A method and system for sorting and managing vehicles in a container terminal stacking area

Abstract

The present invention discloses a method and system for sorting and managing vehicles in a container terminal stacking area, and relates to the technical field of terminal vehicle management, including steps S1: preprocessing the collected raw data, the raw data including internal container truck data, external container truck data and vehicle-road collaboration data; S2: establishing a spatial retrieval process for vehicles based on an R-tree, and mounting the point information in the vehicle positioning data into an electronic fence composed of multiple polygonal areas; S3: filtering the vehicle trajectory data of the longitudinal road and the transverse road through the electronic fence, colliding the vehicle-road collaboration data with the internal container truck data and the external container truck data respectively, and calibrating the result data generated after the collision to form the vehicle trajectory data in the stacking yard. The present invention can cope with the perception problem of sorting container trucks in the container terminal stacking area.

Description

Technical Field

[0001] The invention relates to the technical field of terminal vehicle management, and provides a method and system for sorting and managing vehicles in a container terminal stacking area. Background Art

[0002] With the unmanned operation of automated yard gantry cranes, accurate perception of vehicles in the yard becomes particularly important. Accurate perception of vehicles in the container terminal yard is a complex and critical process that involves many factors, including the location, speed, direction, etc. of the vehicle.

[0003] The internal container truck synchronization part will be equipped with Beidou high-precision positioning terminals, which can obtain real-time location information through Beidou positioning dedicated network and differential services, but the stability of the positioning terminal cannot be guaranteed during the transportation of container trucks. External container trucks are non-port vehicles, and the installation and operation and maintenance of high-precision positioning equipment are difficult to achieve from a commercial perspective. At present, one method is to use video card ports, ETC, RFID and other solutions. Video card ports can provide information about vehicle entry and exit gates, but the points are limited. ETC and RFID equipment, in addition to the need to issue operation and maintenance tag cards, also have strict requirements for equipment deployment. Due to site restrictions, the port area cannot build a gantry across the intersection like ordinary roads, resulting in limited use scenarios, limited coverage, and the inability to obtain vehicle posture information. Another method is to rely on mobile phone APP or other terminals to collect information. This method requires manual service opening, and the monitoring background collects the positioning data of the mobile phone or terminal to construct trajectory information. The stability of the positioning results is greatly affected by human factors and does not fully meet business needs. Summary of the invention

[0004] In order to solve at least one of the technical problems mentioned in the background technology, the purpose of the present invention is to provide a method and system for sorting management of vehicles in a container terminal yard, which can cope with the perception problem of container truck sorting in a container terminal yard.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for sorting and managing vehicles in a container terminal stacking area comprises the following steps:

[0007] S1: pre-processing the collected raw data, which includes internal container truck data, external container truck data and vehicle-road collaboration data;

[0008] S2: Establish a spatial retrieval process for vehicles based on the R-tree, and mount the point information in the vehicle positioning data into an electronic fence composed of multiple polygonal areas;

[0009] S3: Filter the vehicle trajectory data of the longitudinal and transverse roads through the electronic fence, collide the vehicle-road cooperative data with the internal container truck data and the external container truck data respectively, and calibrate the result data generated after the collision to form the vehicle trajectory data in the yard;

[0010] S4: Taking a single vehicle at a moment as a unit, the trajectory generated by the vehicle at the current moment and the result data information of the post-collision calibration are packaged and integrated, and the vehicles are sorted according to the integrated vehicle trajectory data and the preset sorting rules to complete the vehicle queuing in the yard.

[0011] Furthermore, the step S1 comprises:

[0012] S11: Obtain the internal container truck data of vehicles in the port through the positioning terminal;

[0013] S12: Obtaining the external container truck data of non-port vehicles through the mobile terminal APP;

[0014] S13: Acquire vehicle-road collaboration data of vehicles in the port and non-port vehicles through cameras;

[0015] S14: Preprocess the data information in step S11 to step S13, the data information includes the job number, time, longitude and latitude positioning data and electronic fence information; the preprocessing includes consistent field names, calibration data sources, and converting time data into a 13-bit timestamp format.

