is improved.

Description

Technical Field

[0001] The present invention relates to the field of road driving technology, and in particular to a mileage estimation method, device and storage medium based on base station data. Background Art

[0002] With the rapid development of mobile communication technology and intelligent transportation systems, it has become possible to estimate the actual mileage of vehicles using base station data.

[0003] The Nearest Neighbor Algorithm (NNA) is a common base station data road network fitting method in the prior art. It is used to match the location information of the base station with the existing road network to obtain the predicted value of mileage. The basic idea of ​​this method is to map the location point of each base station to the road node closest to it; however, since it ignores the relationship between the base station trajectory and the actual trajectory, the data used for road network fitting is biased, and the accuracy of the prediction result is low. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a mileage estimation method, device and storage medium based on base station data, so as to solve the problem that the nearest neighbor algorithm in the prior art ignores the relationship between the base station trajectory and the actual trajectory, thereby causing the data used for road network fitting to be biased and the prediction results to be less accurate.

[0005] According to a first aspect of an embodiment of the present invention, a mileage estimation method based on base station data is provided, the method comprising:

[0006] Obtain base station data of the target to be predicted within a period of time;

[0007] Acquire each track point data according to the base station data, wherein the track point data includes a timestamp and a geographic coordinate;

[0008] Arrange all trajectory point data in ascending order according to timestamps to obtain ordered base station location data;

[0009] Preprocessing the ordered base station location data to obtain standard base station location data;

[0010] Splitting the standard base station location data into multiple travel trajectories;

[0011] The time consumption of each travel trajectory is obtained according to the timestamp of the first trajectory point and the timestamp of the last trajectory point in each travel trajectory;

[0012] Obtain all the center of gravity points of each segment of the travel trajectory, obtain the center of gravity point trajectory of each segment of the travel trajectory according to all the center of gravity points of each segment of the travel trajectory, and obtain the distance of each segment of the travel trajectory according to the coordinates of each center of gravity in the center of gravity point trajectory;

[0013] Obtain the road density distribution of the route according to the center of gravity trajectory of each travel trajectory;

[0014] The time consumption, distance and road density distribution of each travel trajectory are input into the pre-trained mileage prediction model as feature data to obtain the mileage prediction result of the target to be predicted.

[0015] Preferably,

[0016] The training of the pre-trained mileage prediction model includes:

[0017] Get data of multiple base stations in different time periods;

[0018] Obtain the time, distance and road density distribution of each trip trajectory from multiple base station data;

[0019] Build the MLA model framework and set the loss function of the MLA model framework;

[0020] The time consumption, distance and road density distribution of each travel trajectory of the multiple base station data are used as inputs of the MLA model framework to iteratively train the MLA model framework until the loss function no longer decreases or reaches a preset number of iterations, thereby obtaining the pre-trained mileage prediction model.

[0021] Preferably,

[0022] The preprocessing of the ordered base station location data to obtain standard base station location data comprises:

[0023] According to the geographic coordinates of each track point, the track points whose longitude and latitude are not within the preset longitude and latitude range are removed;

[0024] For any two adjacent track points, determine whether the geographic coordinates of the two adjacent track points are consistent. If they are consistent, remove any track point;

[0025] A preset time window algorithm is used to detect whether there are jump points in all trajectory points, and the detected jump points are removed to obtain the standard base station location data.

[0026] Preferably,

[0027] The method of using a preset time window algorithm to detect whether there is a jump point in all trajectory points includes:

[0028] Defining a sliding window, wherein the sliding window includes trajectory points within a preset number or a preset time period;

[0029] For each track point in the sliding window, obtain the time difference between the current track point and the previous track point, and obtain the geographical distance according to the geographical coordinates of the current track point and the previous track point;

[0030] Obtaining an average speed between two trajectory points according to a time difference and a geographical distance between the current trajectory point and the previous trajectory point;

[0031] If the average speed is greater than the preset speed threshold, the current trajectory point is a jump point.

[0032] Preferably,

[0033] The step of dividing the standard base station location data into multiple travel trajectories includes:

[0034] Get the time difference between any two adjacent trajectory points according to the timestamps of any two adjacent trajectory points;

[0035] The time difference between two adjacent trajectory points is compared with a preset time threshold. If the time difference between two adjacent trajectory points is greater than or equal to the preset time threshold, the two adjacent trajectory points belong to different travel trajectories, and the two adjacent trajectory points are the end trajectory point and the start trajectory point of the two adjacent travel trajectories respectively.

