A port transportation management method and system and a storage medium

Abstract

The present application provides a kind of port transport management method, comprising: collecting port transport vehicle data;Structural digital road network model of port is built;Planning port transport vehicle path;Monitoring port transport abnormal event;When vehicle completes weighing, closed-loop feedback report is generated and port transport management data is synchronized;In the process of planning port transport vehicle path, according to vehicle real-time position data, dynamic road network layer, the type of goods of transport work order, destination and time limit and requirement, the optimal transport path is calculated by multi-objective optimization algorithm, and according to vehicle state and task priority, automatically assign job task, generate scheduling scheme including path guide and task instruction.The present application provides a kind of port transport management system, comprising data acquisition module, road network construction module, path planning module, abnormal monitoring module, weighing management module and data synchronization module.The present application realizes the full-process digital management of port transport operation, improves the efficiency of port transport scheduling.

Description

Technical Field

[0001] This invention relates to the field of port operation management and software technology, specifically to a port evacuation management method, system, and storage medium. Background Technology

[0002] As a multimodal transport hub, the efficiency of port operations directly impacts the overall operation of the logistics chain. Currently, port transport scheduling relies heavily on manual dispatching and experience-based route selection, resulting in high vehicle empty-running rates, delayed task response, and untimely anomaly detection. Although some ports have introduced satellite positioning and electronic map-assisted management, traditional positioning methods have limitations such as signal blind spots and accuracy fluctuations in complex port environments, and static road network models cannot dynamically reflect real-time changes such as lane occupancy and traffic control. Furthermore, the weighing process still predominantly uses paper documents, preventing real-time data integration with the scheduling system, leading to a disconnect between operational progress and vehicle status information, hindering the formation of an effective management loop. Existing technologies have not yet achieved integrated intelligent scheduling from positioning perception, route planning, process monitoring to data feedback, necessitating a port transport system that integrates multi-source positioning, dynamic route optimization, and end-to-end data connectivity.

[0003] Furthermore, the existing port operation vehicle dispatching system has shortcomings such as incomplete information closed-loop management and low efficiency, and the level of intelligence of its port transportation management process and feedback mechanism needs to be further improved. Summary of the Invention

[0004] In view of this, it is necessary to propose a port transportation management method, system and storage medium to address the shortcomings and deficiencies of existing technologies, so as to improve the management efficiency of port transportation, avoid the risks of fragmented port management information and abnormal or untimely feedback of port operations, thereby improving port operation efficiency and management level.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] This invention proposes a port evacuation management method, comprising the following steps:

[0007] Collect port transport vehicle data: Collect real-time location data of port transport vehicles and integrate BeiDou positioning and base station positioning information to generate fused positioning data;

[0008] Constructing a structured digital road network model for the port: Based on the fused positioning data and the port electronic map, construct a structured digital road network model for the port that includes lane attributes and traffic rules, and output real-time vehicle location data and dynamic road network layers;

[0009] Planning port transport vehicle routes: Based on real-time vehicle location data, dynamic road network layers, cargo type of transport work orders, destination and timeliness requirements, the optimal transport route is calculated through a multi-objective optimization algorithm, and work tasks are automatically assigned according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0010] Monitoring abnormal events in port transportation: Based on the scheduling plan and tracking the vehicle's trajectory, the vehicle's position and driving route are compared in real time based on the port's structured digital road network model. When deviation from the route, speeding, or illegal parking is detected, an alarm is triggered, and a warning record with a timestamp and violation type is generated.

[0011] Once the vehicle has completed weighing, a closed-loop feedback report is generated: Based on the warning record and the vehicle arrival signal at the weighbridge, the electronic weighbridge is automatically triggered to complete the weighing, and the weighing data is associated with the transportation work order to generate a closed-loop feedback report including task completion status, weighing information and anomaly statistics.

[0012] Synchronize port transport management data: Synchronize the closed-loop feedback report to the dispatch center database and / or mobile terminal application.

[0013] Furthermore, the step of collecting port transport vehicle data includes:

[0014] By using Beidou positioning terminals and base station communication modules installed on port transport vehicles, Beidou satellite positioning data and cellular base station positioning data of port transport vehicles are collected in real time to generate raw positioning data streams;

[0015] The original positioning data stream is spatiotemporally aligned and fused using Kalman filtering to eliminate multipath effects and positioning jumps caused by base station handover, generating fused positioning data.

[0016] Furthermore, the steps for constructing the port's structured digital road network model include:

[0017] The fused positioning data is matched with the port electronic map, and the vehicle position is projected onto the nearest lane using a map matching algorithm. Lane boundary, traffic sign and traffic rule information are extracted to generate basic road network data including lane attributes.

[0018] Based on basic road network data and real-time fused positioning data, the congestion status and traffic speed of each lane are dynamically updated, and a structured digital road network model including lane attributes and traffic rules is constructed. Finally, real-time vehicle location data and dynamic road network layers are output.

[0019] The fused positioning data is matched with the port electronic map, and the vehicle position is projected onto the nearest lane using a map matching algorithm. Lane boundary, traffic sign and traffic rule information are extracted to generate basic road network data including lane attributes.

[0020] Based on basic road network data and real-time fused positioning data, the congestion status and traffic speed of each lane are dynamically updated, and a structured digital road network model including lane attributes and traffic rules is constructed. Finally, real-time vehicle location data and dynamic road network layers are output.

[0021] Furthermore, in the step of planning port transport vehicle routes, the optimal transport route is dynamically adjusted based on real-time traffic conditions and weather conditions, and tasks are assigned to vehicles in the best condition first.

[0022] Furthermore, in the step of monitoring abnormal events in port transportation, when a vehicle is detected to have deviated from its route, the offset distance and direction are calculated and analyzed, and the offset information is pushed to the dispatcher's terminal.

[0023] In the steps of monitoring abnormal events in port transportation, an alarm is triggered when deviation from the route, speeding, or illegal parking is detected, and the abnormal event information is pushed to the mobile terminal.

[0024] Furthermore, the steps for planning port transport vehicle routes include:

[0025] The task requirement parameter set is generated by analyzing the cargo type, destination coordinates, and time requirements in the transportation work order and combining them with the vehicle's current coordinates and idle status in the real-time vehicle location data.

[0026] Based on the topology and real-time traffic rules in the dynamic road network layer, the A* algorithm is used to pre-calculate the traffic cost of each candidate path and generate a path cost matrix; the traffic cost includes travel distance, estimated time and energy consumption.

[0027] The path cost matrix and task requirement parameter set are input into a multi-objective optimization model. The Pareto optimal path set is solved by using a non-dominated sorting genetic algorithm with an elitist strategy. This comprehensively considers timeliness, economy and safety to generate the optimal path sequence.

[0028] Based on the vehicle's current load capacity, fuel range, and task execution records, and combined with task priority rules, the system automatically assigns tasks to each idle vehicle, ultimately generating a scheduling scheme that includes route guidance and task instructions.

[0029] This invention also proposes a port evacuation management method, comprising the following steps:

[0030] Collect real-time location data of transport vehicles within the port, integrate BeiDou positioning and base station positioning information, and combine with the port's electronic map to construct a structured digital road network model that includes lane attributes and traffic rules, generating real-time vehicle location data and dynamic road network layers;

[0031] It receives real-time vehicle location data and dynamic road network layers, combines the cargo type, destination and timeliness requirements of the transport work order, calculates the optimal transport route through a multi-objective optimization algorithm, and automatically assigns work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0032] It receives dispatch plans and tracks vehicle execution trajectories, compares vehicle positions with driving routes in real time based on digital road network models, and triggers alarms when it detects deviation from the route, speeding, or illegal parking. At the same time, it pushes abnormal event information to mobile terminals and generates warning records with timestamps and violation types.

[0033] Upon receiving early warning records and vehicle arrival signals, the electronic weighbridge is automatically triggered to complete the weighing process. The weighing data is then linked to the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information, and anomaly statistics. This report is simultaneously updated to the dispatch center and mobile terminals, thereby achieving full-process digital management of the transportation operation.

[0034] This invention also proposes a port evacuation management system, comprising:

[0035] The data acquisition module is used to collect port transport vehicle data; the data acquisition module collects real-time location data of port transport vehicles and integrates Beidou positioning and base station positioning information to generate fused positioning data.

[0036] The road network construction module is used to construct a structured digital road network model of the port. Based on the fused positioning data and the port electronic map, the road network construction module constructs a structured digital road network model of the port that includes lane attributes and traffic rules, and outputs real-time vehicle location data and dynamic road network layers.

[0037] The route planning module is used to plan the routes of port transport vehicles. The route planning module receives real-time vehicle location data and dynamic road network layers, combines the cargo type, destination and timeliness requirements of the transport work order, calculates the optimal transport route through a multi-objective optimization algorithm, and automatically assigns work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0038] An anomaly monitoring module is used to monitor abnormal events in port transportation. The anomaly monitoring module receives the scheduling plan and tracks the vehicle's execution trajectory. Based on the port's structured digital road network model, it compares the vehicle's position with the driving route in real time. When it detects deviation from the route, speeding, or illegal parking, it triggers an alarm and generates a warning record with a timestamp and violation type.

[0039] The weighing management module is used to generate a closed-loop feedback report after a vehicle completes weighing. The weighing management module receives the warning record and the vehicle arrival signal at the weighbridge, automatically triggers the electronic weighbridge to complete the weighing, and associates the weighing data with the transportation work order to generate a closed-loop feedback report including task completion status, weighing information and anomaly statistics.

[0040] The data synchronization module is used to synchronize port transportation management data; the data synchronization module synchronizes the closed-loop feedback report to the dispatch center database and / or mobile terminal application.

[0041] This invention proposes a port evacuation management system, comprising:

[0042] The multi-source positioning fusion and digital road source construction module is used to collect real-time location data of transport vehicles in the port, integrate Beidou positioning and base station positioning information, and combine them with the port electronic map to construct a structured digital road network model containing lane attributes and traffic rules, generating real-time vehicle location data and dynamic road network layers.

[0043] The dynamic route planning and intelligent task assignment module is used to receive real-time vehicle location data and dynamic road network layers, combine the cargo type, destination and timeliness requirements of the transport work order, calculate the optimal transport route through a multi-objective optimization algorithm, and automatically assign work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0044] The operation process monitoring and anomaly early warning module is used to receive dispatch plans and track vehicle execution trajectories. Based on the digital road network model, it compares the vehicle position and driving route in real time. When it detects deviation from the route, speeding, or illegal parking, it triggers an alarm and pushes abnormal event information to the mobile terminal, generating an early warning record with timestamp and violation type.

[0045] The paperless weighing and data closed-loop feedback module is used to receive early warning records and vehicle arrival signals, automatically trigger the electronic weighbridge to complete the weighing, and associate the weighing data with the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information and anomaly statistics, which is synchronously updated to the dispatch center and mobile terminals, thereby realizing full-process digital management of transportation operations.

[0046] The present invention further proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the port transport management method as described in any of the preceding claims.

[0047] The beneficial effects of this invention are as follows:

[0048] This invention improves the efficiency of logistics operation management within the port area, significantly enhances port operation efficiency, reduces human error, and lowers safety risks. Furthermore, this invention realizes digital management of the entire port evacuation operation process, improving port evacuation scheduling efficiency and safety control level. Attached Figure Description

[0049] Figure 1 This is a schematic flowchart of a port evacuation management method provided in Embodiment 1 of the present invention;

[0050] Figure 2 This is a schematic diagram of the structure of a port evacuation management system provided in Embodiment 2 of the present invention;

[0051] Figure 3 This is a schematic diagram of the structure of a port evacuation management system provided in Embodiment 6 of the present invention;

[0052] Figure 4 This is a schematic diagram illustrating the implementation process of a dynamic path planning and intelligent task assignment module in a port evacuation management system provided in Embodiment 6 of the present invention.

[0053] Figure 5 This is a schematic diagram illustrating the implementation process of the operation process monitoring and anomaly early warning module of a port evacuation management system provided in Embodiment 6 of the present invention;

[0054] Figure 6 This is a schematic flowchart of a port evacuation management method provided in Embodiment 6 of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be further described clearly and completely below in conjunction with the embodiments of this invention. It should be noted that the described embodiments are merely some embodiments of this invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0056] As used in this specification and the following claims, the words “a,” “an,” and “the” have the meaning of plural references unless the context clearly indicates otherwise.

[0057] The following is a detailed description of embodiments of the invention depicted in the accompanying drawings. The embodiments are detailed in order to clearly convey the invention. However, the amount of detail provided is not intended to limit the contemplative variations of the embodiments; rather, it is intended to cover all modifications, equivalents, and substitutions falling within the spirit and scope of the invention as defined by the appended claims.

[0058] Example 1

[0059] like Figure 1 As shown:

[0060] This embodiment proposes a port evacuation management method, including the following steps:

[0061] S1, Collect port transport vehicle data: Collect real-time location data of port transport vehicles and integrate Beidou positioning and base station positioning information to generate fused positioning data;

[0062] S2, Construct a structured digital road network model for the port: Based on the fused positioning data and the port electronic map, construct a structured digital road network model for the port that includes lane attributes and traffic rules, and output real-time vehicle location data and dynamic road network layers;

[0063] S3, Plan Port Transport Vehicle Routes: Based on real-time vehicle location data, dynamic road network layers, cargo type of transport work orders, destination and timeliness requirements, calculate the optimal transport route through a multi-objective optimization algorithm, and automatically assign work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0064] S4, Monitor abnormal events in port transportation: Based on the scheduling plan and track the vehicle's trajectory, compare the vehicle's position and driving route in real time based on the port's structured digital road network model. When deviation from the route, speeding, or illegal parking is detected, trigger an alarm and generate a warning record with a timestamp and violation type.