[0016] Furthermore, the step S2 comprises:

[0017] S21: define the electronic fence as a polygonal area in a two-dimensional coordinate system, and decompose it into multiple minimum bounding rectangles, and the minimum bounding rectangles are used as leaf nodes of the R-tree;

[0018] S22: When it is necessary to query whether a point is in a certain polygonal area, the R-tree is used for query, and the rectangular area that cannot contain the query point is excluded through the pruning strategy;

[0019] S23: At the leaf node, a polygon inclusion test is performed using a ray method or a crossing number method to determine whether the point is within the polygon area;

[0020] S24: Feedback and transmit the electronic fence information corresponding to the polygon containing the query point.

[0021] Furthermore, the step S3 comprises:

[0022] S31: collide the internal container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is marked as the matched internal container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is marked as the internal container truck data with a large distance error;

[0023] S32: The external container truck data and the vehicle-road cooperative data are collided according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is marked as the matched external container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is marked as the external container truck data with a large distance error;

[0024] S33: The uncollided VIR data is collided with the internal container truck data and the external container truck data according to the yard number and time. If the operation numbers are different, the distance between the internal container truck data or the external container truck data and the VIR data is less than 5 meters, and there is no vehicle within 5 meters around, the calibration is considered a successful match; for the unmatched VIR data, if the identified operation number has an entry record and no exit record, it is calibrated as recalled external container truck data.

[0025] A vehicle sequencing management system in a container terminal stacking area, comprising:

[0026] Data processing module: used to pre-process the collected raw data, which includes internal container truck data, external container truck data and vehicle-road collaboration data;

[0027] Establishing module: Establishing the spatial retrieval process of vehicles based on R-tree, mounting the point information in the vehicle positioning data into the electronic fence composed of multiple polygonal areas;

[0028] Calibration module: used to filter the vehicle trajectory data of the longitudinal and transverse roads through the electronic fence, collide the vehicle-road cooperative data with the internal container truck data and the external container truck data respectively, and calibrate the result data generated after the collision to form the vehicle trajectory data in the yard;

[0029] Sorting module: It is used to package and fuse the trajectory generated by the vehicle at the current moment and the result data information of the post-collision calibration, taking the vehicle at a single moment as a unit. According to the fused vehicle trajectory data and the preset sorting rules, the vehicles are sorted to complete the queuing of vehicles in the yard.

[0030] Furthermore, the data processing module includes:

[0031] The first data acquisition unit is used to acquire the internal container truck data of the vehicle in the port through the positioning terminal;

[0032] The second data acquisition unit is used to acquire the external container truck data of non-port vehicles through the mobile terminal APP;

[0033] The third data acquisition unit is used to acquire the vehicle-road cooperative data of vehicles in the port and non-port vehicles through a camera;

[0034] Processing unit: used to preprocess the data information in the first data acquisition unit, the second data acquisition unit and the third data acquisition unit, the data information includes the operation number, time, longitude and latitude positioning data and electronic fence information; the preprocessing includes consistent field names, calibration data sources, and converting time data into a 13-bit timestamp format.

[0035] Furthermore, the establishment module includes:

[0036] Index building unit: used to define the electronic fence as a polygonal area in a two-dimensional coordinate system and decompose it into multiple minimum bounding rectangles, and the minimum bounding rectangles are used as leaf nodes of the R-tree;

[0037] Query unit: When you need to query whether a point is in a polygonal area, use the R-tree to query and use the pruning strategy to exclude rectangular areas that may not contain the query point;

[0038] Matching unit: used to perform polygon inclusion test at leaf nodes using ray method or intersection number method to determine whether the point is within the polygon area;

[0039] Result return unit: used to feedback and transmit the electronic fence information corresponding to the polygon containing the query point.

