Cargo transportation path planning and tracking method

Abstract

The invention relates to a cargo transportation path planning and tracking method. The method comprises the following steps: acquiring initial target intention data, transportation environment state data and an original transportation path evolution model of a cargo transportation task; according to the initial target intention data and the transportation environment state data, constructing an initial transportation planning path of the cargo transportation task; performing parameter initialization on the original transportation path evolution model according to the initial transportation planning path to obtain a current transportation path evolution model; performing target migration processing on the initial target intention data according to real-time environment sensing data of the cargo transportation task to obtain target migration intention data; inputting the target migration intention data into the current transportation path evolution model to obtain an updated transportation planning path; and tracking the execution condition of the cargo transportation task according to the updated transportation planning path. The method can effectively cope with complex and changeable transportation environments.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method for planning and tracking cargo transportation routes. Background Technology

[0002] In traditional technologies, freight transportation route planning typically relies on human experience or static algorithms based on historical data (such as shortest path and minimum cost flow) to determine transportation routes. This is combined with GPS and GIS systems for location data collection and tracking, and transportation status monitoring is achieved by periodically uploading location information to a central server. During transportation, dispatchers intervene manually and adjust routes according to fixed rules and pre-set contingency plans to address unforeseen events. However, traditional route planning and tracking technologies lack real-time performance and dynamic optimization capabilities, making it difficult to cope with complex and ever-changing transportation environments. Summary of the Invention

[0003] Therefore, it is necessary to provide a method, apparatus, and computer equipment for cargo transportation route planning and tracking that can effectively cope with complex and ever-changing transportation environments, addressing the aforementioned technical problems.

[0004] Firstly, this application provides a method for cargo transportation route planning and tracking, including:

[0005] Acquire initial target intent data, transportation environment status data, and original transportation path evolution model for cargo transportation tasks;

[0006] Based on the initial target intent data and the transportation environment status data, an initial transportation planning path for the cargo transportation task is constructed;

[0007] Based on the initial transportation planning path, the initialization parameters of the original transportation path evolution model are set to obtain the current transportation path evolution model;

[0008] Based on the real-time environmental perception data of the cargo transportation task, the initial target intent data is processed to obtain target migration intent data.

[0009] The target migration intention data is input into the current transportation path evolution model to obtain an updated transportation planning path;

[0010] The execution status of the cargo transportation task is tracked based on the updated transportation planning route.

[0011] Secondly, this application also provides a cargo transportation route planning and tracking device, comprising:

[0012] The transportation data acquisition module is used to acquire the initial target intent data, transportation environment status data, and original transportation path evolution model of the cargo transportation task.

[0013] The transportation route planning module is used to construct the initial transportation planning route for the cargo transportation task based on the initial target intent data and the transportation environment status data.

[0014] The path model initialization module is used to set the initialization parameters of the original transportation path evolution model according to the initial transportation planning path, so as to obtain the current transportation path evolution model.

[0015] The target migration processing module is used to perform target migration processing on the initial target intent data based on the real-time environmental perception data of the cargo transportation task, so as to obtain target migration intent data.

[0016] The transportation route update module is used to input the target migration intention data into the current transportation route evolution model to obtain an updated transportation planning route;

[0017] The transportation task tracking module is used to track the execution status of the cargo transportation task based on the updated transportation planning route.

[0018] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of a cargo transportation route planning and tracking method.

[0019] The aforementioned cargo transportation route planning and tracking method, apparatus, and computer equipment generate an initial transportation planning route that meets task requirements and environmental constraints by fusing acquired initial target intent data, transportation environment status data, and the original transportation route evolution model. The initial transportation planning route is then used to initialize the parameters of the route evolution model, ensuring high initial adaptability. During transportation, real-time environmental perception data is incorporated to perform target migration processing on the initial target intent, enabling the transportation plan to reflect changes in task objectives and environmental conditions in a timely manner. The target migration intent data is input into the current route evolution model to generate an updated transportation planning route that adapts to the latest environmental and task requirements. The task execution process is continuously tracked, achieving dynamic optimization and closed-loop management of the transportation route. This effectively addresses complex and ever-changing transportation environments, significantly improving the flexibility, real-time performance, accuracy, and execution efficiency of route planning in such environments, effectively reducing transportation risks and increasing the overall reliability and success rate of the task. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a diagram illustrating the application environment of a cargo transportation route planning and tracking method in one embodiment.