[0065] S5. After the vehicle completes the weighing, a closed-loop feedback report is generated: Based on the warning record and the vehicle arrival signal at the weighbridge, the electronic weighbridge is automatically triggered to complete the weighing, and the weighing data is associated with the transportation work order to generate a closed-loop feedback report including the task completion status, weighing information and anomaly statistics.

[0066] S6, Synchronize port transportation management data: Synchronize the closed-loop feedback report to the dispatch center database and / or mobile terminal application.

[0067] In some embodiments, S1 is optimized to include:

[0068] S11, through the Beidou positioning terminal and base station communication module installed on the port transport vehicle, collect the Beidou satellite positioning data and cellular base station positioning data of the port transport vehicle in real time, and generate the raw positioning data stream; then execute S12;

[0069] S12 performs spatiotemporal alignment and Kalman filtering fusion processing on the original positioning data stream to eliminate multipath effects and positioning jumps caused by base station handover, generating fused positioning data.

[0070] Further optimized, S2 includes:

[0071] S21, the fused positioning data generated in S12 is matched with the port electronic map, the vehicle position is projected onto the nearest lane using the map matching algorithm, and lane boundary, traffic sign and traffic rule information is extracted to generate basic road network data including lane attributes; then S22 is executed;

[0072] S22, based on basic road network data and real-time fused positioning data, dynamically updates the congestion status and traffic speed of each lane, constructs a structured digital road network model including lane attributes and traffic rules, and finally outputs real-time vehicle location data and dynamic road network layers.

[0073] In some embodiments, in S3, the optimal transportation route is dynamically adjusted based on real-time traffic conditions and weather conditions, and tasks are assigned to the vehicle in the best condition first.

[0074] In some embodiments, in S4, when a vehicle is detected to have deviated from the route, the offset distance and direction are calculated and analyzed, and the offset information is pushed to the dispatcher terminal.

[0075] In some embodiments, an alarm is triggered in S4 when deviation from the route, speeding, or illegal parking is detected, and the abnormal event information is pushed to the mobile terminal.

[0076] In some embodiments, in S5, after weighing is completed, an electronic weighbridge slip is generated and the electronic weighbridge is associated with and stored in the transport work order.

[0077] In some embodiments, in S6, when it is necessary to synchronize the closed-loop feedback report to the scheduling center database and the mobile terminal application through a standardized data interface, a two-way data verification mechanism is adopted to ensure the data consistency between the scheduling center database and the mobile terminal.

[0078] Further optimization of any of the above implementation schemes, wherein S3 includes:

[0079] S31: Parse the cargo type, destination coordinates, and timeliness requirements in the transportation work order, and combine them with the vehicle's current coordinates and idle status in the real-time vehicle location data to generate a set of task requirement parameters; then execute S32.

[0080] S32, based on the topology and real-time traffic rules in the dynamic road network layer, the A* algorithm is used to pre-calculate the traffic cost of each candidate path and generate a path cost matrix; the traffic cost includes travel distance, estimated time and energy consumption; then S33 is executed;

[0081] S33: Input the path cost matrix and task requirement parameter set into the multi-objective optimization model, and use the non-dominated sorting genetic algorithm with elitist strategy to solve the Pareto optimal path set, thereby comprehensively considering timeliness, economy and safety to generate the optimal path sequence; then execute S34.

[0082] S34 automatically assigns work tasks to each idle vehicle based on the vehicle's current load capacity, fuel range, and task execution records, combined with task priority rules, and finally generates a scheduling scheme that includes route guidance and task instructions.

[0083] Example 2

[0084] like Figure 2 As shown:

[0085] This invention also proposes a port evacuation management system, comprising:

[0086] The data acquisition module is used to collect port transport vehicle data; the data acquisition module collects real-time location data of port transport vehicles and integrates Beidou positioning and base station positioning information to generate fused positioning data.

[0087] The road network construction module is used to construct a structured digital road network model of the port. Based on the fused positioning data and the port electronic map, the road network construction module constructs a structured digital road network model of the port that includes lane attributes and traffic rules, and outputs real-time vehicle location data and dynamic road network layers.

[0088] The route planning module is used to plan the routes of port transport vehicles. The route planning module receives real-time vehicle location data and dynamic road network layers, combines the cargo type, destination and timeliness requirements of the transport work order, calculates the optimal transport route through a multi-objective optimization algorithm, and automatically assigns work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0089] An anomaly monitoring module is used to monitor abnormal events in port transportation. The anomaly monitoring module receives the scheduling plan and tracks the vehicle's execution trajectory. Based on the port's structured digital road network model, it compares the vehicle's position with the driving route in real time. When it detects deviation from the route, speeding, or illegal parking, it triggers an alarm and generates a warning record with a timestamp and violation type.

[0090] The weighing management module is used to generate a closed-loop feedback report after a vehicle completes weighing. The weighing management module receives the warning record and the vehicle arrival signal at the weighbridge, automatically triggers the electronic weighbridge to complete the weighing, and associates the weighing data with the transportation work order to generate a closed-loop feedback report including task completion status, weighing information and anomaly statistics.

[0091] The data synchronization module is used to synchronize port transportation management data; the data synchronization module synchronizes the closed-loop feedback report to the dispatch center database and / or mobile terminal application.

[0092] In some optimized embodiments, the data acquisition module includes:

[0093] The raw positioning data stream generation submodule communicates with the Beidou positioning terminal and the base station communication module installed on the port transport vehicle, respectively, to collect Beidou satellite positioning data and cellular base station positioning data of the port transport vehicle in real time and generate the raw positioning data stream;

[0094] The fused positioning data generation submodule is used to perform spatiotemporal alignment and Kalman filtering fusion processing on the original positioning data stream to eliminate positioning jumps caused by multipath effects and base station handover, and generate fused positioning data.

[0095] In some optimized embodiments, the road network construction module includes:

[0096] The basic road network data generation submodule is used to match the fused positioning data with the port electronic map, use a map matching algorithm to project the vehicle position onto the nearest lane, and extract lane boundary, traffic sign and traffic rule information to generate basic road network data including lane attributes.

[0097] The vehicle positioning and road network layer output submodule is used to dynamically update the congestion status and traffic speed of each lane based on basic road network data and real-time fused positioning data, construct a structured digital road network model including lane attributes and traffic rules, and output real-time vehicle location data and dynamic road network layers.

[0098] In some embodiments, the route planning module also dynamically adjusts the optimal transportation route based on real-time traffic conditions and weather conditions, and prioritizes assigning tasks to vehicles in the best condition.

[0099] In some embodiments, the anomaly monitoring module, when it detects a vehicle deviating from its route, calculates and analyzes the offset distance and direction, and pushes the offset information to the dispatcher terminal.

[0100] In some embodiments, the weighing management module generates an electronic weighbridge slip after weighing is completed and stores the electronic weighbridge in association with the transportation work order.

[0101] In some embodiments, when the closed-loop feedback report needs to be synchronized to the dispatch center database and mobile terminal application through a standardized data interface, the data synchronization module adopts a two-way data verification mechanism to ensure data consistency between the dispatch center database and the mobile terminal.

[0102] Further optimization of any of the above implementation schemes, the path planning module includes:

[0103] The task requirement parameter set generation submodule parses the cargo type, destination coordinates, and timeliness requirements in the transportation work order, and combines the vehicle's current coordinates and idle status in the real-time vehicle location data to generate the task requirement parameter set;

[0104] The path cost matrix generation submodule, based on the topology and real-time traffic rules in the dynamic road network layer, uses the A* algorithm to pre-calculate the traffic cost of each candidate path and generate a path cost matrix; the traffic cost includes travel distance, estimated time and energy consumption;

[0105] The optimal path sequence generation submodule inputs the path cost matrix and task requirement parameter set into the multi-objective optimization model, and uses a non-dominated sorting genetic algorithm with elitist strategy to solve for the Pareto optimal path set, thereby comprehensively considering timeliness, economy and safety to generate the optimal path sequence.

[0106] The scheduling scheme generation submodule automatically assigns work tasks to each idle vehicle based on the current status of port transport vehicles, including load capacity, fuel level, and task execution records, combined with task priority rules, and finally generates a scheduling scheme that includes route guidance and task instructions.

[0107] In a further optimized manner, each module of the port evacuation management system in Embodiment 2 executes the corresponding steps of the port evacuation management method in Embodiment 1.

[0108] Example 3

[0109] Example 3 is an optimized design of the corresponding technical solution in Example 1 or Example 2;

[0110] In some specific embodiments, in the path planning module of S3 in Embodiment 1 / Embodiment 2, the real-time vehicle location data obtained by the data acquisition module of S1 in Embodiment 1 / Embodiment 2 and the dynamic road network layer obtained by the road network construction module of S2 in Embodiment 1 / Embodiment 2 are combined with the cargo type, destination and timeliness requirements of the transport work order, and the optimal transport route is calculated by a multi-objective optimization algorithm. The work tasks are automatically assigned according to the vehicle status and task priority, and a scheduling scheme including path guidance and task instructions is generated.

[0111] In some embodiments, specifically in S4 of Embodiment 1 / the anomaly monitoring module of Embodiment 1, the closed-loop feedback report is synchronized to the scheduling center database and / or mobile terminal application through a standardized data interface.

[0112] In some embodiments, S4 of embodiment 1 specifically includes:

[0113] S41 receives route guidance and task instructions from the scheduling scheme through the vehicle terminal, and transmits real-time vehicle location data back at a fixed frequency to generate a sequence of actual vehicle driving trajectory points.

[0114] S42, performs real-time spatial matching between the actual driving trajectory point sequence and the planned path in the port structured digital road network model, calculates the lateral offset and longitudinal progress deviation, and generates the trajectory deviation index.

[0115] S43, compares the real-time speed of the vehicle with the speed limit rules of the road segment in the dynamic road network layer to detect speeding behavior, and detects illegal parking based on the dwell time and the parking area rules in the road network, and generates a list of violation events;

[0116] S44: When the trajectory deviation exceeds the preset threshold or the list of violations is not empty, an alarm is immediately triggered and a warning record with a timestamp and violation type is pushed to the driver's mobile terminal. At the same time, it is stored in the abnormal event database and a warning record is generated.

[0117] In some embodiments, the anomaly monitoring module of embodiment 2 specifically includes:

[0118] The vehicle actual driving trajectory point sequence generation submodule is used to receive path guidance and task instructions in the scheduling scheme through the vehicle terminal, and transmit real-time vehicle location data back at a fixed frequency to generate the vehicle actual driving trajectory point sequence.

[0119] The trajectory deviation index generation submodule is used to perform real-time spatial matching between the actual driving trajectory point sequence and the planned path in the port structured digital road network model, calculate the lateral offset and longitudinal progress deviation, and generate the trajectory deviation index.

[0120] The violation event list generation submodule is used to detect speeding behavior by comparing the real-time speed of the vehicle with the speed limit rules of the road segment in the dynamic road network layer, and to detect illegal parking by comparing the dwell time with the parking area rules in the road network, and to generate a violation event list.

[0121] The warning record generation submodule is used to immediately trigger an alarm and push a warning record with a timestamp and violation type to the driver's mobile terminal when the trajectory deviation exceeds a preset threshold or the list of violations is not empty. At the same time, it is stored in the abnormal event database to generate a warning record.

[0122] In some embodiments, S5 of embodiment 1 specifically includes:

[0123] S51 receives the latest warning information from the warning record database and receives the arrival signal triggered by the infrared sensor when the vehicle arrives at the weighbridge. It automatically starts the electronic weighbridge to perform weighing operations, obtains the vehicle's gross weight and tare weight data, and generates a weighing record.

[0124] S52 associates the weighing record with the current transportation work order through the work order number, updates the work order status to completed, and attaches the weighing time and weight data to generate a task completion record including weighing information.

[0125] S53, summarize all warning records, weighing data and task start and end times of the vehicle during this task, use statistical analysis methods to calculate the number of anomalies and work efficiency, and generate an anomaly statistics report;

[0126] S54 integrates task completion status, weighing information, and anomaly statistics reports to generate a complete closed-loop feedback report. It is then synchronized to the dispatch center database and mobile terminal application via a data interface, thereby achieving digital closed-loop management of the entire transportation operation process.

[0127] In some embodiments, the weighing management module of embodiment 2 specifically includes:

[0128] The weighing record generation submodule is used to receive the latest warning information from the warning record database and receive the arrival signal triggered by the infrared sensor when the vehicle arrives at the weighbridge. It automatically starts the electronic weighbridge to perform weighing operations, obtains the vehicle's gross weight and tare weight data, and generates a weighing record.

[0129] The task completion record generation submodule is used to associate the weighing record with the current transportation work order through the work order number, update the work order status to completed, and attach the weighing time and weight data to generate a task completion record including weighing information.

[0130] The anomaly statistics report generation submodule is used to summarize all warning records, weighing data and task start and end times of the vehicle during this task, calculate the number of anomalies and work efficiency using statistical analysis methods, and generate anomaly statistics reports.

[0131] The closed-loop feedback report generation and data synchronization submodule is used to integrate task completion status, weighing information, and anomaly statistics reports to generate a complete closed-loop feedback report. It is then synchronized and updated to the dispatch center database and mobile terminal application through a data interface, thereby realizing digital closed-loop management of the entire transportation operation process.

[0132] Example 4

[0133] This embodiment proposes a port evacuation method, the method comprising:

[0134] Collect real-time location data of transport vehicles within the port, integrate BeiDou positioning and base station positioning information, and combine with the port's electronic map to construct a structured digital road network model that includes lane attributes and traffic rules, generating real-time vehicle location data and dynamic road network layers;

[0135] It receives real-time vehicle location data and dynamic road network layers, combines the cargo type, destination and timeliness requirements of the transport work order, calculates the optimal transport route through a multi-objective optimization algorithm, and automatically assigns work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0136] It receives dispatch plans and tracks vehicle execution trajectories, compares vehicle positions with driving routes in real time based on digital road network models, and triggers alarms when it detects deviation from the route, speeding, or illegal parking. At the same time, it pushes abnormal event information to mobile terminals and generates warning records with timestamps and violation types.