[0040] Furthermore, the calibration module includes:

[0041] The first calibration unit is used to collide the internal container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is calibrated as the matched internal container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is calibrated as the internal container truck data with a large distance error;

[0042] The second calibration unit is used to collide the external container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is calibrated as the matched external container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is calibrated as the external container truck data with a large distance error;

[0043] The third calibration unit: used to collide the uncollided vehicle-road collaborative data with the internal container truck data and the external container truck data according to the yard number and time. If the operation numbers are different, the distance between the internal container truck data or the external container truck data and the vehicle-road collaborative data is less than 5 meters, and there is no vehicle within 5 meters around, the calibration is considered a successful match; for the unmatched vehicle-road collaborative data, if the identified operation number has an entry record and no exit record, it is calibrated as the recalled external container truck data.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention obtains real-time images, uses image processing and target recognition algorithms to extract and track the location information of trucks, and sorts and queues container trucks based on this information. Vehicle trajectory data fusion and calibration can fuse vehicle trajectory data from different sources to form complete vehicle trajectory information, providing an accurate data basis for vehicle sorting. Vehicle sorting and queuing are based on the fused vehicle trajectory data, combined with preset sorting rules to sort vehicles, and complete vehicle queue management in the yard. This method can more accurately identify the location and status of container trucks, so as to better deal with the perception problem of container truck sorting in the container area of ​​the container terminal yard. At the same time, the technical solution also realizes the rapid retrieval and query of vehicle positioning data by constructing a spatial index and a fast retrieval mechanism, further improving the management efficiency and operation efficiency of the container terminal yard. This method can effectively deal with the perception problem of container truck sorting in the container area of ​​the container terminal yard. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of a method for sorting and managing vehicles in a container terminal stacking area provided in Embodiment 1 of the present invention;

[0047] Figure 2 This is a flow chart of step S1 in the method for sorting and managing vehicles in a container terminal stacking area provided in the first embodiment of the present invention;

[0048] Figure 3 This is a flow chart of step S2 in the method for sorting and managing vehicles in a container terminal stacking area provided in the first embodiment of the present invention;

[0049] Figure 4 This is a flowchart of step S3 in the method for sorting and managing vehicles in a container terminal stacking area provided in the first embodiment of the present invention;

[0050] Figure 5 The structural framework of the vehicle sorting management system in the container terminal stacking area provided by the second embodiment of the present invention Figure 1 ;

[0051] Figure 6 The structural framework of the vehicle sorting management system in the container terminal stacking area provided by the second embodiment of the present invention Figure 2 . DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Embodiment 1

[0054] See also Figure 1 This embodiment provides a method for sorting and managing vehicles in a container terminal stacking area, comprising the steps of:

[0055] S1: pre-processing the collected raw data, which includes internal container truck data, external container truck data and vehicle-road collaboration data;

[0056] S2: Establish a spatial retrieval process for vehicles based on the R-tree, and mount the point information in the vehicle positioning data into an electronic fence composed of multiple polygonal areas;

[0057] S3: Filter the vehicle trajectory data of the longitudinal and transverse roads through the electronic fence, collide the vehicle-road cooperative data with the internal container truck data and the external container truck data respectively, and calibrate the result data generated after the collision to form the vehicle trajectory data in the yard;

[0058] S4: Taking a single vehicle at a moment as a unit, the trajectory generated by the vehicle at the current moment and the result data information of the post-collision calibration are packaged and integrated, and the vehicles are sorted according to the integrated vehicle trajectory data and the preset sorting rules to complete the vehicle queuing in the yard.

[0059] Specifically, data preprocessing step S1: Process the collected raw data, including internal container truck data, external container truck data and vehicle-road collaboration data. Data preprocessing step S1 collects, cleans and converts data from different data sources, unifying the originally scattered and differently formatted data and providing a basis for subsequent analysis.

[0060] Specifically, the electronic fence mounting step S2 based on the R-tree: the vehicle positioning data is mounted into the electronic fence using the R-tree spatial index structure. Spatial index and fast retrieval: The electronic fence mounting step S2 based on the R-tree realizes fast retrieval and query of vehicle positioning data by constructing a spatial index, which greatly improves the efficiency and accuracy of data retrieval.

[0061] Specifically, the vehicle trajectory data fusion and calibration step S3: through collision matching, the vehicle trajectory data from different sources are fused and calibrated. Trajectory data fusion and precise analysis: The vehicle trajectory data fusion and calibration step S3 integrates the data from different sources into complete vehicle trajectory information through data collision and fusion, and calibrates the results to ensure the accuracy and reliability of the trajectory data.