[0022] Figure 2 This is a flowchart illustrating a cargo transportation route planning and tracking method in one embodiment;

[0023] Figure 3 This is a flowchart illustrating a method for updating transportation planning routes in one embodiment;

[0024] Figure 4 This is a flowchart illustrating a path structure family generation method in one embodiment;

[0025] Figure 5 This is a flowchart illustrating the method for obtaining the interaction path ecosystem in one embodiment;

[0026] Figure 6 This is a flowchart illustrating a method for obtaining fitness analysis data in an embodiment.

[0027] Figure 7 This is a flowchart illustrating the method for adjusting the path ecosystem in one embodiment;

[0028] Figure 8 This is a flowchart illustrating a method for obtaining transportation adaptability analysis data in one embodiment;

[0029] Figure 9 This is a flowchart illustrating a method for obtaining target migration intent data in one embodiment. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0031] The cargo transportation route planning and tracking method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other network servers. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0032] In one exemplary embodiment, such as Figure 2 As shown, a method for cargo transportation route planning and tracking is provided, which can be applied to... Figure 1 Taking the server in the example, the explanation includes the following steps 202 to 212. Wherein:

[0033] Step 202: Obtain the initial target intent data, transportation environment status data, and original transportation path evolution model for the cargo transportation task.

[0034] Step 204: Based on the initial target intent data and transportation environment status data, construct the initial transportation planning path for the cargo transportation task.

[0035] Step 206: Based on the initial transportation planning path, set the initialization parameters of the original transportation path evolution model to obtain the current transportation path evolution model.

[0036] Step 208: Based on the real-time environmental perception data of the cargo transportation task, perform target migration processing on the initial target intent data to obtain target migration intent data.

[0037] Step 210: Input the target migration intention data into the current transportation path evolution model to obtain the updated transportation planning path.

[0038] Step 212: Track the execution of cargo transportation tasks based on the updated transportation planning route.

[0039] Among them, the cargo transportation task is a comprehensive transportation execution instruction generated by the transportation dispatch system under specific time, space and operating conditions, which includes cargo information, origin and destination, delivery time limit, transportation mode and constraints.

[0040] The initial target intent data is a set of multi-dimensional transportation target parameters proposed by the task demand side when the cargo transportation task is generated. These parameters include, but are not limited to, indicators such as delivery time priority, transportation cost constraints, risk tolerance, service level requirements, and environmental compliance. They are used to reflect the comprehensive target preferences of the task in the route planning stage.

[0041] Among them, transportation environment status data is static and dynamic environmental information related to the transportation network and external conditions, which is obtained during the route planning stage. This includes road network topology, road segment capacity, meteorological conditions, regional risk level, and regulatory restrictions, providing environmental constraints for constructing the initial transportation planning route.

[0042] The original transportation path evolution model is a computational model pre-built based on historical operational data, transportation network structure, and transportation behavior characteristics. It includes attribute information of path nodes and edges, transportation cost and risk assessment functions, resource allocation strategies, and path evolution rules, and is used to simulate and optimize transportation paths under different objectives and environmental conditions.

[0043] The initial transportation planning path is the preferred path among the set of optimal or second-best transportation paths under the current conditions, calculated by a multi-objective optimization algorithm during the task planning stage based on the initial target intent data and transportation environment status data.

[0044] The current transportation path evolution model is a path evolution calculation model that matches the current task objectives and environmental conditions. It is obtained by initializing and adjusting the structure and parameters of the initial transportation planning path based on the original transportation path evolution model.

[0045] Real-time environmental perception data refers to dynamic environmental information collected during the execution of transportation tasks through IoT sensors, traffic monitoring systems, and external information platforms. This includes real-time traffic flow status, emergency alerts, weather changes, road closure information, and abnormal node passage.

[0046] Among them, target migration processing is a process of dynamically adjusting the target parameters during task execution based on the changes in adaptability between real-time environmental perception data and initial target intent data. Its core lies in making predictive corrections to the target dimensions affected by the environment while keeping non-critical targets stable.

[0047] Among them, the target migration intent data is a set of multi-dimensional target parameters obtained after target migration processing. It can better balance the constraints of time, cost, and risk of transportation tasks under the current environmental conditions and is used as input to the path evolution model to drive the dynamic update of the path structure.

[0048] Among them, the updated transportation planning path is the latest path scheme calculated based on the target migration intention data and real-time environmental factors in the current transportation path evolution model. It can dynamically replace the initial planned path during transportation to improve transportation efficiency, reduce risks, and ensure the achievement of mission objectives.

