Network freight information transparent management platform

Abstract

The invention discloses a network freight information transparent management platform, and relates to the field of intelligent transportation. The data receiving module is used for receiving original global positioning system track data in a vehicle transportation process; the preprocessing module cleans the data and generates a track difference sequence; the feature analysis module calculates a direction stability index and path tortuosity based on the differential sequence; the consistency checking module generates a consistency score through velocity curvature consistency checking; the decision fusion module fuses the direction stability index, the path tortuosity and the consistency score to generate a trajectory counterfeiting probability evaluation value; and the judgment supervision module judges the authenticity of the trajectory data according to a comparison result of the evaluation value and a preset threshold value, and triggers a supervision disposal process. And through multi-dimensional index fusion and decision making, the platform can accurately identify and give an alarm on a forged track in real time, and the data authenticity judgment and supervision efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation, specifically to a transparent management platform for online freight information. Background Technology

[0002] With the rapid development of the logistics industry, especially the rise of online freight, traditional transportation management models are gradually facing problems such as low information transparency and difficulty in ensuring data authenticity. In the freight process, the collection and transmission of transportation data usually rely on the Global Positioning System (GPS) and other related technologies. However, these data may be affected by various factors during transmission and processing, such as GPS signal errors, vehicle malfunctions, and environmental interference, which can lead to the falsification or tampering of trajectory data.

[0003] Currently, most online freight management platforms rely solely on traditional manual review or simple rule-based detection methods to identify forged trajectory data. This approach is not only inefficient but also susceptible to human error and struggles to counter complex forgery techniques. Furthermore, existing technologies fall short in terms of real-time performance, accuracy, and automation, failing to meet the ever-increasing volume of transportation data and stringent regulatory requirements. Summary of the Invention

[0004] Based on the shortcomings of the prior art described above, the purpose of this invention is to provide a transparent management platform for online freight information to solve the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a transparent management platform for online freight information, comprising: Data receiving module: used to receive raw GPS trajectory data generated during vehicle transportation, the trajectory data including the latitude and longitude coordinates, timestamps and speed values ​​of each trajectory point; Preprocessing module: Used to preprocess the received trajectory data and output the cleaned trajectory sequence and trajectory difference sequence; Feature analysis module: used to calculate directional stability index and path tortuosity based on trajectory difference sequences; Consistency verification module: used to generate a consistency score based on the cleaned trajectory sequence and trajectory difference sequence through velocity curvature consistency verification; Decision fusion module: used to fuse directional stability index, path tortuosity, and consistency score to generate a trajectory fraud probability assessment value; Judgment and Supervision Module: This module is used to determine the authenticity of trajectory data based on the comparison between the falsification probability assessment value and the preset probability threshold, and to trigger the supervision and handling process based on the judgment result.

[0006] The present invention is further configured such that the preprocessing module includes: Acquire raw GPS trajectory data consisting of multiple trajectory points, where each trajectory point contains longitude coordinates, latitude coordinates, timestamp, and instantaneous velocity value; Based on preset minimum displacement threshold and minimum time interval threshold, redundancy is determined for continuous trajectory points in trajectory data. When the distance between adjacent trajectory points calculated based on the Haversine formula is less than the minimum displacement threshold and the time interval between adjacent trajectory points is less than the minimum time interval threshold, subsequent trajectory points are marked as redundant data points and filtered out. Based on the filtered trajectory sequence, the displacement, heading angle, time interval difference, and velocity difference between adjacent trajectory points are calculated to generate the corresponding first-order trajectory difference sequence. The displacement is a spherical distance calculated based on the Haversine formula and using the Earth's average radius as a reference.

[0007] The present invention is further configured such that the feature analysis module includes: Based on the heading angle data between adjacent trajectory points in the trajectory difference sequence, calculate the change in heading angle of each trajectory point; The change in heading angle is normalized, and the normalized change in heading angle is weighted and statistically processed using an exponential decay weighted average algorithm to generate a directional stability index. Based on the displacement data in the trajectory difference sequence, the ratio of the total length of the trajectory path to the straight-line distance between the start and end points of the trajectory is calculated to generate a path tortuosity index.