[0062] Specifically, the vehicle sorting and queuing step S4: sorting the vehicles based on the fused vehicle trajectory data and the preset rules. Automated sorting and queuing management: The vehicle sorting and queuing step S4 realizes automated sorting and queuing management based on the fused vehicle trajectory data, providing efficient scheduling and management means for the operation of the container terminal yard.

[0063] Specifically, the data source description is:

[0064] Internal container truck data: Vehicles within the port are acquired through positioning terminals at a frequency of 1000 milliseconds / frame, with electronic fence information.

[0065] External container truck data: non-port vehicles, obtained by manually opening the mobile phone APP, the frequency is 3000 milliseconds / frame, and there is no electronic fence information. Because the vehicle positioning is transmitted manually, there are some objective factors, such as failure to open the terminal APP in time, poor mobile phone signal in the yard, terminal APP being closed, etc., which may cause the vehicle positioning data to fail to upload, have discontinuity and other problems.

[0066] Vehicle-road collaboration data: including internal and external container trucks, obtained by camera data fusion, with a frequency of 200 milliseconds / frame, without electronic fence information. It can effectively supplement the positioning data of internal and external container trucks, but there is also the problem of vehicle operation number recognition error.

[0067] See also Figure 2 , wherein step S1 comprises:

[0068] S11: Obtain the internal container truck data of vehicles in the port through the positioning terminal;

[0069] S12: Obtaining the external container truck data of non-port vehicles through the mobile terminal APP;

[0070] S13: Acquire vehicle-road collaboration data of vehicles in the port and non-port vehicles through cameras;

[0071] S14: Preprocess the data information in step S11 to step S13, the data information includes the job number, time, longitude and latitude positioning data and electronic fence information; the preprocessing includes consistent field names, calibration data sources, and converting time data into a 13-bit timestamp format.

[0072] See also Figure 3 , wherein step S2 comprises:

[0073] S21: define the electronic fence as a polygonal area in a two-dimensional coordinate system, and decompose it into multiple minimum bounding rectangles, and the minimum bounding rectangles are used as leaf nodes of the R-tree;

[0074] S22: When it is necessary to query whether a point is in a certain polygonal area, the R-tree is used for query, and the rectangular area that cannot contain the query point is excluded through the pruning strategy;

[0075] S23: At the leaf node, a polygon inclusion test is performed using a ray method or a crossing number method to determine whether the point is within the polygon area;

[0076] S24: Feedback and transmit the electronic fence information corresponding to the polygon containing the query point.

[0077] Specifically, the electronic fence mounting based on the R-tree: The electronic fence is a polygonal area distributed in the physical space. It is the only description of the location of the vehicle, which provides a guarantee for us to locate which yard the vehicle is in. At the same time, because the goal is to sort the vehicles in the yard, the trajectory data outside the yard is not considered. At this time, using the electronic fence to effectively filter the data outside the yard will also greatly reduce the consumption of computing costs.

[0078] Specifically, among the collected vehicle positioning data, only the internal container truck data contains electronic fence information, while the external container truck data and vehicle-road collaboration data do not have electronic fences. In order to determine the location of the vehicle, it is necessary to bundle the electronic fences where each location is located, that is, electronic fence bundling. Electronic fences are a finite number of polygonal areas, which are distributed in the horizontal roads, vertical roads, corners and yards of the terminal. The vehicle's GPS positioning can be used to calculate which polygonal area it is in, and then determine the electronic fence number it belongs to.

[0079] Specifically, the spatial retrieval mechanism based on the R-tree will mount the point information to a specific spatial area more quickly. The electronic fence is defined as a polygonal area in a two-dimensional coordinate system, and the GPS positioning is represented as a point in the coordinate system. Then, the specific implementation process of the algorithm is as follows:

[0080] Data index construction: First, the polygonal area is decomposed into a series of minimum bounding rectangles. The minimum bounding rectangle corresponding to each polygon contains all vertices. The minimum bounding rectangle is used as the leaf node in the R-tree, and the internal nodes store their child nodes, which are usually the merged bounding rectangles of the child rectangles. During the R-tree construction process, the tree will be kept balanced as much as possible to optimize the query efficiency.