[0049] Specifically, after the server 104 receives the cargo transportation task through the terminal 102, it controls the task data acquisition sensor in the terminal 102 to collect initial target intent data related to the task, including indicators such as delivery time requirements, transportation cost ceiling, cargo category and transportation safety level; at the same time, it calls the environmental perception platform interface to obtain macro traffic conditions (such as road capacity, congestion index), meteorological conditions and potential regional risk information along the transportation route in real time, forming transportation environment status data; and loads the original transportation path evolution model matching the current task from the transportation database or model warehouse. This model has pre-established the transportation path network topology, node attributes, path cost function and historical operation rules.

[0050] In the initial route planning engine, the initial target intent data is used as the optimization constraint, and the transportation environment status data is used as the route cost correction factor to perform multi-objective optimization calculations on the transportation network. The optimization process simultaneously considers multiple trade-offs such as time priority, cost minimization, and risk avoidance. A combination strategy of the A-search algorithm and the multi-scale weighted Dijkstra algorithm is used to generate a set of candidate paths that meet the constraints. Finally, the path with the highest score is selected as the initial transportation planning path through a comprehensive scoring function.

[0051] The node sequence, path segment attributes, and corresponding runtime parameters of the initial transportation planning path are input into the original transportation path evolution model. Based on the geometric features and operating conditions of the path, the path cost matrix, node service capacity, resource allocation priority, and other parameters in the original transportation path evolution model are initialized. At the same time, the historical operating data under the initial transportation planning path is fused with the current environmental data to correct the prediction bias of the model and obtain the current transportation path evolution model that is completely matched with the current task and environmental state.

[0052] During the execution of the cargo transportation task, the server 104 continuously collects real-time environmental perception data through instructions to control the terminal 102 in the transportation process, including information on road emergencies, weather changes, and local traffic flow fluctuations. This real-time environmental perception data is then aligned with the initial target intent data in multiple dimensions, and the target fitness change rate at different scales (macro, meso, and micro) is calculated. When the change rate exceeds a set threshold, a target migration mechanism is triggered to perform multi-scenario prediction and correction on the affected target dimensions (such as time, cost, or risk tolerance), generate a set of migration candidate targets, and finally select the optimal target according to priority to form target migration intent data.

[0053] The target migration intention data is mapped to the migration perturbation input of the current transportation path evolution model, and the path structure, resource allocation rules and node priority parameters of the current transportation path evolution model are dynamically adjusted. After receiving the perturbation, the current transportation path evolution model performs multi-scale adaptive evolution to generate a new path structure family. The transportation adaptability of these new paths in different environments is evaluated through the built-in multi-scenario simulation module, and the best-performing path is selected as the updated transportation planning path based on the adaptability score.

[0054] The updated transportation planning route is sent to the task execution module, and during task execution, the vehicle's location, speed, load status, and route deviation are monitored in real time through IoT terminals, vehicle positioning systems, and sensor networks. When operational behavior that does not conform to the updated transportation planning route or a sudden anomaly is detected, an early warning is immediately generated and the emergency adjustment module is triggered. At the same time, the actual operation data is continuously integrated into the real-time environmental perception data and then sent back to the current transportation route evolution model to provide closed-loop data support for subsequent route optimization and task evaluation.

[0055] In the aforementioned cargo transportation route planning and tracking method, an initial transportation planning route that meets task requirements and environmental constraints is generated by integrating the acquired initial target intent data, transportation environment status data, and the original transportation route evolution model. This initial transportation planning route is then used to initialize the parameters of the route evolution model, ensuring high initial adaptability. During transportation, real-time environmental perception data is used to perform target migration processing on the initial target intent, enabling the transportation plan to reflect changes in task objectives and environmental conditions in a timely manner. The target migration intent data is then input into the current route evolution model to generate an updated transportation planning route that adapts to the latest environment and task requirements. The task execution process is continuously tracked, achieving dynamic optimization and closed-loop management of the transportation route. This effectively addresses complex and ever-changing transportation environments, significantly improving the flexibility, real-time performance, accuracy, and execution efficiency of route planning in such environments, effectively reducing transportation risks and increasing the overall reliability and success rate of the task.