[0036] Preferably,

[0037] The method of obtaining all the center of gravity points of each travel trajectory includes:

[0038] Obtain the average value of the geographical coordinates of any three adjacent track points in each travel track, and use the average value of the geographical coordinates of any three adjacent track points as the centroid point of the three adjacent track points;

[0039] According to the sequence of timestamps of each travel trajectory, all the center of gravity points of each travel trajectory are obtained.

[0040] According to a second aspect of an embodiment of the present invention, a mileage estimation device based on base station data is provided, the device comprising:

[0041] Base station data acquisition module: used to obtain the base station data of the target to be predicted within a period of time;

[0042] Track point acquisition module: used to acquire each track point data according to the base station data, and the track point data includes a timestamp and a geographic coordinate;

[0043] Track point sorting module: used to arrange all track point data in ascending order according to timestamps to obtain ordered base station location data;

[0044] Preprocessing module: used for preprocessing the ordered base station location data to obtain standard base station location data;

[0045] A travel trajectory splitting module is used to split the standard base station location data into multiple travel trajectories;

[0046] Time-consuming acquisition module: used to obtain the time-consuming of each travel trajectory according to the timestamp of the first trajectory point and the timestamp of the last trajectory point in each travel trajectory;

[0047] Distance acquisition module: used to acquire all the center of gravity points of each segment of the travel trajectory, acquire the center of gravity point trajectory of each segment of the travel trajectory according to all the center of gravity points of each segment of the travel trajectory, and acquire the distance of each segment of the travel trajectory according to the coordinates of each center of gravity in the center of gravity point trajectory;

[0048] Road density distribution acquisition module: used to obtain the road density distribution of the route according to the center of gravity trajectory of each travel trajectory;

[0049] Prediction module: used to input the time consumption, distance and road density distribution of each travel trajectory as feature data into the pre-trained mileage prediction model to obtain the mileage prediction result of the target to be predicted.

[0050] According to a third aspect of an embodiment of the present invention, there is provided a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a main controller, each step in the above method is implemented.

[0051] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0052] The present application obtains the base station data of the target to be predicted within a period of time, and obtains the standard base station location data after preprocessing, then splits the trajectory into multiple travel trajectories, and obtains the time consumption of each travel trajectory respectively, and obtains the center of gravity of each travel trajectory at the same time, obtains the distance of each travel trajectory and the road density distribution of the route through the center of gravity trajectory, uses the time consumption, distance and road density distribution of each travel trajectory as the pre-trained model input, and the model outputs the mileage prediction value; the scheme of the present application learns the characteristics of the trajectory shape through the use of an artificial intelligence model, corrects it to the real trajectory based on the base station trajectory, and represents the relationship between the base station trajectory and the real trajectory through the time consumption, distance and road density distribution of each travel trajectory, thereby improving the accuracy of mileage prediction based on base station data.

[0053] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0055] Figure 1 is a flow chart of a method for estimating mileage based on base station data according to an exemplary embodiment;

[0056] Figure 2 is a system schematic diagram of a mileage estimation device based on base station data according to another exemplary embodiment;

[0057] In the attached figure: 1-base station data acquisition module, 2-trajectory point acquisition module, 3-trajectory point sorting module, 4-preprocessing module, 5-trip trajectory splitting module, 6-time acquisition module, 7-distance acquisition module, 8-road density distribution acquisition module, 9-prediction module. DETAILED DESCRIPTION

[0058] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0059] Embodiment 1

[0060] Figure 1 is a flow chart of a method for estimating mileage based on base station data according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0061] S1, obtaining the base station data of the target to be predicted within a period of time;

[0062] S2, acquiring each track point data according to the base station data, wherein the track point data includes a timestamp and a geographic coordinate;

[0063] S3, arranging all trajectory point data in ascending order according to timestamps to obtain ordered base station location data;

[0064] S4, preprocessing the ordered base station location data to obtain standard base station location data;

[0065] S5, splitting the standard base station location data into multiple travel trajectories;

[0066] S6, obtaining the time consumption of each travel trajectory according to the timestamp of the first trajectory point and the timestamp of the last trajectory point in each travel trajectory;