[0081] Query process: When you need to query whether a point is in a polygonal area, first use the smallest rectangle where the point is located, that is, the point itself, as the query rectangle, and then use the R-tree to query.

[0082] Query optimization: R-tree uses pruning strategy to quickly exclude rectangular areas that cannot contain query points. If a rectangular node has no intersection with the rectangle of the query point, then this rectangle and all its child nodes do not need to be checked and are skipped directly, greatly reducing the search space.

[0083] Exact match: When the query process recursively reaches a leaf node, check whether the point is inside the corresponding polygon. This usually involves a polygon inclusion test, which can be accurately determined using geometric algorithms such as the ray method and the intersection number method.

[0084] Result return: If a polygon containing the query point is found, the corresponding area information is returned; if not, it means that the point is not within any polygon.

[0085] Specifically, the efficient indexing mechanism of the R-tree enables this type of query to be quickly located in large-scale spatial data sets, and is suitable for spatial query optimization in the fields of geographic information systems, database indexes, etc.

[0086] Specifically, the time alignment standard, an effective time alignment standard is the primary prerequisite for ensuring the fusion of multi-source trajectories, which enables different data sources to maintain consistency in timing. For internal and external container truck data, the minimum time unit is 1000 milliseconds, while the minimum time unit for vehicle-road collaboration is 1 millisecond. It will also create great difficulties for the alignment of vehicle trajectories in timing, so all times are reduced or amplified, that is, the time is rounded off in 100 millisecond units, so that the minimum unit of time is divided into 100 milliseconds.

[0087] Specifically, the essence of trajectory fusion is to use vehicle-road collaborative data to supplement internal and external container truck data. The link generated by vehicle-road collaborative data is relatively long. The vehicle-road collaborative data is preliminarily fused with the internal container truck data and the external container truck data respectively, and the root mean square error of the distance deviation of the vehicle trajectory fusion point is used as the judgment standard to find a deviation value that minimizes the root mean square error, which is the required deviation value.

[0088] Specifically, linear interpolation is to interpolate the vehicle's driving trajectory in chronological order based on the vehicle dimension to ensure the continuity of time and trajectory. It is a further supplement to the above-mentioned time alignment standard. Based on 100 milliseconds as the basic unit, the trajectory positioning data of the internal container truck data, external container truck data, and vehicle-road collaborative data are interpolated respectively, and the interpolated data are marked, and the speed is calculated, etc., in order to prepare for the formulation of subsequent rules.

[0089] See also Figure 4 , wherein step S3 comprises:

[0090] S31: collide the internal container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is marked as the matched internal container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is marked as the internal container truck data with a large distance error;

[0091] S32: The external container truck data and the vehicle-road cooperative data are collided according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is marked as the matched external container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is marked as the external container truck data with a large distance error;

[0092] S33: The uncollided VIR data is collided with the internal container truck data and the external container truck data according to the yard number and time. If the operation numbers are different, the distance between the internal container truck data or the external container truck data and the VIR data is less than 5 meters, and there is no vehicle within 5 meters around, the calibration is considered a successful match; for the unmatched VIR data, if the identified operation number has an entry record and no exit record, it is calibrated as recalled external container truck data.

[0093] Specifically, vehicles are queued according to the preset sorting rules:

[0094] Vehicle queuing is a rule-based method that can effectively integrate the current vehicle driving data and predict the lost trajectory. Under the condition that the internal and external truck data is met, if the self-vehicle reported data is delayed for more than 7 seconds, or the data timestamp is abnormal, the data is discarded to eliminate the interference of delayed data, the entire vehicle queuing method follows the following process and priority:

[0095] If the received data of a certain vehicle appears in the box area for the first time, the judgment of Table 1 is entered;

[0096] If a certain vehicle data is received and exists in the box area before, the judgment of Table 2 is entered;

[0097] After entering the box area, the vehicle trajectory is likely to disappear and become discontinuous. At this time, the judgment in Table 3 will be entered to predict the trajectory of the vehicle for a period of time.