[0137] Upon receiving early warning records and vehicle arrival signals, the electronic weighbridge is automatically triggered to complete the weighing process. The weighing data is then linked to the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information, and anomaly statistics. This report is simultaneously updated to the dispatch center and mobile terminals, thereby achieving full-process digital management of the transportation operation.

[0138] Further optimized, the real-time location data of transport vehicles within the port is collected, integrated with BeiDou positioning and base station positioning information, and combined with the port's electronic map to construct a structured digital road network model including lane attributes and traffic rules, generating real-time vehicle location data and a dynamic road network layer, including:

[0139] By using the BeiDou positioning terminal and base station communication module installed on the transport vehicle, the BeiDou satellite positioning data and cellular base station positioning data of the vehicle are collected in real time to generate the raw positioning data stream;

[0140] The original positioning data stream is spatiotemporally aligned and fused using Kalman filtering to eliminate multipath effects and positioning jumps caused by base station handover, generating fused positioning data.

[0141] The fused positioning data is matched with the port's electronic map, and the vehicle's position is projected onto the nearest lane using a map matching algorithm. Lane boundaries, traffic signs, and traffic rules information are extracted to generate basic road network data containing lane attributes.

[0142] Based on basic road network data and real-time fused positioning data, the congestion status and traffic speed of each lane are dynamically updated, and a structured digital road network model containing lane attributes and traffic rules is constructed. Finally, real-time vehicle location data and dynamic road network layers are output.

[0143] Further optimized, the received real-time vehicle location data and dynamic road network layer, combined with the cargo type, destination, and timeliness requirements of the transport work order, calculate the optimal transport route using a multi-objective optimization algorithm, and automatically assign work tasks based on vehicle status and task priority, generating a scheduling scheme including route guidance and task instructions, including:

[0144] The task requirement parameter set is generated by analyzing the cargo type, destination coordinates, and time requirements in the transportation work order and combining them with the vehicle's current coordinates and idle status in the real-time vehicle location data.

[0145] Based on the topology and real-time traffic rules in the dynamic road network layer, the A* algorithm is used to pre-calculate the traffic cost of each candidate path and generate a path cost matrix; the traffic cost includes travel distance, estimated time and energy consumption.

[0146] The path cost matrix and task requirement parameter set are input into a multi-objective optimization model. The Pareto optimal path set is solved using a non-dominated sorting genetic algorithm with an elitist strategy. Taking into account timeliness, economy and safety, the optimal path sequence is generated.

[0147] Based on the vehicle's current load capacity, fuel range, and task execution records, and combined with task priority rules, the system automatically assigns tasks to each idle vehicle, ultimately generating a scheduling scheme that includes route guidance and task instructions.

[0148] Example 5

[0149] This embodiment proposes a port transportation system, which includes: a multi-source positioning fusion and digital road network construction module, a dynamic route planning and intelligent task assignment module, an operation process monitoring and anomaly early warning module, and a paperless weighing and data closed-loop feedback module.

[0150] The multi-source positioning fusion and digital road source construction module is used to collect real-time location data of transport vehicles in the port, fuse Beidou positioning and base station positioning information, and combine them with the port electronic map to construct a structured digital road network model containing lane attributes and traffic rules, and generate real-time vehicle location data and dynamic road network layers.

[0151] The dynamic route planning and intelligent task assignment module is used to receive real-time vehicle location data and dynamic road network layers, combine the cargo type, destination and timeliness requirements of the transport work order, calculate the optimal transport route through a multi-objective optimization algorithm, and automatically assign work tasks according to vehicle status and task priority, generating a scheduling scheme including route guidance and task instructions.

[0152] The operation process monitoring and anomaly early warning module is used to receive the dispatch plan and track the vehicle execution trajectory. Based on the digital road network model, it compares the vehicle position and driving route in real time. When it detects deviation from the route, speeding or illegal parking, it triggers an alarm and pushes the abnormal event information to the mobile terminal, generating an early warning record with a timestamp and violation type.

[0153] The paperless weighing and data closed-loop feedback module is used to receive early warning records and vehicle arrival signals, automatically trigger the electronic weighbridge to complete the weighing, and associate the weighing data with the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information and anomaly statistics, which is synchronously updated to the dispatch center and mobile terminal, thereby realizing full-process digital management of transportation operations.

[0154] Further optimized, the real-time location data of transport vehicles within the port is collected, integrated with BeiDou positioning and base station positioning information, and combined with the port's electronic map to construct a structured digital road network model including lane attributes and traffic rules, generating real-time vehicle location data and a dynamic road network layer, including:

[0155] By using the BeiDou positioning terminal and base station communication module installed on the transport vehicle, the BeiDou satellite positioning data and cellular base station positioning data of the vehicle are collected in real time to generate the raw positioning data stream;

[0156] The original positioning data stream is spatiotemporally aligned and fused using Kalman filtering to eliminate multipath effects and positioning jumps caused by base station handover, generating fused positioning data.

[0157] The fused positioning data is matched with the port's electronic map, and the vehicle's position is projected onto the nearest lane using a map matching algorithm. Lane boundaries, traffic signs, and traffic rules information are extracted to generate basic road network data containing lane attributes.

[0158] Based on basic road network data and real-time fused positioning data, the congestion status and traffic speed of each lane are dynamically updated, and a structured digital road network model containing lane attributes and traffic rules is constructed. Finally, real-time vehicle location data and dynamic road network layers are output.

[0159] Further optimized, the received real-time vehicle location data and dynamic road network layer, combined with the cargo type, destination, and timeliness requirements of the transport work order, calculate the optimal transport route using a multi-objective optimization algorithm, and automatically assign work tasks based on vehicle status and task priority, generating a scheduling scheme including route guidance and task instructions, including:

[0160] The task requirement parameter set is generated by analyzing the cargo type, destination coordinates, and time requirements in the transportation work order and combining them with the vehicle's current coordinates and idle status in the real-time vehicle location data.

[0161] Based on the topology and real-time traffic rules in the dynamic road network layer, the A* algorithm is used to pre-calculate the traffic cost of each candidate path and generate a path cost matrix; the traffic cost includes travel distance, estimated time and energy consumption.

[0162] The path cost matrix and task requirement parameter set are input into a multi-objective optimization model. The Pareto optimal path set is solved using a non-dominated sorting genetic algorithm with an elitist strategy. Taking into account timeliness, economy and safety, the optimal path sequence is generated.

[0163] Based on the vehicle's current load capacity, fuel range, and task execution records, and combined with task priority rules, the system automatically assigns tasks to each idle vehicle, ultimately generating a scheduling scheme that includes route guidance and task instructions.

[0164] Further optimized, the system receives the scheduling plan and tracks the vehicle's trajectory, compares the vehicle's position with the driving route in real time based on the digital road network model, and triggers an alarm when deviation from the route, speeding, or illegal parking is detected. Simultaneously, abnormal event information is pushed to the mobile terminal, generating a warning record with a timestamp and violation type, including:

[0165] The vehicle-mounted terminal receives route guidance and task instructions from the scheduling plan and transmits real-time vehicle location data back at a fixed frequency to generate a sequence of actual vehicle driving trajectory points.

[0166] The actual driving trajectory point sequence is spatially matched with the planned path in the structured digital road network model in real time, and the lateral offset and longitudinal progress deviation are calculated to generate the trajectory deviation index.

[0167] Speeding behavior is detected by comparing the vehicle's real-time speed with the speed limit rules of road segments in the dynamic road network layer, and illegal parking is detected by comparing the dwell time with the stop area rules in the road network, generating a list of violation events;

[0168] When the trajectory deviation exceeds the preset threshold or the list of violations is not empty, an alarm is immediately triggered and a warning record with a timestamp and violation type is pushed to the driver's mobile terminal. At the same time, it is stored in the abnormal event database and a warning record is generated.

[0169] In a further optimized manner, the receipt of early warning records and vehicle arrival signals automatically triggers the electronic weighbridge to complete the weighing process. The weighing data is then linked to the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information, and anomaly statistics. This report is simultaneously updated to the dispatch center and mobile terminals, thereby achieving full-process digital management of the transportation operation. This step includes the following sub-steps:

[0170] It receives the latest warning information from the warning record database and the arrival signal triggered by the infrared sensor when the vehicle arrives at the weighbridge. It automatically starts the electronic weighbridge to perform the weighing operation, obtains the vehicle's gross weight and tare weight data, and generates a weighing record.

[0171] Associate the weighing record with the current transportation work order using the work order number, update the work order status to "completed", and attach the weighing time and weight data to generate a task completion record containing weighing information.

[0172] Summarize all warning records, weighing data and task start and end times of the vehicle during this task, use statistical analysis methods to calculate the number of anomalies and work efficiency, and generate an anomaly statistics report.

[0173] By integrating task completion status, weighing information, and anomaly statistics reports, a complete closed-loop feedback report is generated and synchronously updated to the dispatch center database and mobile terminal application through a data interface, realizing digital closed-loop management of the entire transportation operation process.

[0174] Example 6

[0175] Example 6 is an optimized design of the corresponding technical solutions in Examples 1-5;

[0176] like Figures 3-5 As shown, an embodiment of the present invention provides a port transportation management system, including a multi-source positioning fusion and digital route source construction module 101, a dynamic route planning and intelligent task assignment module 102, an operation process monitoring and anomaly early warning module 103, and a paperless weighing and data closed-loop feedback module 104.

[0177] The multi-source positioning fusion and digital road source construction module 101 (in some preferred embodiments, the multi-source positioning fusion and digital road source construction module 101 of this embodiment may include the data acquisition module and road network construction module as described in Embodiment 2) is used to collect real-time location data of port transport vehicles, fuse Beidou positioning (satellite positioning) and base station positioning information, and combine the port electronic map to construct a structured digital road network model including lane attributes and traffic rules, and generate real-time vehicle location data and dynamic road network layers;

[0178] Specifically, the modules corresponding to the steps / systems of the method of the present invention can collect BeiDou satellite positioning data and cellular base station positioning data of the vehicle in real time through the BeiDou positioning terminal and base station communication module installed on the transport vehicle, and generate raw positioning data stream;

[0179] Firstly, one of the core execution steps of the multi-source positioning fusion and digital road source construction module 101 is to rely on the vehicle-mounted hardware terminal to complete the real-time acquisition of multi-source positioning signals, convert the vehicle's physical location into a digital positioning data stream, and provide raw input data for subsequent positioning fusion processing. The specific implementation method is as follows:

[0180] Port transport vehicles are the core carriers of port evacuation operations. Each vehicle is equipped with standardized positioning and data acquisition hardware. The Beidou positioning terminal is responsible for receiving satellite navigation signals, and the base station communication module is responsible for accessing the cellular network to obtain base station positioning information. The two types of hardware work together to form a dual-source positioning and data acquisition system. The hardware deployment follows the protection standards of the port operation environment and is adapted to the complex working conditions such as obstruction, vibration, and humidity in the port area. The positioning data acquisition frequency is set to 10Hz, that is, 10 times of positioning data are collected per second to ensure the real-time and continuous location information and meet the millisecond-level response requirements of dynamic vehicle scheduling in the port area.

[0181] The BeiDou positioning terminal receives radio frequency signals transmitted by BeiDou navigation satellites, calculates the vehicle's spatial position information, and generates BeiDou satellite positioning data. This data includes four core parameters: the vehicle's three-dimensional coordinates, positioning time, number of satellites, and positioning accuracy factor. The three-dimensional coordinates use the WGS-84 geodetic coordinate system, consisting of longitude, latitude, and elevation. The longitude ranges from 73°E to 135°E, the latitude ranges from 18°N to 53°N, and the elevation is based on mean sea level in meters. The positioning time uses UTC (Coordinated Universal Time) with millisecond accuracy. The number of satellites indicates the number of BeiDou satellites successfully locked by the terminal; a larger value indicates stronger positioning stability. During normal operations in the port area, the number of satellites is maintained between 8 and 12. The positioning accuracy factor DOP is used to characterize the positioning error level. A horizontal positioning accuracy factor HDOP value less than 2 indicates excellent horizontal positioning accuracy, and a vertical positioning accuracy factor VDOP value less than 3 indicates reliable elevation positioning accuracy.

[0182] The base station communication module relies on the cellular communication network covering the port area to establish data interaction with surrounding communication base stations. By measuring parameters such as signal propagation time and signal strength between the vehicle and multiple base stations, it calculates cellular base station positioning data. This data includes four types of parameters: base station number, signal arrival time difference, signal received power, and estimated positioning coordinates. The base station number is a unique identifier for each communication base station in the port area, used to distinguish the signal sources of different base stations. The signal arrival time difference (TDOA) is the difference in signal propagation time from the vehicle to different base stations, measured in nanoseconds, and is the core basis for calculating base station positioning. The signal received power (RSSI) is measured in decibels and milliwatts, with a value range of -120dBm to -50dBm. A larger value indicates a higher signal strength. The estimated positioning coordinates are the vehicle's latitude and longitude coordinates initially calculated by the base station algorithm. The accuracy is lower than that of BeiDou positioning but is not limited by satellite obstruction.

[0183] The two types of positioning data are collected synchronously according to a unified time reference. The communication module between the Beidou positioning terminal and the base station has a built-in high-precision temperature-compensated crystal oscillator, and the time synchronization error is controlled within 1 millisecond. This ensures that the two types of data collected at the same time correspond to the same physical location of the vehicle. The collected Beidou satellite positioning data and cellular base station positioning data are packaged in a fixed data format, and the data type, collection time, and core parameter values ​​are labeled in sequence. They are continuously transmitted to form a raw positioning data stream. The data stream adopts a streaming transmission mode with no data loss or breakpoints. All data are continuously spliced ​​to form a complete raw positioning data stream, providing standardized input for subsequent fusion processing.