[0008] The present invention is further configured such that the consistency verification module includes: Based on the cleaned trajectory sequence and the corresponding trajectory difference sequence, the coordinates of the trajectory points are transformed into a local Cartesian coordinate system; In a Cartesian coordinate system, the curvature estimate at each trajectory point is calculated based on the spatial positional relationship and time interval information of adjacent trajectory points. Based on a speed curvature relationship model pre-established by fitting historical real driving data, the theoretical expected speed value at each trajectory point is calculated. The relative deviation between the actual reported velocity value and the theoretical expected velocity value is calculated point by point, and the relative deviation of all trajectory points is weighted and summed to generate a consistency score.

[0009] The present invention is further configured such that the decision fusion module includes: The receiver receives the directional stability index and path tortuosity index calculated by the feature analysis module, and the consistency score calculated by the consistency verification module. Based on directional stability index, path tortuosity index, and consistency score, a pre-set decision-maker is used to calculate the probability assessment value of trajectory fraud.

[0010] The present invention is further configured such that the determination and supervision module includes: Compare the trajectory fraud probability assessment value with the preset probability threshold; When the probability assessment value of trajectory fraud is greater than the preset probability threshold, the corresponding trajectory data is determined to be fake trajectory data; Based on the judgment results, the regulatory handling process was triggered for the trajectory data identified as forged.

[0011] The present invention is further configured such that the preprocessing module further includes: before performing redundancy determination, detecting and filtering velocity mutation points in the trajectory data, wherein the velocity mutation point is a point where the velocity change rate between adjacent trajectory points is greater than a preset acceleration threshold.

[0012] The present invention is further configured such that the judgment and supervision module further includes: when the trajectory data is determined to be forged, generating a visual alarm message containing a trajectory forgery probability assessment value and judgment basis, and pushing the alarm message to the supervision terminal.

[0013] The present invention is further configured such that the platform is also provided with a data storage module for storing the original trajectory data and the final trajectory authenticity determination result, and establishing a corresponding data index for historical query and analysis.

[0014] This invention provides a transparent management platform for online freight information. It comprises a data receiving module for receiving raw GPS trajectory data generated during vehicle transportation, including the latitude and longitude coordinates, timestamps, and speed values ​​of each trajectory point; a preprocessing module for preprocessing the received trajectory data, outputting a cleaned trajectory sequence and a trajectory difference sequence; a feature analysis module for calculating a directional stability index and path tortuosity based on the trajectory difference sequence; a consistency verification module for generating a consistency score based on the cleaned trajectory sequence and trajectory difference sequence through a speed curvature consistency verification; a decision fusion module for fusing the directional stability index, path tortuosity, and consistency score to generate a trajectory fraud probability assessment value; and a judgment and supervision module for determining the authenticity of the trajectory data based on a comparison between the fraud probability assessment value and a preset probability threshold, and triggering a regulatory handling process based on the judgment result. The beneficial effects include: 1. By integrating multi-dimensional indicators and using a decision-maker based on rules or nonlinear functions to calculate the probability of trajectory fraud, the accuracy of identifying fraudulent trajectories is improved. By setting a preset probability threshold, the platform can accurately determine the authenticity of trajectory data, thereby triggering regulatory handling procedures. 2. When trajectory data is determined to be forged, a visual alarm message containing a forgery probability assessment value and the basis for the judgment can be generated and pushed to the regulatory terminal in real time, ensuring that regulatory personnel can understand and handle abnormal situations in a timely manner. The real-time alarm and intuitive display not only improve regulatory efficiency, but also enhance the transparency and timeliness of the regulatory process.

[0015] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is a structural diagram of a transparent management platform for online freight information, as shown in an exemplary embodiment of the present invention. Detailed Implementation

[0017] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0018] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0019] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0020] A transparent management platform for online freight information, such as Figure 1 As shown, it includes: Data receiving module: used to receive raw GPS trajectory data generated during vehicle transportation, the trajectory data including the latitude and longitude coordinates, timestamps and speed values ​​of each trajectory point; Preprocessing module: Used to preprocess the received trajectory data and output the cleaned trajectory sequence and trajectory difference sequence; Feature analysis module: used to calculate directional stability index and path tortuosity based on trajectory difference sequences; Consistency verification module: used to generate a consistency score based on the cleaned trajectory sequence and trajectory difference sequence through velocity curvature consistency verification; Decision fusion module: used to fuse directional stability index, path tortuosity, and consistency score to generate a trajectory fraud probability assessment value; Judgment and Supervision Module: This module is used to determine the authenticity of trajectory data based on the comparison between the falsification probability assessment value and the preset probability threshold, and to trigger the supervision and handling process based on the judgment result.