[0067] S7, obtaining all the center of gravity points of each segment of the travel trajectory, obtaining the center of gravity trajectory of each segment of the travel trajectory according to all the center of gravity points of each segment of the travel trajectory, and obtaining the distance of each segment of the travel trajectory according to the coordinates of each center of gravity in the center of gravity trajectory;

[0068] S8, obtaining a road density distribution of the route according to the trajectory of the center of gravity of each travel trajectory;

[0069] S9, inputting the time consumption, distance and road density distribution of each travel trajectory as feature data into a pre-trained mileage prediction model to obtain a mileage prediction result of the target to be predicted;

[0070] It can be understood that, in this embodiment, a base station location data set of a fixed vehicle SIM code (that is, a target to be predicted) is obtained over a period of time;

[0071] Extract the data of all track points based on the base station location data. Each track point usually contains a timestamp, geographic coordinates (latitude and longitude), and possible altitude information to ensure that the data is read correctly and arranged in chronological order.

[0072] In order to obtain cleaner data, the acquired base station data needs to be preprocessed. The preprocessing steps include:

[0073] First, delete the points with abnormal longitude and latitude (for example, the decimal longitude and latitude in China, the longitude range is (73.550000, 135.083333), and the latitude range is (3.850000, 53.516667). Points not within this range are considered abnormal points and can be excluded);

[0074] Next, delete the repeated latitude and longitude base station location points in adjacent time nodes;

[0075] Finally, calculate the time difference and distance (calculated using longitude and latitude) of adjacent base station data points, and indirectly calculate the speed by dividing the distance by the time difference. Then determine the calculated speed. If the speed is abnormal, the base station data point is considered a jump point and needs to be deleted. Specifically:

[0076] Define a sliding window that covers a certain number of trajectory points or a specific time period. For example, you can select 5 consecutive trajectory points as a window, or set a time window of 30 seconds. This window will be used to evaluate the relationship between each trajectory point and its adjacent points to identify whether there is a jump point.

[0077] For each track point, calculate the time difference Δt (i.e., the difference between two timestamps, in seconds) between it and the previous track point, as well as the geographical distance d between the two points. The geographical distance can be determined using the Havers i ne formula or other distance calculation methods suitable for the curvature of the earth. Then, calculate the average speed v between the two points according to the speed formula v = d / Δt;

[0078] Set a reasonable speed threshold V_thres, which is based on the maximum speed that the vehicle can reach under normal driving conditions (for example, no more than 120 km / h on urban roads). If the speed calculated for a certain trajectory point exceeds this preset threshold, it is considered that the point may be a jump point due to poor GPS signal or other reasons; apply the sliding window algorithm in the above steps to the entire trajectory, gradually move the window to check each trajectory point, and recalculate the speed of all points in the current window every time the window slides forward, and determine whether there is a jump point based on the set standard. If a jump point is found, it is directly deleted;

[0079] After preprocessing, a new, cleaned dataset is obtained. This updated dataset no longer contains obvious jump points and anomalies, and is more suitable for subsequent analysis tasks.

[0080] Splitting trips: Distinguish trips based on the intervals between adjacent time points in the original data. In order to reasonably split the continuous trajectory data into multiple independent trips, this embodiment introduces a time interval-based criterion. Specifically, if the time difference between two adjacent trajectory points exceeds a preset threshold (e.g., 25 minutes), the two points are considered to belong to different trips. This time interval-based method can effectively identify natural breakpoints between trips, such as parking for a rest, waiting at a traffic light, etc.;

[0081] For each split trip, the time taken can be calculated by simply taking the timestamp of the last point of the trip minus the timestamp of the first point, and the result is expressed in seconds or minutes;

[0082] Then calculate the center of gravity based on the travel data, and convert the base station data into real coordinate position information as much as possible. The calculation formula of the center of gravity is as follows:

[0083]

[0084] In the formula, G represents the center of gravity, λ represents the longitude of the trajectory point, Indicates the latitude of the trajectory point. The subscripts 1, 2, and 3 indicate three adjacent trajectory points. For example, a travel trajectory consists of (1, 2, 3, 4, 5, and 6) trajectory points in the order of timestamps. Then, a center of gravity point can be calculated based on trajectory points 1, 2, and 3, and a center of gravity point can be calculated based on trajectory points 2, 3, and 4. The last center of gravity point is calculated from trajectory points 4, 5, and 6. Through all the center of gravity points of a travel trajectory, the center of gravity trajectory can be obtained, and the distance information of the travel trajectory can be calculated based on the center of gravity trajectory.