[0184] The original positioning data stream is spatiotemporally aligned and fused using Kalman filtering to eliminate multipath effects and positioning jumps caused by base station handover, generating fused positioning data.

[0185] Secondly, one of the core execution steps of the multi-source positioning fusion and digital path source construction module 101 is to perform unified temporal and spatial calibration on the original dual-source positioning data, suppress positioning errors and signal jumps through the Kalman filter algorithm, integrate the advantages of the two types of positioning data, and generate high-precision fused positioning data. The specific implementation method is as follows:

[0186] In the original positioning data stream, there are slight time and spatial coordinate deviations between BeiDou positioning and base station positioning data. Spatiotemporal alignment is the core operation to eliminate these deviations, and it consists of two parts: time alignment and spatial alignment. Time alignment uses the UTC time of BeiDou positioning as the reference time and interpolates and calibrates the acquisition time of the base station positioning data to the reference time. The interpolation algorithm adopts the linear interpolation method. For base station positioning data with a time deviation of more than 5 milliseconds, the time and coordinates of two adjacent sets of data are linearly fitted to generate the base station positioning coordinates corresponding to the reference time, ensuring that the two types of data correspond to the same location under the same timestamp. Spatial alignment transforms the coordinate system of the two types of positioning data into the local plane rectangular coordinate system of the port. This coordinate system has the port center as the origin, the X-axis points due east, the Y-axis points due north, and the unit is meters. The transformation parameters are calculated through the coordinate transformation model measured in the port area to eliminate the difference between the geodetic coordinate system and the local coordinate system. After completing the spatiotemporal alignment, the two types of data have the basis for fusion.

[0187] Multipath effect refers to the positioning error caused by the reflection of BeiDou signals from containers, cranes, and buildings in the port area, leading to random shifts in positioning coordinates. Base station switching is the positioning jump caused when a vehicle connects to different base stations during its journey, manifesting as an instantaneous shift of coordinates of several meters to tens of meters. Kalman filter fusion processing is the core algorithm for solving these problems. This algorithm consists of two stages: prediction and update. By establishing the vehicle motion state equation and observation equation, it achieves the optimal estimation of positioning data. The vehicle motion state equation uses the vehicle's position and velocity as state variables, and sets the state vector X_k=[x [_k, y_k, v_xk, v_yk], where x_k and y_k are the vehicle's planar coordinates at time k, and v_xk and v_yk are the vehicle's velocities in the X and Y directions at time k. The state transition matrix adopts a uniform motion model. The observation equation uses the spatiotemporally aligned BeiDou and base station positioning coordinates as the observation values. The observation vector Z_k = [x_observation_k, y_observation_k]. The observation noise matrix is ​​set according to the accuracy of the two types of positioning data. The BeiDou positioning observation noise R_1 is appropriately selected, and the base station positioning observation noise R_2 is appropriately selected to reflect the higher accuracy of BeiDou positioning.

[0188] In the prediction phase of Kalman filtering, the state at time k-1 is estimated based on the optimal state at time k-1, and the prior state estimate and prior error covariance matrix at time k are calculated. In the update phase, the Kalman gain K_k is calculated by combining the observation value at time k, and the prior state estimate is corrected to obtain the posterior state estimate at time k, i.e., the optimal positioning coordinates. The algorithm iteration frequency is consistent with the data acquisition frequency of 10Hz, and each set of positioning data is processed in real time. This can effectively suppress the random error caused by multipath effects, control the positioning offset within 0.3 meters, and smooth the coordinate jump caused by base station switching, thereby avoiding sudden changes in positioning data. After the filtering and fusion processing is completed, the optimal coordinates, positioning time, and fusion accuracy parameters are integrated to generate fused positioning data. The fusion accuracy parameter is the horizontal positioning error after fusion, in meters, with a value range of 0.1 meters to 0.5 meters, representing the reliability of the fused positioning data. This data eliminates the errors and jumps of the original data and has the characteristics of high accuracy and high stability, providing core positioning basis for subsequent map matching and road network construction.

[0189] The fused positioning data is matched with the port electronic map for coordinate matching. The map matching algorithm is used to project the vehicle position onto the nearest lane and extract lane boundary, traffic sign and traffic rule information to generate basic road network data containing lane attributes.

[0190] Thirdly, one of the core execution steps of the multi-source positioning fusion and digital road source construction module 101 is to associate high-precision fused positioning data with the port electronic map, achieve precise binding of vehicle position and lane through map matching algorithm, extract basic road network information, and construct basic road network data containing lane attributes. The specific implementation method is as follows:

[0191] The port electronic map is a digital carrier of port area geographic information, covering the spatial information of all roads, lanes, storage yards, weighbridges, and operating areas within the port area. The map accuracy is 0.1 meters, and it is stored in vector data format. It contains a three-layer structure: spatial coordinate layer, attribute information layer, and rule information layer. The spatial coordinate layer records the geometric coordinates of all road network elements, the attribute information layer marks the lane number, lane width, and lane type, and the rule information layer stores traffic signs, traffic restrictions, driving directions, and other rules. The map data is pre-surveyed and calibrated in the port area to ensure complete consistency with the physical road network. It is the base map for coordinate matching and map matching.

[0192] The coordinate matching process involves matching the planar coordinates of the fused positioning data with the spatial coordinates of the port electronic map. The matching threshold is set to 5 meters. When the distance between the fused positioning coordinates and the map road network coordinates is less than 5 meters, it is considered a valid match. If the distance exceeds the threshold, it is considered a positioning anomaly, triggering the data verification process. During the coordinate matching process, abnormal coordinates that are outside the port area map range are automatically filtered to ensure that the vehicle position is always within the port area road network range and to complete the valid coordinate matching.

[0193] The map matching algorithm uses a weighted probabilistic matching algorithm. Its core is to calculate the matching probability between the fused positioning coordinates and the surrounding lanes, and project the vehicle position onto the lane with the highest probability. The input of the algorithm is the fused positioning coordinates, the geometric center line coordinates of the surrounding lanes, and the lane width. The output of the algorithm is the target lane where the vehicle is located and the projected coordinates. The matching probability calculation considers three dimensions: distance weight, direction weight, and historical trajectory weight. The distance weight is the distance from the vehicle coordinates to the lane center line. The smaller the distance, the higher the weight. The direction weight is the angle between the vehicle's driving direction and the lane's driving direction. The smaller the angle, the higher the weight. The historical trajectory weight is the lane where the vehicle was in at the previous moment. The weight ratio increases when the vehicle is driving continuously. The total matching probability is obtained by weighted summation. The lane with the highest total matching probability is the lane where the vehicle is currently driving. The algorithm finally projects the vehicle position accurately onto the center line of the corresponding lane.

[0194] After map matching is completed, the core information of the target lane is extracted from the port's electronic map. The lane boundary information is the coordinate sequence of the left and right sides of the lane, defining the physical range of the lane. The lane width is the vertical distance between the two sides of the lane. The width of the evacuation lanes in the port area is uniformly 4.5 meters. Traffic sign information includes speed limit signs, directional signs, and no-entry signs. The speed limit for the evacuation lane is 20 kilometers per hour. The directional signs indicate that the lane is for one-way straight travel. Traffic rule information includes vehicle type restrictions, operating time restrictions, and meeting rules. Only transport vehicles are allowed to pass through the evacuation lane. It is open for 24-hour unobstructed operation and reverse driving is prohibited. All extracted information is integrated in a fixed format, and the lane number, lane boundary coordinates, lane width, traffic signs, and traffic rules are labeled to generate basic road network data containing lane attributes. This data completely describes the physical attributes and traffic rules of a single lane, providing basic materials for dynamic road network updates.

[0195] Based on basic road network data and real-time fused positioning data, the congestion status and traffic speed of each lane are dynamically updated, and a structured digital road network model including lane attributes and traffic rules is constructed. Finally, real-time vehicle location data and dynamic road network layers are output.

[0196] Fourthly, one of the core execution steps of the multi-source positioning fusion and digital road source construction module 101 is to integrate basic road network data and real-time positioning data, dynamically perceive the road network traffic status, construct a standardized structured digital road network model, output the real-time vehicle location and dynamic road network layer, and complete the entire process of multi-source positioning fusion and digital road network construction. The specific implementation method is as follows:

[0197] The real-time traffic speed and congestion status involved in the multi-source positioning fusion and digital road source construction module 101 (or the corresponding steps of the port transportation management method involved in Embodiment 1) are the core indicators of the dynamic road network. The multi-source positioning fusion and digital road source construction module 101 (or the corresponding steps of the port transportation management method involved in Embodiment 1) calculates the average traffic speed of each lane based on all port lanes covered by basic road network data, with an update cycle of 30 seconds, combined with real-time fused positioning data. The calculation method is the average real-time speed of all vehicles traveling on that lane within the statistical period. The real-time speed of the vehicle is obtained by dividing the distance between the fused positioning coordinates of two adjacent moments by the time interval, in kilometers per hour. The congestion status is divided into three levels according to the average traffic speed: smooth, slow, and congested. The smooth status is an average speed greater than 15 kilometers per hour, the slow status is an average speed of 5 to 15 kilometers per hour, and the congested status is an average speed less than 5 kilometers per hour. Each level corresponds to a different road network visualization label, which is convenient for the dispatch center to identify intuitively and realize the dynamic quantification of the traffic status of each lane.

[0198] The structured digital road network model uses basic road network data as its framework, incorporating dynamically updated congestion status, traffic speed, and real-time vehicle location information. It adopts a hierarchical structured design, consisting of three layers: a lane foundation layer, a dynamic status layer, and a vehicle positioning layer. The lane foundation layer stores static attributes such as lane number, lane boundary, lane width, and traffic rules, and the data remains stable. The dynamic status layer stores dynamic information for each lane, such as real-time average speed, congestion status, and update time, and is automatically refreshed periodically. The vehicle positioning layer stores real-time data for all transport vehicles, including their fused positioning coordinates, lane, vehicle number, and direction of travel, and is updated synchronously with the vehicle positioning data collection frequency. These three layers of data are interconnected, forming a logically clear and data-complete structured model. The model supports horizontal expansion to add lanes and vertical refinement of road network rules, adapting to the needs of port area road network adjustments and operational expansion.

[0199] The real-time vehicle location data is the core output of the vehicle positioning layer in the structured digital road network model. It integrates vehicle number, fused positioning coordinates, lane location, speed, and positioning time. The data accuracy is consistent with the fused positioning data, with coordinate accuracy of 0.1 meters, speed accuracy of 0.1 km / h, and time accuracy at the millisecond level. This data reflects the real-time operating position and status of a single vehicle. The dynamic road network layer is the visualization output carrier of the structured digital road network model. Based on the port electronic map base map, it is generated by overlaying lane basic attributes, dynamic traffic status, and real-time vehicle location. The layer uses a rendering method combining raster and vector, with unobstructed lanes marked in green and slow-moving lanes marked in yellow. Congested lanes are marked in red, and vehicle positions are displayed as dynamic icons. The icon size adjusts according to the vehicle type, and text labels such as lane number, speed limit, and direction of travel are overlaid. The layer refresh frequency is consistent with the road network status update cycle and can be selected as 30 seconds to ensure that the road network information viewed by the dispatch center and terminal equipment is real-time and effective. Finally, the real-time vehicle location data and the dynamic road network layer are output synchronously. The real-time vehicle location data is used for subsequent route planning and task assignment, while the dynamic road network layer is used for operation monitoring and visual dispatch. Together, they constitute the output of the multi-source positioning fusion and digital road network construction module, providing stable and accurate underlying data support for the subsequent functional modules of the port transportation system.

[0200] The dynamic route planning and intelligent task assignment module 102 (i.e., the route planning module of embodiment 2) is used to receive real-time vehicle location data and dynamic road network layers, combine the cargo type, destination and timeliness requirements of the transport work order, calculate the optimal transport route through a multi-objective optimization algorithm, and automatically assign work tasks according to vehicle status and task priority, generating a scheduling scheme including route guidance and task instructions; specifically, the dynamic route planning and intelligent task assignment module 102 of this embodiment performs the following sequential steps S1021-S1023 (optionally, S3 of embodiment 1 includes steps S1021-S1024):

[0201] S1021: Analyze the cargo type, destination coordinates, and timeliness requirements in the transportation work order, and combine this with the vehicle's current coordinates and idle status from real-time vehicle location data to generate a set of task requirement parameters. The core of this step is to complete the standardized parsing of transportation work orders and vehicle status information, transforming the work order business attributes and vehicle real-time operation attributes into a quantifiable and calculable set of parameters, providing basic input data for subsequent route planning and task assignment. The specific implementation method is as follows:

[0202] The transport work order is the core business certificate for port evacuation operations, carrying all business needs for cargo transportation. The parsing process requires connecting to the port business management platform via a system data interface to automatically extract structured data from the work order, avoiding errors and delays caused by manual entry. Data extraction accuracy must cover all key fields without missing information. Cargo type is a core parameter distinguishing the characteristics of transport operations. Based on typical port evacuation business scenarios, it can be divided into four categories: bulk cargo, general cargo, containers, and hazardous materials. Different cargo types correspond to different road network access restrictions and operational requirements. For example, bulk cargo transport vehicles can travel on the port's internal main roads and dedicated freight lanes, while hazardous material transport vehicles can only travel on designated dedicated closed lanes. In this example, the cargo type of the currently parsed transport work order is set to containers, and the corresponding cargo type parameter is marked as G_1. This parameter directly determines the lane adaptation rules in subsequent route planning. Destination coordinates are used to accurately locate the final position of the transport vehicle. They are marked using the port's local Cartesian coordinate system, with coordinates divided into horizontal (X) and vertical (Y) coordinates, all in meters, with an accuracy controlled to 0.1 meters to ensure the vehicle can accurately reach the work point. In this example, the destination coordinates for this work order are set to X_1 = 326.8 meters and Y_1 = 514.3 meters, corresponding to the designated loading and unloading point in the container storage area within the port. Timeliness requirements are used to define the urgency of the transport task, divided into three levels: urgent work orders, regular work orders, and ordinary work orders, each corresponding to different transport timeliness thresholds. Urgent work orders require a transport time not exceeding 15 minutes, regular work orders not exceeding 30 minutes, and ordinary work orders not exceeding 45 minutes. The timeliness parameter is marked as T_1. In this example, this work order is a regular work order, with a timeliness threshold set to 30 minutes. This parameter is the core basis for the timeliness index in subsequent multi-objective path optimization.