[0021] The present invention is further configured such that the preprocessing module includes: Acquire raw GPS trajectory data consisting of multiple trajectory points, where each trajectory point contains longitude coordinates, latitude coordinates, timestamp, and instantaneous velocity value; Based on preset minimum displacement threshold and minimum time interval threshold, redundancy is determined for continuous trajectory points in trajectory data. When the distance between adjacent trajectory points calculated based on the Haversine formula is less than the minimum displacement threshold and the time interval between adjacent trajectory points is less than the minimum time interval threshold, subsequent trajectory points are marked as redundant data points and filtered out. Based on the filtered trajectory sequence, the displacement, heading angle, time interval difference, and velocity difference between adjacent trajectory points are calculated to generate the corresponding first-order trajectory difference sequence. The displacement is a spherical distance calculated based on the Haversine formula and using the Earth's average radius as a reference. The invention further specifies that the preprocessing module includes: detecting and filtering velocity abrupt change points in the trajectory data before redundancy determination; these velocity abrupt change points are points where the rate of velocity change between adjacent trajectory points exceeds a preset acceleration threshold. Specifically, data reception and parameter initialization are performed, receiving a raw GPS trajectory data sequence uploaded by the vehicle terminal. This data sequence contains multiple trajectory points arranged in chronological order, each trajectory point recording longitude coordinates, latitude coordinates, a timestamp, and an instantaneous velocity value; and initializing the minimum displacement threshold, minimum time interval threshold, and acceleration threshold. For example, the minimum displacement threshold is preset to 5 meters, the minimum time interval threshold to 10 seconds, and the acceleration threshold to 3 meters per second. These preset threshold parameters are used for initial screening and processing of trajectory data. Users can flexibly adjust these thresholds through the configuration interface provided by the platform to adapt to different application scenarios. When detecting and filtering velocity abrupt changes, the trajectory data sequence is traversed, and the rate of change of velocity between each pair of adjacent trajectory points is calculated, which is the absolute value of the velocity difference divided by the time interval. When the calculated acceleration value is greater than the preset acceleration threshold, the trajectory point is marked as a velocity abrupt change point. To distinguish between real rapid acceleration and data anomalies, a sliding window verification mechanism is adopted, taking the current point as the center and examining the velocity changes of two points before and after it. In this mode, if most points within the window display abnormalities, it is confirmed as a genuine anomaly; if it is an isolated point, it is determined to be data acquisition noise. For confirmed velocity abrupt changes, linear interpolation correction is performed using the velocity values ​​of the preceding and following normal points, or the abnormal point is directly removed. When filtering redundant trajectory points, the Haversine distance formula is used to calculate the actual ground distance between every two adjacent trajectory points. This calculation takes into account the Earth's curvature and uses the Earth's average radius as a benchmark. Simultaneously, the time interval between adjacent trajectory points is calculated. When both the distance difference and the time interval are less than the minimum displacement threshold and the time interval is less than the minimum time interval threshold, the subsequent trajectory point is marked as a redundant point. To improve the filtering effect, a bidirectional scanning strategy is adopted, i.e., scanning first in chronological order. Then, the process is repeated in reverse chronological order until no new redundant points are found. After filtering redundant trajectory points, a trajectory difference sequence is generated. For the trajectory sequence after filtering redundant trajectory points, the displacement difference, direction change, time difference, and velocity difference between every two adjacent trajectory points are calculated point by point. The direction change is calculated using an azimuth algorithm, taking into account the changes in latitude and longitude and the influence of latitude on the direction calculation to ensure accurate calculation of the forward direction of each trajectory segment. All difference data are organized into a new difference sequence, and each difference data item contains four dimensions: displacement, heading angle, time interval, and velocity change. During the processing, when the number of consecutive abnormal points exceeds a preset proportion of the total number of points, a data quality alarm is automatically triggered.For general anomaly data, an appropriate imputation strategy is selected based on the anomaly type, including forward imputation for time series interruptions and backward imputation for missing trajectory points, to ensure data continuity. All processing procedures and anomaly situations are recorded in the log for subsequent analysis.