[0056] In one exemplary embodiment, such as Figure 3 As shown, the target migration intention data is input into the current transportation path evolution model to obtain the updated transportation planning path, including steps 302 to 308. Wherein:

[0057] Step 302: Map the target migration intention data to the multidimensional path evolution topology space of the cargo transportation task and construct the migration intention perturbation function.

[0058] Step 304: Based on the migration intention perturbation function, drive the path structure in the current transportation path evolution model to undergo multi-scale self-fractal evolution to generate a path structure family.

[0059] Step 306: Perform simulation analysis on the path structure family in the current transportation path evolution model environment to obtain transportation adaptability analysis data.

[0060] Step 308: Based on the transportation adaptability analysis data, extract the trend path structure from the evolutionary stable region of the current transportation path evolution model to obtain the updated transportation planning path.

[0061] Among them, the multidimensional path evolution topology space is a high-dimensional topology structure that comprehensively models the freight transportation network in multiple dimensions such as time, space, cost, risk, and resources. In this space, nodes represent transportation nodes or hubs, and edges represent transportation paths or channels. Each dimension carries corresponding attribute parameters, thereby supporting multidimensional synchronous optimization and dynamic adjustment of paths at different evolution stages.

[0062] Among them, the migration intention perturbation function is a dynamic function constructed based on the difference between the target migration intention data and the current transportation path state. It is used to apply directional driving or inhibiting effects during the path evolution process, thereby guiding the path structure to shift towards the expected target state in the multi-dimensional path evolution topology space, and can generate perturbation effects of different intensities and frequencies at multiple scales.

[0063] In the current transportation path evolution model, the path structure is the transportation network form composed of specific node connection relationships, edge weight distribution, flow allocation and resource parameters. It includes macro-level path layout patterns, meso-level regional connection methods and micro-level local node organization characteristics.

[0064] Among them, multi-scale self-fractal evolution is a path evolution method based on fractal geometry and multi-scale network growth mechanism. By applying similar but proportionally different topological growth rules at different scales such as macro, meso and micro, the path network can maintain self-similarity in both local details and overall structure, and can form multi-level mutation and optimization results under perturbation drive.

[0065] Among them, the path structure family is a set of multiple candidate path structures that are differentiated in terms of topological features, resource distribution and operational performance, generated by multi-scale self-fractal evolution.

[0066] Among them, the transportation adaptability analysis data is a set of data that records and statistically analyzes the operational performance of a route structure family under different types and intensities of disturbances in a multi-field dynamic simulation environment. It usually includes multi-dimensional indicators such as transportation timeliness, cost consumption, risk exposure, and resource utilization efficiency, and is used to evaluate the comprehensive adaptability of the route structure.

[0067] Among them, the evolutionary stable region is the path state space in the current transportation path evolution model that can maintain the performance index fluctuating within a set threshold range after multiple rounds of disturbance simulation and adaptive evaluation. The path structure in this region has high stability and sustainability in the face of environmental changes.

[0068] Among them, the trend path structure is extracted from the evolutionary stable region, which shows consistently excellent performance under different time periods and perturbation scenarios, and shows a stable or optimized trend path structure in the adaptive analysis.

[0069] Specifically, in the current transportation path evolution model, the multidimensional target parameters (including time, cost, risk, service level, etc.) of the target migration intention data are mapped to the attribute vectors of the corresponding nodes and edges in the multidimensional path evolution topology space according to the weights and constraints of each dimension; and a quantifiable migration intention perturbation function is generated by calculating the deviation value and direction between the intention parameters and the existing path state in the mapped parameters.

[0070] By using the migration intention perturbation function, the current transportation path structure is perturbed at three scales: macro, meso, and micro. This causes the path network to generate multi-level and multi-variable path structures according to fractal growth rules in terms of topology, connection density, and resource distribution. Local variant path sets are generated at each scale and combined into diverse path structure families during the cross-scale fusion stage.

[0071] The path structure family is placed in a dynamic simulation environment that includes multiple coupled fields such as real-time weather, traffic flow and sudden event disturbances. Different types of transportation units are simulated on each path through heterogeneous agents. During the simulation, data such as the running efficiency, resource consumption and risk exposure of the path are recorded under different time periods and disturbance conditions. These data are classified and statistically analyzed according to the type and intensity of disturbance to form a transportation adaptability analysis dataset.