[0098] Specifically, when the vehicle appears in the box area for the first time, it enters the first entry judgment, as shown in Table 1:

[0099]

[0100] Table 1

[0101] Specifically, when the vehicle is already in the queue, the trajectory update logic is entered, as shown in Table 2:

[0102]

[0103] Table 2

[0104] Specifically, when there is no vehicle with new data, the prepared prediction logic is entered, as shown in Table 3:

[0105]

[0106] Table 3

[0107] The vehicle sorting output method within the container terminal yard is a scalable multi-source trajectory fusion method. It uses internal container truck data, external container truck data and vehicle-road collaboration data as data sources. Through data preprocessing, R-tree-based electronic fence mounting, time alignment standard, linear interpolation, in-yard trajectory fusion calibration, rule-based sorting and other methods, it quickly and effectively realizes vehicle queuing within the yard.

[0108] This method obtains real-time images, uses image processing and target recognition algorithms to extract and track the location information of trucks, and sorts and queues container trucks based on this information. Vehicle trajectory data fusion and calibration can fuse vehicle trajectory data from different sources to form complete vehicle trajectory information, providing an accurate data basis for vehicle sorting. Vehicle sorting and queuing are based on the fused vehicle trajectory data, combined with preset sorting rules to sort vehicles and complete vehicle queue management in the yard. This method can more accurately identify the location and status of container trucks, so as to better deal with the perception problem of container truck sorting in the container area of ​​the container terminal yard. At the same time, this technical solution also realizes the rapid retrieval and query of vehicle positioning data by constructing a spatial index and a fast retrieval mechanism, further improving the management efficiency and operation efficiency of the container terminal yard. This method can effectively deal with the perception problem of container truck sorting in the container area of ​​the container terminal yard.

[0109] Embodiment 2

[0110] See also Figure 5 This embodiment provides a sorting management system for vehicles in a container terminal stacking area, including:

[0111] Data processing module: used to pre-process the collected raw data, which includes internal container truck data, external container truck data and vehicle-road collaboration data;

[0112] Establishing module: Establishing the spatial retrieval process of vehicles based on R-tree, mounting the point information in the vehicle positioning data into the electronic fence composed of multiple polygonal areas;

[0113] Calibration module: used to filter the vehicle trajectory data of the longitudinal and transverse roads through the electronic fence, collide the vehicle-road cooperative data with the internal container truck data and the external container truck data respectively, and calibrate the result data generated after the collision to form the vehicle trajectory data in the yard;

[0114] Sorting module: It is used to package and fuse the trajectory generated by the vehicle at the current moment and the result data information of the post-collision calibration, taking the vehicle at a single moment as a unit. According to the fused vehicle trajectory data and the preset sorting rules, the vehicles are sorted to complete the queuing of vehicles in the yard.

[0115] See also Figure 6 , wherein the data processing module includes:

[0116] The first data acquisition unit is used to acquire the internal container truck data of the vehicle in the port through the positioning terminal;

[0117] The second data acquisition unit is used to acquire the external container truck data of non-port vehicles through the mobile terminal APP;

[0118] The third data acquisition unit is used to acquire the vehicle-road cooperative data of vehicles in the port and non-port vehicles through a camera;

[0119] Processing unit: used to preprocess the data information in the first data acquisition unit, the second data acquisition unit and the third data acquisition unit, the data information includes the operation number, time, longitude and latitude positioning data and electronic fence information; the preprocessing includes consistent field names, calibration data sources, and converting time data into a 13-bit timestamp format.

[0120] The establishment modules include:

[0121] Index building unit: used to define the electronic fence as a polygonal area in a two-dimensional coordinate system and decompose it into multiple minimum bounding rectangles, and the minimum bounding rectangles are used as leaf nodes of the R-tree;

[0122] Query unit: When you need to query whether a point is in a polygonal area, use the R-tree to query and use the pruning strategy to exclude rectangular areas that may not contain the query point;

[0123] Matching unit: used to perform polygon inclusion test at leaf nodes using ray method or intersection number method to determine whether the point is within the polygon area;

[0124] Result return unit: used to feedback and transmit the electronic fence information corresponding to the polygon containing the query point.