[0022] The present invention is further configured such that the feature analysis module includes: Based on the heading angle data between adjacent trajectory points in the trajectory difference sequence, calculate the change in heading angle of each trajectory point; The change in heading angle is normalized, and the normalized change in heading angle is weighted and statistically processed using an exponential decay weighted average algorithm to generate a directional stability index. Based on the displacement data in the trajectory difference sequence, the ratio of the total length of the trajectory path to the straight-line distance between the start and end points of the trajectory is calculated to generate a path tortuosity index. Specifically, when calculating the heading angle change, based on the preprocessed trajectory difference sequence, the heading angle data of each trajectory segment is read, and the heading angle change between adjacent trajectory segments is calculated sequentially. The heading angle change is the difference between the heading angle of the subsequent trajectory segment and the heading angle of the preceding trajectory segment. To ensure the accuracy of the angle difference calculation, the calculated angle difference is normalized to ensure that all heading angle changes are between -180 degrees and +180 degrees. When calculating the directional stability index, the obtained heading angle change is normalized and converted into a relative rate of change between 0 and 1. The relative rate of change is the ratio of the absolute value of the heading angle change to 180 degrees. An exponential decay weighted algorithm is used to weight the relative rate of change of heading angle, assigning a weight coefficient to each heading angle change. The weight coefficients for each trajectory segment are as follows: The directional stability index is determined by the displacement length and the preset decay time constant. For example, in a specific implementation, the decay time constant is set to 10 trajectory segments, meaning that every 10 trajectory segments forward, the weight decays to about one-third of its original value. The relative change rates of all weighted heading angles are summed and divided by the total weights, then a negative exponential function is used to obtain the directional stability index. The closer the directional stability index value is to 1, the more stable the trajectory direction. The displacement data between all adjacent trajectory points are extracted from the trajectory difference sequence, and the displacement data are accumulated and summed to obtain the total length of the trajectory path. The Haversine distance calculation method based on the Earth's spherical model is used to calculate the straight-line distance between the starting and ending points of the trajectory. A small positive constant is added to the straight-line distance value for smoothing to prevent division by zero errors during the calculation process. The total length of the trajectory path is divided by the smoothed straight-line distance value to obtain the path tortuosity index value. The larger the path tortuosity index value, the more tortuous the trajectory.

[0023] The present invention is further configured such that the consistency verification module includes: Based on the cleaned trajectory sequence and the corresponding trajectory difference sequence, the coordinates of the trajectory points are transformed into a local Cartesian coordinate system; In a Cartesian coordinate system, the curvature estimate at each trajectory point is calculated based on the spatial positional relationship and time interval information of adjacent trajectory points. Based on a speed curvature relationship model pre-established by fitting historical real driving data, the theoretical expected speed value at each trajectory point is calculated. The relative deviation between the actual reported velocity value and the theoretical expected velocity value is calculated point by point, and the relative deviations of all trajectory points are weighted and summed to generate a consistency score. Specifically, the longitude and latitude coordinates of each trajectory point in the preprocessed trajectory sequence are transformed to a local Cartesian coordinate system using a general transverse Mercator projection coordinate transformation method to obtain X and Y coordinate values ​​in meters. In the Cartesian coordinate system, a curvature estimation algorithm based on three-point geometric relationships is used. For each intermediate trajectory point, the previous point, the current point, and the next point are taken as the calculation unit. The vector direction of two adjacent line segments is calculated based on the coordinate positions of the three points. Then, the instantaneous curvature value at that point is calculated through vector cross product and dot product operations. To reduce noise interference, a sliding window smoothing mechanism is used. For example, with the current point as the center and a window size of 5 points, the calculated curvature is smoothed. The values ​​are processed using a moving average filter. A speed curvature relationship model, pre-trained using historical real driving data, is invoked. The model parameters include maximum speed on straightaways, curvature sensitivity, and minimum speed. The curvature values ​​at each trajectory point are input into the model to obtain the corresponding theoretical expected speed values. This model can reflect the normal driving behavior characteristics of different types of vehicles in typical road network environments. All trajectory points are traversed, and the relative percentage deviation between the actual reported speed value and the theoretical expected speed value is calculated point by point. A weighted strategy considering the length of the trajectory segment and the time sampling density is adopted to perform weighted averaging on the relative deviation values ​​of each point. The deviations of long trajectory segments and densely sampled points are given higher weights. The weighted average result is mapped to the numerical range of 0 to 1 through a linear transformation to generate the final consistency score. The closer the consistency score is to 1, the better the trajectory kinematic consistency.