[0072] Based on transportation adaptability analysis data, path structures exhibiting high stability, high efficiency, and low risk under different disturbance scenarios are identified, and the long-term feasibility trend of these paths in the evolutionary stable region is determined. Based on the trend analysis results, the path structure with the best comprehensive performance is selected as the updated transportation planning path for the current transportation task and submitted to the execution module for real-time replacement and task scheduling.

[0073] In this embodiment, by mapping target migration intention data to a multi-dimensional path evolution topology space and constructing a migration intention perturbation function, the path evolution process can simultaneously respond to target changes at multiple scales (macro, meso, and micro). This drives the current transportation path evolution model to undergo multi-scale self-fractal evolution and generate a group of path structures with differentiated characteristics. Furthermore, the transportation adaptability under various perturbation scenarios is evaluated in simulation analysis, and trend path structures are extracted from the evolutionary stable region as updated planning paths. This achieves dynamic optimization and precise selection of path schemes, which can significantly improve the robustness, flexibility, and operational efficiency of the transportation network in complex and ever-changing environments.

[0074] In one exemplary embodiment, such as Figure 4 As shown, based on the migration intention perturbation function, the path structure in the current transportation path evolution model is driven to undergo multi-scale self-fractal evolution to generate a path structure family, including steps 402 to 412.

[0075] in:

[0076] Step 402: Based on the current transportation path evolution model, construct the initial path ecosystem in different scale environments.

[0077] Step 404: Based on the migration intention perturbation function and real-time environmental perception data, dynamically adjust the resource density parameters and resource competition coefficients of each initial path ecosystem to obtain each adjusted path ecosystem.

[0078] Step 406: Perform resource competition analysis on each adjusted path ecosystem to obtain the corresponding ecosystem fitness analysis data.

[0079] Step 408: Based on the fitness analysis data of each ecosystem, perform cross-scale information interaction processing on each adjustment path ecosystem to obtain each interaction path ecosystem.

[0080] Step 410: If any interactive path ecosystem has not reached a stable state, return to the step of performing resource competition analysis on each adjusted path ecosystem to obtain the corresponding ecosystem fitness analysis data, until each interactive path ecosystem reaches a stable state and obtains each stable path ecosystem.

[0081] Step 412: Identify paths from each stable path ecological field whose morphological difference information and ecological fitness difference information meet the preset thresholds, and use them as path structure groups.

[0082] Among them, the initial path ecosystem is a virtual ecological space formed by environmental abstraction and mathematical modeling of path structure and its resource distribution, traffic flow density, key node status and other elements in a transportation network environment at a specific scale. It is used to characterize the interaction between the path and the environment.

[0083] Among them, the resource density parameter is used to quantify the spatial concentration of available transportation resources (such as transport capacity, traffic capacity, and priority) in the path ecosystem. It reflects the abundance or scarcity of resources in a unit space and directly affects the availability and stability of the path in competition.

[0084] Among them, the resource competition coefficient is a coefficient used to characterize the intensity of competition for limited resources between different paths or nodes in the path ecosystem. The higher the value, the more intense the resource competition, which may lead to more frequent conflicts and concessions between paths.

[0085] Among them, the adjusted path ecosystem is an updated ecosystem obtained by dynamically correcting its resource density parameters and resource competition coefficients based on the migration intention perturbation function and real-time environmental perception data, on the basis of the initial path ecosystem. It can more accurately reflect the impact of real-time environmental changes on the path.

[0086] Among them, resource competition analysis is a process of resource allocation and competition among various paths in the adjusted path ecosystem through multidimensional dynamic game or simulation calculation methods, thereby obtaining the survivability, efficiency and stability indicators of the paths under the current competitive conditions.

[0087] Among them, the ecological field fitness analysis data is a multi-dimensional data set generated after resource competition analysis to quantify the overall adaptability of the path ecological field. It typically includes indicators such as resource utilization efficiency, conflict dissipation speed, risk resilience, and structural diversity.

[0088] Cross-scale information interaction processing is a process of bidirectional information transmission and adjustment between different scales of path ecosystems, such as macro, meso, and micro. By sharing key adaptive indicators and environmental feedback, it optimizes the resource allocation and structural layout of paths at various scales.

[0089] Among them, the interactive path ecosystem is a fusion path ecosystem formed after cross-scale information interaction processing. It includes resource distribution and competition patterns of multi-scale collaborative optimization and can simultaneously reflect the global, regional and local operational characteristics.

[0090] Among them, the stable state is the state in which the resource allocation, competition relationship and structural evolution of the path ecosystem tend to converge in time and no longer fluctuate significantly, which means that the ecosystem has reached dynamic equilibrium or long-term sustainable operation conditions.