[0125] Among them, the calibration module includes:

[0126] The first calibration unit is used to collide the internal container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is calibrated as the matched internal container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is calibrated as the internal container truck data with a large distance error;

[0127] The second calibration unit is used to collide the external container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is calibrated as the matched external container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is calibrated as the external container truck data with a large distance error;

[0128] The third calibration unit: used to collide the uncollided vehicle-road collaborative data with the internal container truck data and the external container truck data according to the yard number and time. If the operation numbers are different, the distance between the internal container truck data or the external container truck data and the vehicle-road collaborative data is less than 5 meters, and there is no vehicle within 5 meters around, the calibration is considered a successful match; for the unmatched vehicle-road collaborative data, if the identified operation number has an entry record and no exit record, it is calibrated as the recalled external container truck data.

[0129] This system obtains real-time images, uses image processing and target recognition algorithms to extract and track the location information of trucks, and sorts and queues container trucks based on this information. Vehicle trajectory data fusion and calibration can fuse vehicle trajectory data from different sources to form complete vehicle trajectory information, providing an accurate data basis for vehicle sorting. Vehicle sorting and queuing are based on the fused vehicle trajectory data, combined with preset sorting rules to sort vehicles and complete vehicle queue management in the yard. This method can more accurately identify the location and status of container trucks, so as to better deal with the perception problem of container truck sorting in the container area of ​​the container terminal yard. At the same time, this technical solution also realizes the rapid retrieval and query of vehicle positioning data by constructing a spatial index and a fast retrieval mechanism, further improving the management efficiency and operation efficiency of the container terminal yard. This method can effectively deal with the perception problem of container truck sorting in the container area of ​​the container terminal yard.

[0130] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

Claims

1. A method for sorting and managing vehicles in a container terminal area, characterized in that: Includes steps: S1: pre-processing the collected raw data, which includes internal container truck data, external container truck data and vehicle-road collaboration data; S2: Establish a spatial retrieval process for vehicles based on the R-tree, and mount the point information in the vehicle positioning data into an electronic fence composed of multiple polygonal areas; S3: Filter the vehicle trajectory data of the longitudinal and transverse roads through the electronic fence, collide the vehicle-road cooperative data with the internal container truck data and the external container truck data respectively, and calibrate the result data generated after the collision to form the vehicle trajectory data in the yard; S4: Taking a single vehicle at a moment as a unit, the trajectory generated by the vehicle at the current moment and the result data information of the post-collision calibration are packaged and integrated, and the vehicles are sorted according to the integrated vehicle trajectory data and the preset sorting rules to complete the vehicle queuing in the yard.

2. The method for sorting and managing vehicles in a container terminal area according to claim 1, characterized in that: The step S1 comprises: S11: Obtain the internal container truck data of vehicles in the port through the positioning terminal; S12: Obtaining the external container truck data of non-port vehicles through the mobile terminal APP; S13: Acquire vehicle-road collaboration data of vehicles in the port and non-port vehicles through cameras; S14: Preprocess the data information in step S11 to step S13, the data information includes the job number, time, longitude and latitude positioning data and electronic fence information; the preprocessing includes consistent field names, calibration data sources, and converting time data into a 13-bit timestamp format.

3. The method for sorting and managing vehicles in a container terminal area according to claim 1, characterized in that: The step S2 comprises: S21: define the electronic fence as a polygonal area in a two-dimensional coordinate system, and decompose it into multiple minimum bounding rectangles, and the minimum bounding rectangles are used as leaf nodes of the R-tree; S22: When it is necessary to query whether a point is in a certain polygonal area, the R-tree is used for query, and the rectangular area that cannot contain the query point is excluded through the pruning strategy; S23: At the leaf node, a polygon inclusion test is performed using a ray method or a crossing number method to determine whether the point is within the polygon area; S24: Feedback and transmit the electronic fence information corresponding to the polygon containing the query point.