[0024] The present invention is further configured such that the decision fusion module includes: The receiver receives the directional stability index and path tortuosity index calculated by the feature analysis module, and the consistency score calculated by the consistency verification module. Based on directional stability index, path tortuosity index, and consistency score, a trajectory fraud probability assessment value is calculated using a pre-set decision-maker. Specifically, the system receives the directional stability index, path tortuosity index, and consistency score from the aforementioned processing steps. These are pre-processed and transformed. The directional stability index is inversely transformed by subtracting the directional stability index from the value of 1 to obtain the directional instability component. A larger value of this component indicates stronger fluctuations in trajectory direction, and more frequent and unstable direction changes. The path tortuosity index is benchmarked by subtracting the theoretical benchmark value of 1 to obtain the path tortuosity deviation component. The theoretical benchmark value of 1 represents the ideal straight path. A larger value of this component indicates a higher degree of detour in the actual driving path relative to the straight-line distance from the start and end points, and a more severe deviation from the straight line. The consistency score is calculated by subtracting the consistency score from the value 1 to obtain the motion consistency deviation component. The larger the value of this component, the more serious the violation of the kinematic laws. The obtained directional instability component, path tortuosity deviation component, and motion consistency deviation component are input into a preset decision unit. For example, in this embodiment, a linear weighted fusion algorithm is used to calculate the preliminary anomaly score. The three components are multiplied by preset weight coefficients and then summed. The weight coefficients are determined through a large number of experiments based on the contribution of each component to the trajectory fraud suspicion, and the sum of the weight coefficients is equal to 1. The weighted summation result is normalized. First, the weighted sum is divided by the sum of the weight coefficients to obtain the standardized anomaly score. Then, the score is limited to between 0 and 1 by the max(0,min(1,value)) function to generate the final trajectory fraud probability assessment value.

[0025] The present invention is further configured such that the determination and supervision module includes: Compare the trajectory fraud probability assessment value with the preset probability threshold; When the probability assessment value of trajectory fraud is greater than the preset probability threshold, the corresponding trajectory data is determined to be fake trajectory data; Based on the judgment results, the regulatory handling process is triggered for trajectory data identified as forged. Specifically, the trajectory forgery probability assessment value is compared with the preset probability threshold in real time. When the trajectory forgery probability assessment value is greater than the preset threshold, the corresponding trajectory data is automatically determined to be forged data. For trajectory data determined to be forged, a multi-level regulatory handling process is immediately triggered. The original trajectory data is marked as forged and persistently stored in the database. Then, according to the preset forgery severity threshold, the corresponding business interception mechanism is initiated. The interception mechanism includes, but is not limited to, handling measures such as suspension of transportation tasks, freezing of freight settlement, or downgrading of credit rating. All regulatory handling processes are pushed to the terminal devices of relevant regulatory personnel in real time through the message push interface integrated by the platform.

[0026] The invention is further configured such that the judgment and supervision module includes: when the trajectory data is determined to be forged, generating a visual alarm message containing a trajectory forgery probability assessment value and judgment basis, and pushing the alarm message to the supervision terminal; specifically, when the judgment and supervision module detects that the trajectory forgery probability assessment value is greater than a preset probability threshold, it automatically triggers an alarm generation mechanism. This mechanism integrates and analyzes information based on the calculation results of various sub-indicators, such as the directional stability index, path tortuosity index, and consistency score, and generates a visual alarm report containing the trajectory forgery probability assessment value and the deviation of each sub-indicator; subsequently, through the message push interface integrated into the platform, the structured alarm message is sent to the terminal device of the supervisor in real time. The alarm message adopts a color-coded visual presentation method, such as red representing high risk, and supports clicking to view detailed trajectory comparison analysis charts and violation point location markers, providing supervisors with intuitive decision support basis.

[0027] The invention further includes a data storage module for storing raw trajectory data and final trajectory authenticity determination results, and establishing corresponding data indexes for historical querying and analysis. Specifically, the data storage module categorizes and stores raw GPS trajectory data, preprocessed cleaned trajectory data, and final trajectory authenticity determination results. The raw trajectory data uses a columnar storage format to optimize storage efficiency and improve large-scale data processing performance, while the preprocessed cleaned trajectory data and trajectory authenticity determination results are stored using a document-oriented database to support fast retrieval and flexible querying. Simultaneously, a multi-dimensional index is established based on timestamps, vehicle IDs, and spatial geographic locations, and a range index is set for the trajectory fraud probability assessment value, thereby achieving efficient retrieval and analysis.