[0085] In order to calculate and analyze the interval road density distribution in a specific area, the map can be divided into square grids and the road network length in each grid can be calculated by the following simplified steps:

[0086] Gridding: Create a square grid system covering the entire study area at set intervals (e.g. 500 m or 1 km) on the map, with each square having the same side length.

[0087] Extract road network data: For each grid, extract the road segment information located inside it or crossing its boundary. This usually requires the use of Geographic Information System (GIS) software or the corresponding API interface to process road vector data.

[0088] Calculate the length of the road network: According to the grids passed by the trajectory of the center of gravity, calculate the total length of all road segments in each grid. This can be the length of the road segment directly in the grid, or it can include the length of the road segment that partially passes through the grid boundary.

[0089] Assign to road network density: Divide the total road network length in each grid that the center of gravity trajectory passes by the area of ​​the grid to obtain the road network density of the grid. The road network density reflects the length of roads per unit area and is an indicator of the degree of development of transportation infrastructure.

[0090] The time, distance and road density distribution of each trip trajectory are input into the pre-trained mileage prediction model as feature data to obtain the mileage prediction result of the target to be predicted;

[0091] It is worth mentioning that the training of the mileage prediction model includes:

[0092] Collect data from multiple base stations, and then use the data processing method in the above steps to obtain the time, distance and road density distribution of each trip trajectory, which is the training set data of the model; build the MLA model framework, select the mean absolute error (MAE) as the loss function, and use the feature data (that is, the training set data) and the corresponding real mileage as input and output to train the model. During training, back propagation automatic differentiation is used to obtain the derivatives of the loss function with respect to the model parameters to obtain the gradients of each parameter. The Adam optimizer adaptively adjusts the learning rate based on the second-order moment estimate of the gradient, sets the update step size according to different parameter characteristics, and optimizes the encoder Vi T architecture parameters, position embedding parameters and decoder layer parameters; the final mileage prediction model is obtained through iterative training.

[0093] Embodiment 2

[0094] Figure 2 is a system schematic diagram of a mileage estimation device based on base station data according to another exemplary embodiment, the device comprising:

[0095] Base station data acquisition module 1: used to obtain the base station data of the target to be predicted within a period of time;

[0096] Track point acquisition module 2: used to acquire each track point data according to the base station data, and the track point data includes a timestamp and a geographic coordinate;

[0097] Track point sorting module 3: used to arrange all track point data in ascending order according to timestamps to obtain ordered base station location data;

[0098] Preprocessing module 4: used to preprocess the ordered base station location data to obtain standard base station location data;

[0099] The travel track splitting module 5 is used to split the standard base station location data into multiple travel tracks;

[0100] The time-consuming acquisition module 6 is used to acquire the time-consuming of each travel trajectory according to the timestamp of the first trajectory point and the timestamp of the last trajectory point in each travel trajectory;

[0101] Distance acquisition module 7: used to acquire all the center of gravity points of each travel trajectory, acquire the center of gravity trajectory of each travel trajectory according to all the center of gravity points of each travel trajectory, and acquire the distance of each travel trajectory according to the coordinates of each center of gravity in the center of gravity trajectory;

[0102] Road density distribution acquisition module 8: used for acquiring the road density distribution of the route according to the center of gravity trajectory of each travel trajectory;

[0103] Prediction module 9: used to input the time consumption, distance and road density distribution of each travel trajectory as feature data into the pre-trained mileage prediction model to obtain the mileage prediction result of the target to be predicted.

[0104] Embodiment three:

[0105] This embodiment provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a main controller, each step in the above method is implemented;

[0106] It is understandable that the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0107] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0108] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0109] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0111] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0112] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0113] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0114] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" 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 representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0115] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A mileage estimation method based on base station data, characterized in that: The method comprises: Obtain base station data of the target to be predicted within a period of time; Acquire each track point data according to the base station data, wherein the track point data includes a timestamp and a geographic coordinate; Arrange all trajectory point data in ascending order according to timestamps to obtain ordered base station location data; Preprocessing the ordered base station location data to obtain standard base station location data; Splitting the standard base station location data into multiple travel trajectories; The time consumption of each travel trajectory is obtained according to the timestamp of the first trajectory point and the timestamp of the last trajectory point in each travel trajectory; Obtain all the center of gravity points of each segment of the travel trajectory, obtain the center of gravity point trajectory of each segment of the travel trajectory according to all the center of gravity points of each segment of the travel trajectory, and obtain the distance of each segment of the travel trajectory according to the coordinates of each center of gravity in the center of gravity point trajectory; Obtain the road density distribution of the route according to the center of gravity trajectory of each travel trajectory; The time consumption, distance and road density distribution of each travel trajectory are input into the pre-trained mileage prediction model as feature data to obtain the mileage prediction result of the target to be predicted.