[0203] The real-time vehicle location data is output by a multi-source positioning fusion module. During parsing, two core pieces of information need to be extracted: the vehicle's current coordinates and its idle status. The vehicle's current coordinates also adopt the port's local Cartesian coordinate system, maintaining a consistent reference with the destination coordinates to facilitate subsequent path distance calculations. In the example, the current coordinates of the vehicle to be dispatched are X_2=142.5 meters and Y_2=207.6 meters. The coordinates are updated once per second to ensure data timeliness. The vehicle's idle status is a key indicator for determining whether a vehicle can undertake a new task. It is divided into four states: idle, in operation, under maintenance, and awaiting inspection. Only vehicles in the idle state can participate in task assignment. The status parameter is marked as S_1. In the example, the vehicle's current state is idle, which meets the basic conditions for undertaking a task.

[0204] The parsed cargo type parameter G_1, destination coordinate parameters X_1 and Y_1, timeliness parameter T_1, vehicle current coordinate parameters X_2 and Y_2, and vehicle status parameter S_1 are integrated and encapsulated according to a fixed data logic order to form a task requirement parameter set. This parameter set is stored in a standardized data structure, and all parameters are labeled with name, value, unit, and attribute type, with no redundant data or missing information. It becomes the core data carrier connecting work order requirements and vehicle resources, providing complete input conditions for subsequent path cost calculation.

[0205] S1022, based on the topology and real-time traffic rules in the dynamic road network layer, the A* algorithm is used to pre-calculate the traffic cost of each candidate path, generating a path cost matrix; the traffic cost includes travel distance, estimated time, and energy consumption; the core of this step is to rely on the real-time status of the dynamic road network to complete the preliminary screening and cost quantification of candidate paths through a classic path search algorithm, transforming abstract road network traffic conditions into numerical cost indicators, forming matrix data that can be used for optimization calculations. The specific implementation method is as follows:

[0206] The dynamic road network layer is a real-time road network data carrier constructed by the multi-source positioning fusion module. The topology refers to the connection relationship and connectivity between lanes, intersections, and work points within the port, presented in the form of road network nodes and road segments. Nodes represent key locations such as intersections, loading and unloading points, and weighbridges, while road segments represent the traffic lanes between adjacent nodes. The topology clearly defines the range of drivable paths for vehicles, avoiding invalid paths that are impassable. The real-time traffic rules are dynamically updated road network constraints, including speed limits for road segments, lane closures, temporary congestion, and one-way traffic. The rule data is adjusted in real time according to the port's on-site operations. For example, if a road segment is temporarily closed due to loading and unloading operations, the traffic rules will be updated to a prohibited state to ensure that the route planning conforms to the actual on-site traffic conditions.

[0207] The A* algorithm is the core search algorithm for pre-computing candidate paths. This algorithm combines the advantages of cost search and heuristic search, and can quickly filter out the optimal candidate path in complex road networks. The core formula of the algorithm is f(n)=g(n)+h(n), where f(n) represents the total travel cost from the starting point to the destination, g(n) represents the actual travel cost from the vehicle's current location node to the current search node, and h(n) represents the estimated travel cost from the current search node to the destination node. The estimated cost is calculated using Euclidean distance to ensure a balance between the algorithm's search efficiency and accuracy. In the port road network scenario, the road network node corresponding to the vehicle's current coordinates is taken as the starting point, and the road network node corresponding to the destination coordinates is taken as the ending point. The A* algorithm traverses the road network topology, eliminates road segments that do not meet the real-time traffic rules, and pre-computes three feasible candidate paths, which are marked as path L_1, path L_2, and path L_3, respectively. All three paths meet the lane passage requirements corresponding to the cargo type and have no prohibited or illegal road segments.

[0208] The toll cost is a core indicator for evaluating the merits of candidate routes, comprising three dimensions: travel distance, estimated time, and energy consumption. All three dimensions are quantified numerically to form a unified quantitative standard. Travel distance refers to the total length traveled by the vehicle along the candidate route, measured in meters, calculated by summing the lengths of road network segments. In the example, the travel distance for path L_1 is 865 meters, for path L_2 it is 920 meters, and for path L_3 it is 790 meters. Estimated time refers to the time it takes for the vehicle to complete the route at the speed limit of each road segment, measured in minutes. The formula for calculating estimated time is: Estimated Time = Travel Distance ÷ Average Travel Speed. The average travel speed is calculated based on the real-time vehicle speed of the road segments in the dynamic road network layer. In the example, the estimated time for path L_1 is 22 minutes, for path L_2 it is 25 minutes, and for path L_3 it is 18 minutes. Energy consumption refers to the fuel or electricity consumption during vehicle operation, measured in liters or kilowatt-hours. It is related to the travel distance, road conditions, gradient, and vehicle load. In the example, the energy consumption for path L_1 is 6.2 liters, for path L_2 it is 6.8 liters, and for path L_3 it is 5.5 liters.

[0209] The three candidate paths are arranged according to their path number and cost type to form a two-dimensional path cost matrix. The rows of the matrix correspond to the candidate paths, and the columns correspond to the three cost indicators: travel distance, estimated time, and energy consumption. Each value in the matrix is ​​a quantitative result after precise calculation, with no estimation error. This path cost matrix realizes the digital presentation of the travel costs of candidate paths, transforming complex road network traffic conditions into intuitive numerical data, and providing a standardized computational basis for subsequent multi-objective optimization algorithms to solve for the optimal path.

[0210] S1023: Input the path cost matrix and task requirement parameter set into the multi-objective optimization model, and use a non-dominated sorting genetic algorithm with an elitist strategy to solve for the Pareto optimal path set, thereby comprehensively considering timeliness, economy, and safety to generate the optimal path sequence. The core of this step is to balance multi-dimensional transportation objectives through intelligent optimization algorithms, select the path set that meets the comprehensive optimal conditions from the candidate paths, and then form the final optimal path sequence by sorting the paths by objective weights, taking into account the multiple needs of the transportation operation. The specific implementation method is as follows:

[0211] The multi-objective optimization model is a mathematical model used to balance the three objectives of timeliness, economy, and safety. This model uses the travel distance, estimated time, and energy consumption in the path cost matrix as basic input variables, and combines them with constraints such as timeliness requirements and cargo type in the task requirement parameter set to construct three optimization objective functions: a timeliness objective function F_1, an economy objective function F_2, and a safety objective function F_3. The timeliness objective function F_1 uses estimated time as the core variable; a smaller value indicates better timeliness, and it must meet the work order timeliness threshold constraint. The economy objective function F_2 uses energy consumption and travel distance as core variables; a smaller value indicates lower transportation costs. The safety objective function F_3 uses the violation risk and road condition complexity in the path as core variables, and determines the safety weight based on the cargo type. The safety weight for container transportation is appropriately set, and the safety weight for hazardous chemical transportation is appropriately adjusted; a smaller value indicates higher safety.

[0212] The non-dominated sorting genetic algorithm with an elitist strategy is the core algorithm for solving multi-objective optimization problems, abbreviated as NSGA-II algorithm. The elitist strategy is to retain the best individuals in each generation of computation, avoid losing high-quality solutions during iteration, and improve the convergence speed and solution accuracy of the algorithm. When the algorithm runs, it first converts the path cost matrix and task requirement parameter set into chromosome codes that the algorithm can recognize. Each chromosome corresponds to a candidate path. Then, the path solutions are iteratively optimized through selection, crossover, and mutation operations. The number of iterations is set to 50 to ensure the stability of the solution results. During the iteration process, the algorithm performs non-dominated sorting of the path solutions according to the three objective functions, and selects the set of paths that are not surpassed by other path solutions in all objectives, namely the Pareto optimal path set. The paths in this set have achieved local optima in terms of timeliness, economy, and safety, and there are no absolutely inferior paths. In the example, after the algorithm solves the problem, the Pareto optimal path set includes two paths, L_1 and L_3. Path L_2 is excluded from the optimal set because its expected time and energy consumption are higher than the other two paths.

[0213] Based on the Pareto optimal path set and combined with the actual operational preferences of port transportation, the three optimization objectives are weighted and comprehensively ranked. The weights for timeliness, economy, and safety are appropriately set. The comprehensive score of each optimal path is calculated by weighted summation. The higher the score, the better the comprehensive performance of the path. In the example, the comprehensive score of path L_1 is 78 points and the comprehensive score of path L_3 is 92 points. The optimal path sequence is generated by arranging the paths from high to low. The first path in the sequence is path L_3, which has the best comprehensive performance and is the priority transportation path for vehicles. The second path is path L_1, which is the backup path in case of path deviation or congestion. The optimal path sequence satisfies the timeliness requirements of the work order while taking into account transportation costs and operational safety, thus achieving a balanced optimization of multiple objectives.

[0214] S1024: Based on the vehicle's current load capacity, fuel range, and task execution records, and combined with task priority rules, the system automatically assigns work tasks to each idle vehicle, ultimately generating a scheduling plan that includes route guidance and task instructions. The core of this step is to achieve precise matching between vehicle resources and transportation tasks. It intelligently assigns tasks based on vehicle hardware conditions, historical performance, and task urgency, integrating route information and task requirements to form an executable scheduling plan. The specific implementation method is as follows:

[0215] The vehicle's current status includes three core indicators: load capacity, fuel remaining range, and task execution record. These are crucial for determining whether the vehicle is suitable for the current task. Load capacity refers to the vehicle's rated load-bearing capacity, measured in tons, and must match the weight of the goods in the transport order. In the example, the vehicle to be assigned has a rated load capacity of 35 tons, and the current order's goods weight is 20 tons, meeting the load requirement and eliminating the risk of overloading. Fuel remaining range refers to the remaining mileage of the vehicle, measured in kilometers, and must be greater than the mileage of the longest path in the optimal path sequence. In the example, the vehicle has 80 kilometers of remaining fuel, and the longest mileage of the optimal path is 0.92 kilometers, indicating sufficient range to complete the transport task. The task execution record is the vehicle's historical operational data, including historical task completion rate, number of path deviations, and number of violations. A higher completion rate and fewer violations indicate stronger vehicle operational reliability. In the example, the vehicle's historical task completion rate is 98%, with no violations, demonstrating excellent operational performance.

[0216] The task priority rule is the core sorting basis for task assignment. The rule divides priority levels according to the timeliness requirements of the work order and the importance of the goods. Urgent work orders are level 1, marked as P_1, regular work orders are level 2, marked as P_2, and ordinary work orders are level 3, marked as P_3. The system prioritizes assigning work orders with higher priority. Work orders with the same priority are assigned in the order of their creation time. In the example, the current work order is a regular level 2 work order. It is ranked second in priority among the tasks to be assigned in the system. The system automatically matches the available vehicle with the best status and the closest location to accept the task, thus achieving optimal matching between the task and the vehicle.

[0217] After task assignment is completed, the system integrates route information from the optimal route sequence, operational requirements of the transport work order, and task details of vehicle execution to generate a standardized scheduling plan. The route guidance in the scheduling plan includes the optimal route's lane driving sequence, turning points, speed limits, and estimated arrival time, presented in visual text and location descriptions for easy viewing by drivers and the dispatch center. The task instructions include core content such as cargo loading and unloading points, operation timeliness, precautions, and work order numbers, clearly defining the requirements for the entire operation process. The scheduling plan is simultaneously pushed to the vehicle terminal and the dispatch center's large screen. The vehicle terminal displays route navigation and task reminders in real time, while the dispatch center monitors the task assignment status and route execution progress in real time, achieving real-time synchronization of scheduling information. This scheduling plan, through intelligent matching and standardized output, replaces the traditional extensive mode of manual scheduling, improving the scheduling efficiency and accuracy of port transportation operations and ensuring the efficient operation of the entire transportation system.

[0218] The operation process monitoring and anomaly warning module 103 (i.e. the anomaly monitoring module of embodiment 2) is used to receive the scheduling plan and track the vehicle execution trajectory, compare the vehicle position and driving route in real time based on the digital road network model, and trigger an alarm when it is detected that the vehicle deviates from the route, speeds, or illegally stops, and generate an alarm record with a timestamp and violation type; optionally, at the same time as triggering the alarm, the abnormal event information is pushed to the mobile terminal.