[0091] Among them, a stable path ecosystem is a path ecosystem that has reached a stable state, in which its resource distribution, competition structure, and fitness indicators remain within a relatively constant range.

[0092] Among them, morphological difference information is data describing the differences in structural morphology of different paths, including path length, node distribution pattern, topological connectivity, redundancy, etc., which is used to measure the geometric and structural diversity of paths.

[0093] Among them, the ecological adaptability difference information is the difference data of different paths in terms of adaptability indicators, including resource utilization efficiency, risk resilience, conflict dissipation speed, etc., which is used to measure the diversity of paths in terms of environmental adaptability.

[0094] Among them, the preset threshold is a numerical boundary condition set in advance during the path selection and decision-making process. It is used to determine whether the morphological difference information and ecological fitness difference information meet the requirements, so as to select a set of paths that meet the conditions.

[0095] Specifically, based on the current transportation path evolution model, according to different scales of macro (overall transportation network), meso (regional transportation network) and micro (nodes and local paths), the resource distribution, traffic flow density, key bottlenecks and other elements of the path network are abstracted to form an initial path ecosystem corresponding to different scales and having resource status, competitive relationships and topological characteristics.

[0096] By combining the migration intention perturbation function (reflecting the direction and intensity of the intervention of the change of target intention on the path evolution) with real-time environmental perception data (real-time traffic conditions, weather, emergencies, etc.), the resource density parameters (resource distribution concentration) and resource competition coefficients (competition intensity between path nodes) in each initial path ecosystem are dynamically corrected, thereby obtaining each adjusted path ecosystem that can reflect the changes in the real-time environment and target.

[0097] In each adjusted path ecosystem, the competition process for limited transportation resources (capacity, traffic capacity, priority, etc.) between different paths is simulated based on a multidimensional dynamic game model. Indicators such as resource utilization efficiency, conflict dissipation speed, and risk resistance are calculated, and the ecosystem fitness analysis data of the ecosystem is generated in combination to measure its survival and evolution potential under the current conditions.

[0098] By utilizing fitness analysis data of ecosystems at different scales, bidirectional information transmission and feedback can be carried out between macro, meso, and micro scales. For example, macro bottleneck information can be transmitted to the meso scale to optimize regional allocation, and efficient path patterns at the micro scale can be fed back to the meso and macro scales, forming cross-scale coupling and optimization, and ultimately generating ecosystems of various interactive paths that integrate multi-level information.

[0099] The resource allocation and competition status of each interaction path ecosystem are continuously monitored. If any of the interaction path ecosystems fails to reach a stable state, i.e., if there is excessive resource fluctuation or the competition relationship does not converge, the process returns to the step of "conducting resource competition analysis on each adjustment path ecosystem to obtain the corresponding ecosystem fitness analysis data" for iterative correction until all interaction path ecosystems at all scales reach a stable state, resulting in stable path ecosystems at each scale that can be used for long-term prediction and trend analysis.

[0100] Morphological differences (path structure, layout characteristics, etc.) and ecological adaptability differences (resource utilization efficiency, risk resilience, etc.) were analyzed for paths in each stable path ecosystem. The set of paths that met the preset standards in both differences and adaptability was selected as the final path structure group.

[0101] In this embodiment, by constructing an initial path ecosystem under different scale environments and dynamically adjusting resource density parameters and resource competition coefficients in conjunction with migration intention perturbation functions and real-time environmental perception data, real-time adaptation of the path ecosystem is achieved. Multidimensional ecosystem fitness data is obtained through resource competition analysis and gradually tends towards a stable state in the closed-loop iteration of cross-scale information interaction processing, effectively ensuring the collaborative optimization of the path structure at global, regional, and local multi-scale levels. Finally, the path structure family obtained by screening based on preset thresholds of morphological differences and ecological fitness differences can simultaneously take into account diversity and high adaptability, significantly improving the stability, robustness, and overall operational efficiency of transportation paths in complex dynamic environments.

[0102] In one exemplary embodiment, such as Figure 5 As shown, based on the fitness analysis data of each ecosystem, cross-scale information interaction processing is performed on each adjustment path ecosystem to obtain each interaction path ecosystem, including steps 502 to 516. Wherein:

[0103] Step 502: Based on the fitness analysis data of each ecosystem, construct a three-layer bidirectional fitness mapping network; the three-layer bidirectional fitness mapping network includes macro-macro mapping function, mid-macro mapping function, mid-micro mapping function, micro-mid mapping function, and macro-micro direct connection mapping function.