2. The method according to claim 1, characterized in that: The training of the pre-trained mileage prediction model includes: Get data of multiple base stations in different time periods; Obtain the time, distance and road density distribution of each trip trajectory from multiple base station data; Build the MLA model framework and set the loss function of the MLA model framework; The time consumption, distance and road density distribution of each travel trajectory of the multiple base station data are used as inputs of the MLA model framework to iteratively train the MLA model framework until the loss function no longer decreases or reaches a preset number of iterations, thereby obtaining the pre-trained mileage prediction model.

3. The method according to claim 2, characterized in that The preprocessing of the ordered base station location data to obtain standard base station location data comprises: According to the geographic coordinates of each track point, the track points whose longitude and latitude are not within the preset longitude and latitude range are removed; For any two adjacent track points, determine whether the geographic coordinates of the two adjacent track points are consistent. If they are consistent, remove any track point; A preset time window algorithm is used to detect whether there are jump points in all trajectory points, and the detected jump points are removed to obtain the standard base station location data.

4. The method according to claim 3, characterized in that The method of using a preset time window algorithm to detect whether there is a jump point in all trajectory points includes: Defining a sliding window, wherein the sliding window includes trajectory points within a preset number or a preset time period; For each track point in the sliding window, obtain the time difference between the current track point and the previous track point, and obtain the geographical distance according to the geographical coordinates of the current track point and the previous track point; Obtaining an average speed between two trajectory points according to a time difference and a geographical distance between the current trajectory point and the previous trajectory point; If the average speed is greater than the preset speed threshold, the current trajectory point is a jump point.

5. The method according to claim 4, characterized in that The step of dividing the standard base station location data into multiple travel trajectories includes: Get the time difference between any two adjacent trajectory points according to the timestamps of any two adjacent trajectory points; The time difference between two adjacent trajectory points is compared with a preset time threshold. If the time difference between two adjacent trajectory points is greater than or equal to the preset time threshold, the two adjacent trajectory points belong to different travel trajectories, and the two adjacent trajectory points are the end trajectory point and the start trajectory point of the two adjacent travel trajectories respectively.

6. The method according to claim 5, characterized in that The method of obtaining all the center of gravity points of each travel trajectory includes: Obtain the average value of the geographical coordinates of any three adjacent track points in each travel track, and use the average value of the geographical coordinates of any three adjacent track points as the centroid point of the three adjacent track points; According to the sequence of timestamps of each travel trajectory, all the center of gravity points of each travel trajectory are obtained.

7. A mileage estimation device based on base station data, characterized in that: The device comprises: Base station data acquisition module: used to obtain the base station data of the target to be predicted within a period of time; Track point acquisition module: used to acquire each track point data according to the base station data, and the track point data includes a timestamp and a geographic coordinate; Track point sorting module: used to arrange all track point data in ascending order according to timestamps to obtain ordered base station location data; Preprocessing module: used for preprocessing the ordered base station location data to obtain standard base station location data; A travel trajectory splitting module is used to split the standard base station location data into multiple travel trajectories; Time-consuming acquisition module: used to obtain the time-consuming of each travel trajectory according to the timestamp of the first trajectory point and the timestamp of the last trajectory point in each travel trajectory; Distance acquisition module: used to acquire all the center of gravity points of each segment of the travel trajectory, acquire the center of gravity point trajectory of each segment of the travel trajectory according to all the center of gravity points of each segment of the travel trajectory, and acquire the distance of each segment of the travel trajectory according to the coordinates of each center of gravity in the center of gravity point trajectory; Road density distribution acquisition module: used to obtain the road density distribution of the route according to the center of gravity trajectory of each travel trajectory; Prediction module: used to input the time consumption, distance and road density distribution of each travel trajectory as feature data into the pre-trained mileage prediction model to obtain the mileage prediction result of the target to be predicted.

8. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the main controller, each step of the mileage estimation method based on base station data as described in any one of claims 1 to 6 is implemented.

Images