[0219] Specifically, the execution steps corresponding to the work process monitoring and anomaly early warning module 103 may include the sequential execution of S1031-S1034:

[0220] S1031: The vehicle-mounted terminal receives route guidance and task instructions from the scheduling plan and transmits real-time vehicle location data back at a fixed frequency, generating a sequence of actual vehicle driving trajectory points. The core of this step is to rely on the vehicle-mounted terminal to complete the reception of scheduling information and the transmission of location data. Through stable communication transmission and regular data collection, continuous and traceable vehicle driving trajectory data is formed, providing a basic data source for subsequent route comparison and anomaly detection. The specific implementation method is as follows:

[0221] The vehicle-mounted terminal is an integrated hardware unit installed on port transport vehicles, combining data reception, positioning acquisition, and wireless transmission functions. It is the core interactive device connecting the vehicle and the dispatch center. The terminal adopts a standardized wireless communication protocol to establish a stable connection with the port's internal local area network, with data transmission latency controlled within 100 milliseconds, ensuring real-time interaction between the dispatch plan and location data. The terminal receives the dispatch plan issued by the dynamic path planning and intelligent task assignment module through a dedicated data interface, and fully obtains the path guidance and task instruction information. The path guidance includes the coordinate node sequence of the planned path, road segment turning prompts, and passage node identification, while the task instructions include core information such as cargo transportation targets, operation time requirements, and vehicle operation numbers. After receiving the data, the terminal automatically parses the data content, converting the textual path and instructions into visual navigation information displayed on the vehicle screen. At the same time, it performs data integrity verification. Missing or incorrect data will trigger a local temporary storage mechanism, and the data will be re-requested after the network is restored, ensuring that no dispatch information is lost.

[0222] The fixed frequency is the core parameter for controlling the return of location data. It refers to the periodic collection and upload of positioning data by the terminal according to the set time interval. This frequency is set to 5 Hz according to the accuracy requirements of port operations, that is, data collection and return are completed every 200 milliseconds. The higher the frequency value, the higher the density of trajectory points and the higher the accuracy of trajectory reconstruction, which is suitable for the operation scenarios in the port where the lanes are narrow and the driving routes are complex. The real-time location data returned is not a single coordinate information, but a complete dataset containing multi-dimensional parameters, specifically including vehicle fused positioning coordinates, data collection timestamp, real-time driving speed, vehicle heading angle, and vehicle operating status identifier. Among them, the fused positioning coordinates are the two-dimensional plane coordinates of the vehicle in the port's digital road network, in meters, with an accuracy controlled within 0.5 meters; the timestamp adopts the format of year-month-day hour:minute:second, with a millisecond format and an accuracy of 10 milliseconds, which is used to mark the collection time of each trajectory point to ensure the temporal continuity of the trajectory; the real-time driving speed is in kilometers per hour with an accuracy of 0.1 kilometers per hour; the heading angle represents the direction of vehicle travel, in degrees, with a value range of 0 to 360 degrees, which is used to help determine the vehicle's driving trend.

[0223] The actual vehicle trajectory point sequence is a dataset formed by arranging continuously transmitted location data in chronological order. Each trajectory point corresponds to the complete vehicle status information at a given time of acquisition. The sequence is generated following a time-incrementing principle, with newly acquired trajectory points automatically appended to the end of the sequence. Simultaneously, the system automatically removes duplicate data and invalid data with abnormal coordinates to ensure the sequence's validity. In the example, when a transport vehicle is performing a task, the onboard terminal transmits data at a frequency of 5 Hz, generating 50 trajectory points within 10 seconds. Each trajectory point contains information such as coordinates, timestamp, and speed at the corresponding time. By concatenating these trajectory points in chronological order, the vehicle's actual driving route during that time period can be completely reconstructed, forming a continuous actual driving trajectory point sequence. This provides accurate raw data for subsequent spatial matching and path comparison.

[0224] S1032, perform real-time spatial matching between the actual driving trajectory point sequence and the planned path in the structured digital road network model, calculate the lateral offset and longitudinal progress deviation, and generate a trajectory deviation index. The core of this step is to achieve accurate comparison between the actual trajectory and the planned path through spatial matching algorithms, and to form a deviation index that can intuitively judge the compliance of the trajectory by quantitatively calculating the offset and progress deviation, providing a quantitative basis for route deviation warning. The specific implementation method is as follows:

[0225] The planned paths stored in the structured digital road network model are vector paths formed by connecting a series of continuous road network coordinate nodes. They contain key information such as the centerline coordinates, road segment lengths, and lane ranges. Real-time spatial matching is the core process of associating each trajectory point in the actual driving trajectory point sequence with the vector data of the planned path. The matching operation is completed using a projection matching algorithm. The core logic of this algorithm is to vertically project each actual trajectory point onto the centerline of the planned path and find the corresponding matching point, thus achieving a precise association between the actual position and the planned path. The matching accuracy of the algorithm is controlled within 0.3 meters, which can adapt to the detection requirements of small distance offsets in ports. The matching process is executed in real time and synchronously. Each time a new trajectory point is added, a matching calculation is immediately completed without waiting for the sequence to accumulate, ensuring the real-time performance of trajectory comparison. After the matching is completed, a correspondence table between trajectory points and planned paths is generated, which clearly defines the corresponding position of each actual position on the planned path.

[0226] The lateral offset is a core parameter for measuring the vehicle's deviation from the planned path centerline, denoted by the symbol D_lateral. It refers to the vertical straight-line distance from the actual trajectory point to its projection point on the planned path centerline, in meters. This parameter directly reflects whether the vehicle is driving beyond the planned lane. The larger the offset value, the more serious the deviation from the route. In the example, the actual coordinates of a trajectory point are (156.2, 89.7), and the coordinates of its projection point on the planned path centerline are (155.8, 90.1). The lateral offset D_lateral is calculated to be 0.6 meters using the planar distance calculation formula, which is within the normal driving range. The longitudinal progress deviation is a parameter for measuring the difference between the vehicle's driving progress and the planned progress, denoted by the symbol D_longitudinal. It refers to the difference between the mileage actually traveled by the vehicle and the corresponding mileage in the planned path, in meters. The calculation first counts the cumulative mileage of the planned path from the starting point to the matching point, and then counts the cumulative mileage actually traveled by the vehicle. The difference between the two is the progress deviation. A positive deviation indicates that the vehicle is ahead of schedule, and a negative deviation indicates that the vehicle is behind schedule.

[0227] The trajectory deviation index is a quantitative evaluation indicator that integrates lateral offset and longitudinal progress deviation, denoted by K_deviation. It is generated through weighted calculation and can uniformly measure the compliance level of vehicle trajectories. The calculation model is K_deviation = α × D_lateral + β × |D_longitudinal|, where α is the lateral offset weight coefficient (appropriately selected), and β is the longitudinal progress deviation weight coefficient (appropriately selected). The weight coefficients are set according to the port operation priority; lateral offset has a greater impact on driving compliance, hence its higher weight. Absolute value processing is to eliminate the influence of positive and negative progress deviation values, focusing only on the deviation magnitude. The system calculates the trajectory deviation index in real time for each trajectory point. The smaller the index value, the higher the fit between the vehicle's driving trajectory and the planned path; the larger the index value, the more severe the trajectory deviation. In the example, the lateral offset is 0.6 meters and the longitudinal progress deviation is -2 meters. Substituting these values ​​into the formula, we get K_deviation = 0.6 × 0.6 + 0.4 × 2 = 1.16. This index value can be directly used for subsequent deviation threshold comparison and judgment.

[0228] S1033: Speeding is detected by comparing the vehicle's real-time speed with the speed limit rules in the dynamic road network layer. Illegal parking is detected based on dwell time and stop area rules in the road network, generating a list of violations. The core of this step is to automate the detection of two types of abnormal behaviors—speeding and illegal parking—based on road network rules. Through parameter comparison and rule judgment, violations are filtered out and standardized records are formed, providing direct evidence for anomaly alarms. The specific implementation method is as follows:

[0229] The dynamic road network layer contains detailed traffic rules for all lanes within the port. Among these, the speed limit rules are speed limits set for different road sections, denoted by the symbol V_speed_limit, in kilometers per hour. Speed ​​limits vary in different areas of the port: 15 km / h on main roads, 5 km / h near loading / unloading areas, and 8 km / h at curves and intersections. Speed ​​limit data is linked to road section coordinates, allowing the system to automatically match the corresponding speed limit based on vehicle location. Speeding detection is achieved through real-time speed comparison; the vehicle's real-time speed is represented by the symbol V_real-time, determined by the vehicle's position. The terminal collects data in real time. During detection, the real-time speed limit (V_) is compared with the corresponding speed limit (V_) of the road segment. The speeding judgment threshold is set to 1.1 times the speed limit. That is, when V_real-time > V_speed limit × 1.1, it is judged as speeding. This threshold can avoid false judgments caused by small errors in speed collection, balancing detection accuracy and rationality. In the example, the vehicle travels to a road segment near the loading and unloading area. The speed limit of this road segment is 5 km / h. The real-time speed limit (V_) collected is 6.2 km / h. 6.2 > 5 × 1.1 = 5.5, which meets the speeding judgment condition. The system immediately marks this behavior as a speeding violation.

[0230] The stopping area rules in the dynamic road network layer divide the port area into permitted stopping areas and prohibited stopping areas. Permitted stopping areas include vehicle waiting areas, the area around weighbridges, and loading / unloading operation points. Prohibited stopping areas include main roads, the middle of lanes, fire lanes, intersections, and other areas that affect traffic flow. Stopping area information is linked to digital road network coordinates, allowing for precise determination of the area type based on vehicle location. Dwell time is the core parameter for detecting illegal stopping, denoted by the symbol T_dwell, which refers to the duration for which a vehicle remains stationary in the same position without moving, measured in seconds. The system... The dwell time is calculated based on the coordinate changes of continuous trajectory points. When the difference between the coordinates of multiple adjacent trajectory points is less than 0.2 meters, the vehicle is determined to be in a stopped state, and the cumulative time is T_dwell. The rule for judging illegal parking is that the vehicle is in a no-parking zone and T_dwell > 30 seconds. 30 seconds is the set threshold for judging parking. Brief stops to avoid obstacles will not be judged as illegal. In the example, the vehicle stayed continuously in the no-parking zone of the main road for 45 seconds. T_dwell = 45 seconds > 30 seconds, which meets the judgment condition for illegal parking. The system marks this behavior as illegal parking.

[0231] The violation event list is a standardized dataset storing all detected violations. Each violation record in the list contains complete behavioral information, including the vehicle number, timestamp of the violation, coordinates of the location of the violation, violation type, and violation quantification value. Violation types are divided into speeding and illegal parking. The violation quantification value is the difference between the actual speeding speed and the actual parking time for speeding and illegal parking, respectively. The list adopts a dynamic update mechanism. Each violation is immediately added to the list and marked with a processing status. Unprocessed violations are always retained in the list, forming a complete violation event record, providing a direct basis for subsequent alarm triggering.

[0232] S1034, when the trajectory deviation exceeds the preset threshold or the list of violations is not empty, an alarm is immediately triggered and a warning record with a timestamp and violation type is pushed to the driver's mobile terminal. Simultaneously, it is stored in the abnormal event database, generating a warning record. The core of this step is to trigger an alarm based on the deviation threshold and violation event determination, complete the push and storage of warning information, form a traceable warning record, and achieve real-time handling and data retention of abnormal situations. The specific implementation method is as follows:

[0233] The preset threshold is a critical value for judging whether the trajectory deviation is abnormal, denoted by the symbol K_threshold. It is set to 5 meters according to the port lane width and operation specifications. The system uses OR logic to determine the alarm trigger condition, that is, the trajectory deviation index K_deviation > K_threshold, or there are unprocessed violation records in the violation event list (the list is not empty). If either condition is met, the alarm is triggered immediately. The alarm trigger response time is controlled within 50 milliseconds to ensure that abnormal situations are detected as soon as possible. The alarm methods are divided into vehicle terminal local alarm and mobile terminal remote alarm. The vehicle terminal reminds the driver through sound and light signals. The sound and light alarm intensity is adapted to the port operation environment and can clearly convey the warning information. The driver's mobile terminal receives the warning through pop-up prompts and voice broadcasts. The voice content includes the violation type and location prompts, which makes it easy for the driver to quickly know the abnormal situation and correct the behavior.

[0234] The timestamp is the core identifier parameter of the warning record. It adopts a high-precision format consistent with the trajectory acquisition to accurately mark the time when the alarm is triggered, ensuring that the warning time corresponds completely with the time when the abnormality occurs. The violation type is directly associated with the detected abnormality type, which is divided into three categories: route deviation, speeding, and illegal parking, clearly defining the abnormal behavior attributes. The warning record is standardized data that integrates alarm information. In addition to the timestamp and violation type, it also includes complete information such as vehicle number, abnormal location coordinates, trajectory deviation value, speeding speed or dwell time. Each warning record has a unique identifier, which facilitates subsequent querying and statistics.

[0235] The abnormal event database is a dedicated database storing all early warning records. It employs a real-time write mechanism, where early warning records are immediately and synchronously stored after generation. The database supports persistent data storage and rapid retrieval, with storage time covering the entire port operation cycle, facilitating subsequent operation review and violation statistics. While early warning records are pushed to the driver's mobile terminal, they are also simultaneously sent to the dispatch center's monitoring interface. Dispatchers can view the abnormal situations of all vehicles within the port in real time, achieving remote supervision. In the example, the vehicle trajectory deviation K_deviation is 6.2 meters, exceeding the K_threshold of 5 meters. The system immediately triggers an alarm, generating an early warning record with a timestamp of 2026-03-15 09:25:36.120 and a violation type of route deviation. This record is pushed to the driver's mobile terminal and stored in the abnormal event database, ultimately completing the generation of a complete early warning record and achieving closed-loop management of abnormal operations throughout the entire process.