[0104] Step 504: Based on the macro-mapping function, the resource density and competition coefficient of the meso-path ecosystem are adjusted using the resource density, global bottleneck, and risk distribution of the macro-path ecosystem to obtain the rule-reconstructed ecosystem.

[0105] Step 506: Based on the meso-macro mapping function, the global resource allocation of the macro-path ecosystem is adjusted using the transportation efficiency pattern and fitness trend of the meso-path ecosystem to obtain the global resource adjustment ecosystem.

[0106] Step 508: Based on the meso-micro mapping function, the resource fluctuations and flow changes of the meso-path ecosystem are used to adjust the partial resource utilization strategy of the micro-path ecosystem to obtain a local resource-optimized ecosystem.

[0107] Step 510: Based on the micro-meta-macro mapping function, the fitness path pattern and resource utilization efficiency of the micro-path ecosystem are used to adjust the node resources and competition priorities of the meso-path ecosystem to obtain the node resource adjustment ecosystem.

[0108] Step 512: Based on the macro-micro direct connection mapping function, the fitness data and resource data of the macro-path ecosystem and the micro-path ecosystem are used for synchronization processing to obtain the macro-synchronous ecosystem and the micro-synchronous ecosystem.

[0109] Step 514 involves conflict coordination processing of the rule reconstruction ecosystem, global resource adjustment ecosystem, local resource optimization ecosystem, node resource adjustment ecosystem, macro-synchronization ecosystem, and micro-synchronization ecosystem to obtain the ecosystems of each interaction path.

[0110] Among them, the macro-path ecosystem is a dynamic distribution model of resources and paths established at the national or regional transportation network level. It focuses on the overall transportation pattern, long-distance transportation corridors, cross-regional transportation bottlenecks, and the distribution of systemic risks, and is used to guide transportation optimization at the global level.

[0111] Among them, the meso-level path ecosystem is modeled at the regional or urban agglomeration scale, reflecting the regional transportation nodes, trunk resource distribution, inter-regional competitive relationships, and medium-range risks and event impacts. It is an intermediate layer model connecting macro and micro level information.

[0112] Among them, the micro-path ecosystem is modeled at the scale of a single node, a single line, or a local road segment, focusing on the occupation and release of local resources, real-time traffic flow status, changes in local path patterns, and the impact of minor disturbances between nodes.

[0113] Among them, the three-layer bidirectional fitness mapping network is an information interaction structure that spans the macro, meso, and micro levels. It supports the bidirectional transmission and dynamic adjustment of fitness and resource data at different scales, ensuring that each layer of the ecosystem can be optimized in a coordinated manner when it changes.

[0114] Among them, the macro-meta-macro mapping function is used to convert the global characteristics of the macro-path ecosystem (such as resource density, bottleneck distribution, and risk level) into the parameter adjustment signal function of the meso-path ecosystem, so that the regional level can quickly adapt to macro changes.

[0115] Among them, the meso-macro mapping function is a signal function used to feed back the regional transportation efficiency pattern and fitness trend of the meso-path ecosystem to the macro-path ecosystem, thereby realizing the dynamic correction of global resource allocation.

[0116] Among them, the meso-micro mapping function is a signal function used to transmit information on resource fluctuations and flow changes in the meso-path ecosystem to the micro-path ecosystem, thereby driving the real-time optimization of local nodes and routes.

[0117] Among them, the micro-meta-macro mapping function is a signal function used to transmit fitness path patterns and resource utilization efficiency information in the micro-path ecosystem back to the meso-path ecosystem, and adjust the resource allocation and competition priority of regional nodes.

[0118] Among them, the macro-micro direct connection mapping function is a signal function used to directly synchronize fitness and resource data between the macro-path ecosystem and the micro-path ecosystem, bypassing the meso-level to achieve a fast global-local response.

[0119] Resource density refers to the degree of concentration of available transportation resources (such as transport capacity, vehicles, and warehousing) within a specific spatial range or route unit, and is a parameter for measuring the availability of resources in the route ecosystem.

[0120] Among them, the global bottleneck is the key node or section in the macro transportation network that restricts the overall transport capacity. Its congestion or failure will significantly reduce the overall transportation efficiency.