4. The method for sorting and managing vehicles in a container terminal area according to claim 2, characterized in that: The step S3 comprises: S31: collide the internal container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is marked as the matched internal container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is marked as the internal container truck data with a large distance error; S32: The external container truck data and the vehicle-road cooperative data are collided according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is marked as the matched external container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is marked as the external container truck data with a large distance error; S33: The uncollided VIR data is collided with the internal container truck data and the external container truck data according to the yard number and time. If the operation numbers are different, the distance between the internal container truck data or the external container truck data and the VIR data is less than 5 meters, and there is no vehicle within 5 meters around, the calibration is considered a successful match; for the unmatched VIR data, if the identified operation number has an entry record and no exit record, it is calibrated as recalled external container truck data.

5. A vehicle sequencing management system in a container terminal stacking area, characterized in that: include: Data processing module: used to pre-process the collected raw data, including internal container truck data, external container truck data and vehicle-road collaboration data; Establishing module: Establishing the spatial retrieval process of vehicles based on R-tree, mounting the point information in the vehicle positioning data into the electronic fence composed of multiple polygonal areas; Calibration module: used to filter the vehicle trajectory data of the longitudinal and transverse roads through the electronic fence, collide the vehicle-road cooperative data with the internal container truck data and the external container truck data respectively, and calibrate the result data generated after the collision to form the vehicle trajectory data in the yard; Sorting module: It is used to package and fuse the trajectory generated by the vehicle at the current moment and the result data information of the post-collision calibration, taking the vehicle at a single moment as a unit. According to the fused vehicle trajectory data and the preset sorting rules, the vehicles are sorted to complete the queuing of vehicles in the yard.

6. The vehicle sequencing management system in the container terminal area according to claim 5, characterized in that: The data processing module comprises: The first data acquisition unit is used to acquire the internal container truck data of the vehicle in the port through the positioning terminal; The second data acquisition unit is used to acquire the external container truck data of non-port vehicles through the mobile terminal APP; The third data acquisition unit is used to acquire the vehicle-road cooperative data of vehicles in the port and non-port vehicles through a camera; Processing unit: used to preprocess the data information in the first data acquisition unit, the second data acquisition unit and the third data acquisition unit, the data information includes the operation number, time, longitude and latitude positioning data and electronic fence information; the preprocessing includes consistent field names, calibration data sources, and converting time data into a 13-bit timestamp format.

7. The vehicle sequencing management system in the container terminal area according to claim 5, characterized in that: The establishment module includes: Index building unit: used to define the electronic fence as a polygonal area in a two-dimensional coordinate system and decompose it into multiple minimum bounding rectangles, and the minimum bounding rectangles are used as leaf nodes of the R-tree; Query unit: When you need to query whether a point is in a polygonal area, use the R-tree to query and use the pruning strategy to exclude rectangular areas that may not contain the query point; Matching unit: used to perform polygon inclusion test at leaf nodes using ray method or intersection number method to determine whether the point is within the polygon area; Result return unit: used to feedback and transmit the electronic fence information corresponding to the polygon containing the query point.

8. The vehicle sequencing management system in the container terminal area according to claim 6, characterized in that: The calibration module comprises: The first calibration unit is used to collide the internal container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is calibrated as the matched internal container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is calibrated as the internal container truck data with a large distance error; The second calibration unit is used to collide the external container truck data with the vehicle-road cooperative data according to the operation number and time. If the operation number is the same and the position error is less than 5 meters, it is calibrated as the matched external container truck data; if the operation number is the same and the position error is greater than or equal to 5 meters, it is calibrated as the external container truck data with a large distance error; The third calibration unit: used to collide the uncollided vehicle-road collaborative data with the internal container truck data and the external container truck data according to the yard number and time. If the operation numbers are different, the distance between the internal container truck data or the external container truck data and the vehicle-road collaborative data is less than 5 meters, and there is no vehicle within 5 meters around, the calibration is considered a successful match; for the unmatched vehicle-road collaborative data, if the identified operation number has an entry record and no exit record, it is calibrated as the recalled external container truck data.

Images