[0236] The paperless weighing and data closed-loop feedback module 104 (in some preferred embodiments, the paperless weighing and data closed-loop feedback module 104 of this embodiment may include the weighing management module and data synchronization module as described in Embodiment 2) is used to receive early warning records and vehicle arrival signals at the weighbridge, automatically trigger the electronic weighbridge to complete the weighing, associate the weighing data with the transportation work order, generate a closed-loop feedback report containing task completion status, weighing information and anomaly statistics, and synchronously update it to the dispatch center and mobile terminal, thereby realizing full-process digital management of the transportation operation;

[0237] Specifically, it can receive the latest warning information from the warning record database and the arrival signal triggered by the infrared sensor when a vehicle arrives at the weighbridge. It then automatically starts the electronic weighbridge to perform the weighing operation, acquiring the vehicle's gross weight and tare weight data and generating a weighing record. The core of this step is to achieve real-time retrieval of warning information and automatic detection of vehicle arrival at the weighbridge. By linking the infrared sensor and the electronic weighbridge equipment, it achieves unmanned weighing operations, accurately collecting vehicle weight-related data and generating standardized weighing records. This lays the foundation for subsequent work order association and data statistics. The specific implementation method is as follows:

[0238] The early warning record database is the core storage unit for storing all abnormal early warning data during port operations. It adopts a distributed real-time database architecture, featuring high-concurrency read / write and low-latency query capabilities. Data write latency is controlled within 100 milliseconds. It can synchronize all early warning records generated by the operation monitoring and abnormal early warning modules in real time, ensuring data timeliness and integrity. The database uses a structured data storage model. Each early warning record contains core fields such as a unique identifier (ID), vehicle number, timestamp, early warning type, violation location, and offset value. The timestamp uses second-level precision, formatted as year-month-day hour:minute:second, accurately recording the moment the early warning occurred. Early warning types are divided into three categories: route deviation, speeding, and illegal parking. The offset value quantifies the distance of route deviation. The unit is meters, and the numerical precision is retained to one decimal place. In this step, the system uses the real-time database query interface to retrieve all the latest warning information generated by the corresponding vehicle in this transportation task, using the vehicle number as the only search condition. The query interface sets the data filtering threshold T_1 to 30 minutes, that is, it only retrieves the warning data of the vehicle within 30 minutes from receiving the task to arriving at the weighbridge, to avoid interference from invalid historical data and ensure that the obtained warning information is highly correlated with the current operation task. In the example, the latest warning information retrieved for the transport vehicle with vehicle number V_20260315008 is a route deviation warning, with a timestamp of 2026-03-15 14:22:36, an offset value of 2.5 meters, and a warning location in lane L_07 of the port digital road network.

[0239] The signal of a vehicle arriving at the weighbridge is triggered by an infrared sensor. The infrared sensor is deployed on both sides of the entrance and exit of the electronic weighbridge and adopts the through-beam infrared detection principle. When the body of the transport vehicle completely enters the detection area of ​​the infrared sensor, it will block the transmission of infrared light. The sensor will then generate a high-level trigger signal, which is the signal of the vehicle arriving at the weighbridge. The signal transmission adopts a wired communication method, and the transmission delay is controlled within 50 milliseconds to ensure the stability of the signal transmission. The infrared sensor has a detection height threshold H_1 of 0.8 meters to 3.5 meters and a detection width threshold W_1 of 3 meters to 5 meters, which can adapt to the body size of different types of transport vehicles in the port. This avoids false or missed signal triggering due to differences in vehicle size. At the same time, the sensor has built-in signal anti-shake processing logic, with the anti-shake time parameter D_1 set to 2 seconds. That is, the arrival signal is only confirmed to be valid when the vehicle is detected to be obstructed for 2 consecutive seconds. This eliminates false signals caused by factors such as bumps and temporary stops during vehicle movement. In the example, after the transport vehicle with vehicle number V_20260315008 completely drove into the weighbridge, the infrared sensor continuously detected the obstruction for 2 seconds, successfully generated a valid arrival signal and transmitted it to the system control unit.

[0240] Upon receiving the latest warning information and valid arrival signal simultaneously, the system immediately sends a start command to the electronic weighbridge, triggering automatic weighing. The entire weighing process requires no manual intervention, achieving the core of paperless weighing. The electronic weighbridge's weighing process consists of two stages: empty tare weight detection and full-load gross weight detection. First, tare weight data is collected when the vehicle is unloaded. Tare weight is the vehicle's own weight, measured in kilograms, with an accuracy controlled within ±5 kilograms. Then, gross weight data is collected when the vehicle is loaded with goods. Gross weight is the total weight of the vehicle and its goods, and the weighing accuracy remains consistent with the tare weight. During the process, the electronic weighbridge incorporates a built-in weight filtering algorithm to eliminate external interference factors such as vibrations during vehicle parking and minor ground subsidence. The algorithm employs a moving average filtering method, setting the filtering window parameter F_1 to 10 sampling points. This means that after continuously collecting weight data 10 times, the average value is calculated as the final weighing result, ensuring the accuracy of the weight data. In the example, for the transport vehicle with vehicle number V_20260315008, the tare weight data collected was 12,500 kg, and the gross weight data collected was 38,600 kg. Both sets of weight data were confirmed to be valid after being processed by the filtering algorithm.

[0241] After the weighing data is collected, the system integrates information such as vehicle number, tare weight, gross weight, weighing timestamp, and weighbridge number to generate a standardized weighing record. The weighing record is stored in a unified data format, and each field is clearly labeled with its meaning and unit. The weighing timestamp is consistent with the timestamp of the vehicle arrival signal. The weighbridge number is used to identify the electronic weighbridge used for this weighing. The weighing record generated in the example is: Vehicle number V_20260315008, tare weight 12500 kg, gross weight 38600 kg, weighing time 2026-03-15 14:25:10, weighbridge number P_03. This weighing record is stored in real time in the local cache and weighing database to provide accurate weight data support for subsequent work order association.

[0242] Specifically, the paperless weighing and data closed-loop feedback module 104 of this embodiment (in some preferred embodiments, the paperless weighing and data closed-loop feedback module 104 of this embodiment may include the weighing management module and data synchronization module as described in Embodiment 2) associates the weighing record with the current transportation work order through the work order number, updates the work order status to "completed", and attaches the weighing time and weight data to generate a task completion record including weighing information. The core of this step is to achieve precise binding between weighing data and transportation work orders through a unique work order number, complete the real-time update of work order status and integrate weighing-related information, generate a standardized task completion record, and establish a data link between weighing data and transportation tasks. The specific implementation method is as follows:

[0243] The transport work order is the core task certificate for port transportation operations. It is stored in the transport work order database. Each work order is assigned a unique work order number, which consists of three parts: date code, task sequence code, and vehicle association code. It is unique and identifiable and is the core identifier for linking weighing records with work orders. In the example, the transport work order number corresponding to vehicle number V_20260315008 is WO_202603150068. This work order contains core information such as cargo type, destination coordinates, time requirement, vehicle number, and initial status of the work order. The initial status is pending execution by default. During the task execution process, it will be updated to executing. After weighing is completed, it will be updated to completed.

[0244] A data matching algorithm is used to associate weighing records with transportation work orders. The algorithm uses the work order number as the unique matching field. First, it extracts the associated vehicle number from the weighing record. Then, it uses the vehicle number to retrieve the unique work order number currently being executed for the corresponding vehicle from the transportation work order database. Finally, it writes the work order number into the association field of the weighing record, completing the binding between the two. The matching algorithm sets the matching precision parameter M_1 to 100%, only considering a successful match when the vehicle number and the vehicle association code in the work order are completely identical, avoiding work order mismatches. The response time of the matching process is controlled within 200 milliseconds to ensure the real-time nature of the data association. In the example, vehicle number V_20260315008 is matched with work order number WO_202603150068. The system automatically binds this work order number to the weighing record, confirming the correspondence between the two.

[0245] Work order status updates are a crucial link in the closed-loop transportation operation. After the weighing record is associated with the work order, the status of the transportation work order is immediately updated from "in execution" to "completed." The status update operation is synchronized to the transportation work order database and the dispatch center database to ensure data synchronization and consistency across the entire system. At the same time, the weighing time, tare weight data, gross weight data, and net weight data of the goods in the weighing record are appended to the transportation work order. The net weight of the goods is calculated by subtracting the tare weight from the gross weight. The calculation formula is: Net weight = Gross weight - Tare weight. In the example, the gross weight is 38,600 kg, and the tare weight is 12,500 kg, resulting in a net weight of 26,100 kg. The net weight data is added to the work order as core cargo information to fully record the weight of the goods transported in this operation.

[0246] After updating the work order status and attaching information, the basic information, task execution information, and weighing information of the transportation work order are integrated to generate a task completion record that includes weighing information. The task completion record covers all core content such as work order number, vehicle number, cargo type, destination, task start and end time, weighing time, tare weight, gross weight, net weight, and work order status. The data format is standardized, and all fields are complete and without missing information. In the example, the generated task completion record is: Work Order Number WO_202603150068, Vehicle Number V_202603150 08, cargo type is bulk cargo, destination is port C_02 cargo yard, task start time 2026-03-15 14:05:20, task end time 2026-03-15 14:25:10, weighing time 2026-03-15 14:25:10, tare weight 12500 kg, gross weight 38600 kg, net weight 26100 kg, work order status is completed. This record is stored in real time to the task management database as the core voucher for task completion, providing basic data for subsequent anomaly statistics and report generation.

[0247] This step involves summarizing all warning records, weighing data, and task start and end times for the vehicle during its current mission. Statistical analysis methods are then used to calculate the number of anomalies and operational efficiency, generating an anomaly statistics report. The core of this step is to aggregate and integrate the entire process data of the vehicle's mission, quantify operational anomalies and efficiency through professional statistical analysis methods, and generate intuitive anomaly statistics reports. This provides data support for the optimized management of transportation operations. The specific implementation method is as follows:

[0248] The data aggregation stage is a prerequisite for statistical analysis. The full data of the corresponding vehicle for this task is retrieved from the early warning record database, the weighing database, and the task management database. The retrieval range is based on the start and end times of the task execution. The start time is the moment when the vehicle receives the dispatch plan, and the end time is the moment when the vehicle completes the weighing. This ensures that the aggregated data belongs only to this transportation task and that no data from other tasks is mixed in. The aggregated data includes three categories. The first category is warning records, which contain all warning entries for route deviation, speeding, and illegal parking. Information such as warning type, timestamp, offset value, and speeding value for each record is fully extracted. The second category is weighing data, which includes core weight information such as tare weight, gross weight, net weight, and weighing time. The third category is task execution time data, namely the task start time and task end time, used to calculate the total operation time. In the example, vehicle number V_20260315008 has one route deviation warning record, complete weighing data, and task start and end times of 2026-03-15 14:05:20 and 2026-03-15 14:25:10 respectively.

[0249] The calculation of the number of anomalies uses a classification statistical method, classifying warning records into three categories based on warning type: route deviation, speeding, and illegal parking. The frequency of each type of warning is counted separately, and the total number of anomalies is the sum of the warnings for all three categories. The system sets the statistical threshold parameter S_1 to 1, meaning that for each warning record of a corresponding type, the anomaly count for that type is incremented by 1. The statistical process avoids duplicate counting, ensuring the accuracy of the anomaly count. In the example, the vehicle only experienced one route deviation warning in this task, with zero speeding and illegal parking incidents; therefore, the route deviation count is 1, and the total number of anomalies is 1.

[0250] The calculation of operational efficiency employs a time efficiency analysis algorithm. The core metrics are the total task execution time and the unit cargo transportation time. The total task execution time is the task end time minus the task start time, measured in minutes, with precision rounded to one decimal place. The unit cargo transportation time is the total task execution time divided by the net weight of the cargo, measured in minutes per ton. This measure evaluates the efficiency of a vehicle transporting a unit weight of cargo, directly reflecting operational efficiency. The algorithm uses a time difference conversion formula, converting the year-month-day hour:minute:second format to a second-level timestamp, calculating the difference, and then converting it back to minutes to avoid time calculation errors. In the example, the task start timestamp is 1741932320, and the end timestamp is 1741933510, resulting in a time difference of 1190 seconds, which translates to approximately 19.8 minutes. With a net cargo weight of 26.1 tons, the unit cargo transportation time is approximately 19.8 ÷ 26.1 ≈ 0.76 minutes per ton. This value can be compared with the port's operational efficiency standard to determine if the current operation meets the standard.

[0251] After calculating the number of anomalies and operational efficiency, the summarized raw data and calculated statistical indicators are integrated to generate an anomaly statistics report. The report uses a combination of textual description and numerical presentation, covering core information such as vehicle number, work order number, total task execution time, number of various anomalies, total number of anomalies, net weight of goods, and unit cargo transportation time. The data is clear and intuitive, with a well-defined logical hierarchy. The anomaly statistics report generated in the example is as follows: Vehicle number V_20260315008, work order number WO_202603150068, total task execution time 19.8 minutes, route deviation anomaly 1 time, speeding anomaly 0 times, illegal parking anomaly 0 times, total number of anomalies 1 time, net weight of goods 26.1 tons, unit cargo transportation time 0.76 minutes / ton. This report is generated in real time and stored in the statistical database, providing core statistical data for the generation of closed-loop feedback reports.

[0252] Specifically, the paperless weighing and data closed-loop feedback module 104 of this embodiment (in some preferred embodiments, the paperless weighing and data closed-loop feedback module 104 of this embodiment may include the weighing management module and data synchronization module as described in Embodiment 2) integrates task completion status, weighing information, and abnormal statistical reports to generate a complete closed-loop feedback report, and synchronizes and updates it to the dispatch center database and mobile terminal application through a data interface, thereby realizing digital closed-loop management of the entire evacuation operation process; the core of this step is to integrate the core data of all previous links, generate a closed-loop feedback report covering the entire task process, and realize multi-terminal data synchronization through a standardized data interface, ultimately completing the digital closed-loop management of the entire port evacuation operation process. The specific implementation method is as follows:

[0253] The data integration phase deeply integrates the task completion status and weighing information from the task completion record with the abnormal data and efficiency indicators from the anomaly statistics report. The closed-loop feedback report, as a summary document of the entire transportation operation process, must cover all stages from task receipt, execution, weighing to anomaly monitoring, ensuring data integrity and comprehensive information. The report's structure consists of five parts: basic task information, task execution status, detailed weighing information, anomaly statistics, and operational efficiency analysis. Each part corresponds to previously generated standardized data, with no missing or biased information, achieving end-to-end data connectivity from task assignment to final feedback. The integrated closed-loop feedback report in the example includes basic information such as work order number, vehicle number, cargo type, and destination; the work order's completed execution status; weighing information such as tare weight, gross weight, and net weight; anomaly statistics for one route deviation; and an efficiency analysis of 19.8 minutes of operation time and 0.76 minutes / ton, comprehensively presenting the entire process of this transportation operation.