[0028] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A transparent management platform for online freight information, characterized in that, include: Data receiving module: used to receive raw GPS trajectory data generated during vehicle transportation, the trajectory data including the latitude and longitude coordinates, timestamps and speed values ​​of each trajectory point; Preprocessing module: Used to preprocess the received trajectory data and output the cleaned trajectory sequence and trajectory difference sequence; Feature analysis module: used to calculate directional stability index and path tortuosity based on trajectory difference sequences; Consistency verification module: used to generate a consistency score based on the cleaned trajectory sequence and trajectory difference sequence through velocity curvature consistency verification; Decision fusion module: used to fuse directional stability index, path tortuosity, and consistency score to generate a trajectory fraud probability assessment value; Judgment and Supervision Module: This module is used to determine the authenticity of trajectory data based on the comparison between the falsification probability assessment value and the preset probability threshold, and to trigger the supervision and handling process based on the judgment result.

2. The online freight information transparency management platform according to claim 1, characterized in that, The preprocessing module includes: Acquire raw GPS trajectory data consisting of multiple trajectory points, where each trajectory point contains longitude coordinates, latitude coordinates, timestamp, and instantaneous velocity value; Based on preset minimum displacement threshold and minimum time interval threshold, redundancy is determined for continuous trajectory points in trajectory data. When the distance between adjacent trajectory points calculated based on the Haversine formula is less than the minimum displacement threshold and the time interval between adjacent trajectory points is less than the minimum time interval threshold, subsequent trajectory points are marked as redundant data points and filtered out. Based on the filtered trajectory sequence, the displacement, heading angle, time interval difference, and velocity difference between adjacent trajectory points are calculated to generate the corresponding first-order trajectory difference sequence. The displacement is a spherical distance calculated based on the Haversine formula and using the Earth's average radius as a reference.

3. The online freight information transparency management platform according to claim 1, characterized in that, The feature analysis module includes: Based on the heading angle data between adjacent trajectory points in the trajectory difference sequence, calculate the change in heading angle of each trajectory point; The change in heading angle is normalized, and the normalized change in heading angle is weighted and statistically processed using an exponential decay weighted average algorithm to generate a directional stability index. Based on the displacement data in the trajectory difference sequence, the ratio of the total length of the trajectory path to the straight-line distance between the start and end points of the trajectory is calculated to generate a path tortuosity index.

4. The online freight information transparency management platform according to claim 1, characterized in that, The consistency verification module includes: Based on the cleaned trajectory sequence and the corresponding trajectory difference sequence, the coordinates of the trajectory points are transformed to a local Cartesian coordinate system; In a Cartesian coordinate system, the curvature estimate at each trajectory point is calculated based on the spatial positional relationship and time interval information of adjacent trajectory points. Based on a speed curvature relationship model pre-established by fitting historical real driving data, the theoretical expected speed value at each trajectory point is calculated. The relative deviation between the actual reported velocity value and the theoretical expected velocity value is calculated point by point, and the relative deviation of all trajectory points is weighted and summed to generate a consistency score.

5. The online freight information transparency management platform according to claim 1, characterized in that, The decision fusion module includes: The receiver receives the directional stability index and path tortuosity index calculated by the feature analysis module, and the consistency score calculated by the consistency verification module. Based on directional stability index, path tortuosity index, and consistency score, a pre-set decision-maker is used to calculate the probability assessment value of trajectory fraud.

6. The online freight information transparency management platform according to claim 1, characterized in that, The judgment and supervision module includes: Compare the trajectory fraud probability assessment value with the preset probability threshold; When the probability assessment value of trajectory fraud is greater than the preset probability threshold, the corresponding trajectory data is determined to be fake trajectory data; Based on the judgment results, the regulatory handling process was triggered for the trajectory data identified as forged.

7. The online freight information transparency management platform according to claim 2, characterized in that, The preprocessing module further includes: detecting and filtering velocity mutation points in the trajectory data before redundancy determination, wherein the velocity mutation point is a point where the velocity change rate between adjacent trajectory points is greater than a preset acceleration threshold.

8. A transparent management platform for online freight information according to claim 6, characterized in that, The judgment and supervision module also includes: when the trajectory data is determined to be forged, generating a visual alarm message containing a trajectory forgery probability assessment value and judgment basis, and pushing the alarm message to the supervision terminal.

9. A transparent management platform for online freight information according to claim 1, characterized in that, The platform is also equipped with a data storage module for storing raw trajectory data and final trajectory authenticity determination results, and for establishing corresponding data indexes for historical query and analysis.

Images