[0254] After the closed-loop feedback report is generated, data is synchronized and updated through standardized data interfaces. These interfaces are divided into two categories: the dispatch center database interface and the mobile terminal application interface. Both use a common data transmission protocol to ensure data transmission stability and compatibility. The dispatch center database interface synchronizes the report to the port dispatch center's core database, with synchronization latency controlled within 150 milliseconds. The dispatch center can view all vehicle operation feedback reports in real time and grasp the overall evacuation operation status. The mobile terminal application interface pushes the report to the mobile terminals of drivers and dispatchers. Upon receiving the report, the mobile terminals automatically update their local data, supporting real-time viewing and historical queries to meet the information viewing needs of on-site personnel. A data verification mechanism is used during data synchronization. Verification parameter C_1 is set for data integrity verification, checking whether all fields in the report are complete. If no fields are missing, synchronization is considered successful; if data is missing, automatic resynchronization is performed to ensure complete consistency between the data at both ends and the system.

[0255] Through the generation of closed-loop feedback reports and multi-terminal synchronization, the port transportation system has achieved full-process digital management, from multi-source positioning, route planning, and operation monitoring to paperless weighing and data feedback. All operation data is collected in real time, processed automatically, and fed back in a closed loop, eliminating the need for manual filling out of forms and statistical data, thus completely eliminating errors and delays caused by manual operation. At the same time, all data is traceable, statistical, and analyzable, providing precise data support for optimizing port transportation efficiency, adjusting routes, and scheduling vehicles, and comprehensively improving the intelligent and digital management level of the port transportation system.

[0256] This embodiment also provides a port evacuation management method, such as... Figure 6 As shown, the method may include:

[0257] S401 collects real-time location data of port transport vehicles, integrates Beidou positioning and base station positioning information, and combines the port electronic map to construct a structured digital road network model including lane attributes and traffic rules, generating real-time vehicle location data and dynamic road network layers.

[0258] S402 receives real-time vehicle location data and dynamic road network layers. Combining the cargo type, destination, and timeliness requirements of the transport work order, it calculates the optimal transport route through a multi-objective optimization algorithm and automatically assigns work tasks based on vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions.

[0259] S403 receives the dispatch plan and tracks the vehicle's execution trajectory. Based on the digital road network model, it compares the vehicle's position with the driving route in real time. When it detects deviation from the route, speeding, or illegal parking, it triggers an alarm and pushes abnormal event information to the mobile terminal, generating a warning record with a timestamp and violation type.

[0260] The S404 receives warning records and vehicle arrival signals, automatically triggers the electronic weighbridge to complete the weighing, and associates the weighing data with the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information and anomaly statistics. This report is then synchronously updated to the dispatch center and mobile terminals, thereby achieving full-process digital management of the transportation operation.

[0261] Example 7

[0262] The present invention further proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any one of the technical solutions of the port evacuation management method in Embodiment 1, Embodiment 3, Embodiment 4, or Embodiment 6.

[0263] The present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor executes the computer program to implement the steps of any one of the technical solutions of the port evacuation management method in Embodiment 1, Embodiment 3, or Embodiment 4.

[0264] The present invention also proposes a computer program product in which the instructions are executed by the processor of an electronic device, causing the electronic device to perform the steps of any one of the technical solutions for port evacuation management methods in Embodiment 1, Embodiment 3 or Embodiment 4.

[0265] The processor involved in this embodiment can have various specific implementation forms. For example, the processor may include one or more combinations of a central processing unit (CPU), GPU, NPU, TPU, or DPU, etc., and this application embodiment does not impose specific limitations. The processor can also be a single-core processor or a multi-core processor. The processor can be a combination of a CPU and hardware chips. The aforementioned hardware chips can be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. The aforementioned PLDs can be complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), generic array logic (GALs), or any combination thereof. The processor can also be implemented using logic devices with built-in processing logic, such as FPGAs or digital signal processors (DSPs).

[0266] The memory involved in this embodiment can be volatile memory or non-volatile memory, or it can include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DRRAM).

[0267] Embodiments of the present invention include various steps, which will be described below. These steps may be performed by hardware components or may be contained in machine-executable instructions, which may be used by a general-purpose or special-purpose processor programmed with the instructions to perform these steps. Alternatively, the steps may be performed by a combination of hardware, software, and firmware and / or by a human operator. The processor involved in the embodiments of this application may be a chip. For example, it may be a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), a system-on-chip (SoC), a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0268] The present invention can be practiced by combining one or more machine-readable storage media containing code according to the invention with suitable standard computer hardware to execute the code contained therein. Apparatus for implementing various embodiments of the invention may include one or more computers (or one or more processors within a single computer) and a storage system containing or having network access to computer programs encoded according to the various methods described herein, and the method steps of the invention may be performed by modules, routines, subroutines, or sub-parts of a computer program product.

[0269] The system, device, and storage medium in this invention are based on multiple aspects of the same inventive concept as the method in the foregoing embodiments. The implementation process of the method has been described in detail above, so those skilled in the art can clearly understand the structure and implementation process of the system, device, and storage medium in this embodiment based on the foregoing description. For the sake of brevity, it will not be described again here.

[0270] In some embodiments, the system involved in this invention can be configured as a distributed system, wherein one or more components of the system are distributed across one or more networks of a cloud computing system.

[0271] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A port transportation management method, characterized in that, Includes the following steps: Collect port transport vehicle data: Collect real-time location data of port transport vehicles and integrate BeiDou positioning and base station positioning information to generate fused positioning data; Constructing a structured digital road network model for the port: Based on the fused positioning data and the port electronic map, construct a structured digital road network model for the port that includes lane attributes and traffic rules, and output real-time vehicle location data and dynamic road network layers; Planning port transport vehicle routes: Based on real-time vehicle location data, dynamic road network layers, cargo type of transport work orders, destination and timeliness requirements, the optimal transport route is calculated through a multi-objective optimization algorithm, and work tasks are automatically assigned according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions. Monitoring abnormal events in port transportation: Based on the scheduling plan and tracking the vehicle's trajectory, the vehicle's position and driving route are compared in real time based on the port's structured digital road network model. When deviation from the route, speeding, or illegal parking is detected, an alarm is triggered, and a warning record with a timestamp and violation type is generated. Once the vehicle has completed weighing, a closed-loop feedback report is generated: Based on the warning record and the vehicle arrival signal at the weighbridge, the electronic weighbridge is automatically triggered to complete the weighing, and the weighing data is associated with the transportation work order to generate a closed-loop feedback report including task completion status, weighing information and anomaly statistics. Synchronize port transport management data: Synchronize the closed-loop feedback report to the dispatch center database and / or mobile terminal application.

2. The port evacuation management method according to claim 1, characterized in that, The steps for collecting port transport vehicle data include: By using Beidou positioning terminals and base station communication modules installed on port transport vehicles, Beidou satellite positioning data and cellular base station positioning data of port transport vehicles are collected in real time to generate raw positioning data streams; The original positioning data stream is spatiotemporally aligned and fused using Kalman filtering to eliminate multipath effects and positioning jumps caused by base station handover, generating fused positioning data.

3. The port evacuation management method according to claim 2, characterized in that, The steps for constructing a port structured digital road network model include: The fused positioning data is matched with the port electronic map, and the vehicle position is projected onto the nearest lane using a map matching algorithm. Lane boundary, traffic sign and traffic rule information are extracted to generate basic road network data including lane attributes. Based on basic road network data and real-time fused positioning data, the congestion status and traffic speed of each lane are dynamically updated, and a structured digital road network model including lane attributes and traffic rules is constructed. Finally, real-time vehicle location data and dynamic road network layers are output.

4. The port evacuation management method according to claim 1, characterized in that, In the step of planning port transport vehicle routes, the optimal transport route is dynamically adjusted based on real-time traffic conditions and weather conditions, and tasks are assigned to vehicles in the best condition first.

5. The port evacuation management method according to claim 1, characterized in that, In the steps of monitoring abnormal events in port transportation, when a vehicle is detected to have deviated from its route, the offset distance and direction are calculated and analyzed, and the offset information is pushed to the dispatcher's terminal. In the steps of monitoring abnormal events in port transportation, an alarm is triggered when deviation from the route, speeding, or illegal parking is detected, and the abnormal event information is pushed to the mobile terminal.

6. The port evacuation management method according to any one of claims 1-5, characterized in that, The steps for planning port transport vehicle routes include: The task requirement parameter set is generated by analyzing the cargo type, destination coordinates, and time requirements in the transportation work order and combining them with the vehicle's current coordinates and idle status in the real-time vehicle location data. Based on the topology and real-time traffic rules in the dynamic road network layer, the A* algorithm is used to pre-calculate the traffic cost of each candidate path and generate a path cost matrix; the traffic cost includes travel distance, estimated time and energy consumption. The path cost matrix and task requirement parameter set are input into a multi-objective optimization model. The Pareto optimal path set is solved by using a non-dominated sorting genetic algorithm with an elitist strategy. This comprehensively considers timeliness, economy and safety to generate the optimal path sequence. Based on the vehicle's current load capacity, fuel range, and task execution records, and combined with task priority rules, the system automatically assigns tasks to each idle vehicle, ultimately generating a scheduling scheme that includes route guidance and task instructions.

7. A port transportation management method, characterized in that, Includes the following steps: Collect real-time location data of transport vehicles within the port, integrate BeiDou positioning and base station positioning information, and combine with the port's electronic map to construct a structured digital road network model that includes lane attributes and traffic rules, generating real-time vehicle location data and dynamic road network layers; It receives real-time vehicle location data and dynamic road network layers, combines the cargo type, destination and timeliness requirements of the transport work order, calculates the optimal transport route through a multi-objective optimization algorithm, and automatically assigns work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions. It receives dispatch plans and tracks vehicle execution trajectories, compares vehicle positions with driving routes in real time based on digital road network models, and triggers alarms when it detects deviation from the route, speeding, or illegal parking. At the same time, it pushes abnormal event information to mobile terminals and generates warning records with timestamps and violation types. Upon receiving early warning records and vehicle arrival signals, the electronic weighbridge is automatically triggered to complete the weighing process. The weighing data is then linked to the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information, and anomaly statistics. This report is simultaneously updated to the dispatch center and mobile terminals, thereby achieving full-process digital management of the transportation operation.

8. A port transportation management system, characterized in that, include: The data acquisition module is used to collect port transport vehicle data; the data acquisition module collects real-time location data of port transport vehicles and integrates Beidou positioning and base station positioning information to generate fused positioning data. The road network construction module is used to construct a structured digital road network model of the port. Based on the fused positioning data and the port electronic map, the road network construction module constructs a structured digital road network model of the port that includes lane attributes and traffic rules, and outputs real-time vehicle location data and dynamic road network layers. The route planning module is used to plan the routes of port transport vehicles. The route planning module receives real-time vehicle location data and dynamic road network layers, combines the cargo type, destination and timeliness requirements of the transport work order, calculates the optimal transport route through a multi-objective optimization algorithm, and automatically assigns work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions. The anomaly monitoring module is used to monitor abnormal events in port transportation. The anomaly monitoring module receives the scheduling plan and tracks the vehicle's execution trajectory. Based on the port's structured digital road network model, it compares the vehicle's position with the driving route in real time. When it detects deviation from the route, speeding, or illegal parking, it triggers an alarm and generates a warning record with a timestamp and violation type. The weighing management module is used to generate a closed-loop feedback report after a vehicle completes weighing. The weighing management module receives the warning record and the vehicle arrival signal at the weighbridge, automatically triggers the electronic weighbridge to complete the weighing, and associates the weighing data with the transportation work order to generate a closed-loop feedback report including task completion status, weighing information and anomaly statistics. The data synchronization module is used to synchronize port transportation management data; the data synchronization module synchronizes the closed-loop feedback report to the dispatch center database and / or mobile terminal application.

9. A port transportation management system, characterized in that, include: The multi-source positioning fusion and digital road source construction module is used to collect real-time location data of transport vehicles in the port, integrate Beidou positioning and base station positioning information, and combine them with the port electronic map to construct a structured digital road network model containing lane attributes and traffic rules, generating real-time vehicle location data and dynamic road network layers. The dynamic route planning and intelligent task assignment module is used to receive real-time vehicle location data and dynamic road network layers, combine the cargo type, destination and timeliness requirements of the transport work order, calculate the optimal transport route through a multi-objective optimization algorithm, and automatically assign work tasks according to vehicle status and task priority, generating a scheduling scheme that includes route guidance and task instructions. The operation process monitoring and anomaly early warning module is used to receive dispatch plans and track vehicle execution trajectories. Based on the digital road network model, it compares the vehicle position and driving route in real time. When it detects deviation from the route, speeding, or illegal parking, it triggers an alarm and pushes abnormal event information to the mobile terminal, generating an early warning record with timestamp and violation type. The paperless weighing and data closed-loop feedback module is used to receive early warning records and vehicle arrival signals, automatically trigger the electronic weighbridge to complete the weighing, and associate the weighing data with the transportation work order to generate a closed-loop feedback report containing task completion status, weighing information and anomaly statistics, which is synchronously updated to the dispatch center and mobile terminals, thereby realizing full-process digital management of transportation operations.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the steps of the port evacuation management method as described in any one of claims 1 to 7.

Images