[0121] Risk distribution refers to the spatial distribution of probabilities of different path units or regions encountering risks such as natural disasters, traffic accidents, and emergencies within a certain period of time.

[0122] The competition coefficient is used to measure the intensity of resource competition between different transportation paths or nodes. A high competition coefficient means more conflicts and preemptive actions.

[0123] Among them, the rule-reconstructed ecosystem is a meso-path ecosystem formed after the adjustment of the macro-mapping function. Its structure and parameters have been optimized and reconstructed by macro-conditions, making it more in line with global transportation rules and resource layout.

[0124] Among them, the transportation efficiency model refers to the dynamic pattern of transportation speed, loading and unloading efficiency, vehicle turnover rate, etc. in a region or node within a certain period.

[0125] Among them, fitness trend refers to the direction of change in the survival and performance capabilities of a path or node in a changing environment, and is used to predict the sustainability of resource allocation.

[0126] Among them, global resource allocation is a holistic planning and scheduling strategy for transportation tasks and resources in different regions by the macro-path ecosystem.

[0127] Among them, the global resource adjustment ecosystem is the macroscopic path ecosystem after the action of the macro-macro mapping function, and its resource allocation strategy is optimized and updated after regional feedback.

[0128] Resource fluctuation refers to the dynamic fluctuations in the amount of available transportation resources on a path or at a node within a certain period of time.

[0129] Among them, flow direction change refers to the change in the direction pattern of transportation resources (such as vehicle flow and freight flow) in the path network, including direction adjustment and flow redistribution.

[0130] Among them, the local resource utilization strategy is a micro-level scheme for the use and scheduling of resources to ensure the maximization of local transportation operation efficiency.

[0131] Among them, the local resource optimization ecosystem is a micro-path ecosystem after the adjustment of the micro-mapping function, and its resource use scheme has been optimized for local changes.

[0132] Among them, the fitness path pattern is the path combination and operation mode in the micro-path ecosystem that can maintain high adaptability and high efficiency in the current environment.

[0133] Resource utilization efficiency refers to the effective output generated by a unit of resources in a transportation task, reflecting the degree of matching between resource input and output.

[0134] Among them, node resources are the total amount of resources such as transportation capacity, warehousing and operating equipment that a single node (such as a port, logistics center, or hub station) in the transportation network can control.

[0135] Among them, competition priority is the priority ranking given to regions or nodes in the competition for resources, which directly affects resource allocation decisions.

[0136] Among them, the node resource adjustment ecosystem is the meso-path ecosystem after the micro-macro mapping function is adjusted, and its node resource allocation has been optimized based on micro-feedback.

[0137] Fitness data is a quantitative result that measures the performance of a path or node in a specific environment, including indicators such as efficiency, stability, and risk tolerance.

[0138] Resource data is a comprehensive description of the quantity, location, status, and other information of various transportation resources in the path ecosystem.

[0139] Among them, the macro-synchronous ecosystem field is the macro-path ecosystem field after the macro-micro direct connection mapping function is synchronously adjusted, and its global parameters are kept in real time consistent with micro-changes.

[0140] Among them, the micro-synchronous ecological field is the micro-path ecological field after the macro-micro direct connection mapping function is synchronously adjusted, and its local operation strategy is synchronized with macro-control.

[0141] Among them, conflict coordination and handling is the process of identifying, assessing and resolving conflicts between different scale ecosystems in terms of resource allocation, path selection and priority setting, so as to ensure the stable operation of the system.

[0142] Specifically, after identifying the corresponding fitness analysis data at the macro, meso, and micro levels in the fitness analysis data of each of the aforementioned ecological fields, a three-layer bidirectional fitness mapping network is established, consisting of macro-meso mapping functions, meso-macro mapping functions, meso-micro mapping functions, micro-meso mapping functions, and macro-micro direct connection mapping functions.

[0143] At the macro-level of the path ecosystem, the core dataset reflecting the overall transportation pattern is first extracted, including a global resource density distribution map, spatial location information of major transportation bottlenecks, and multi-dimensional statistical characteristics of risk distribution. Using a macro-macro mapping function, this macro data is mapped and transformed into parameter inputs that can be directly applied to the meso-level path ecosystem. The focus is on numerical optimization and structural reconstruction of the resource density parameter matrix and competition coefficient matrix. This process comprehensively considers the long-term impact of macro-trend changes on regional transportation patterns, enabling meso-level resource allocation strategies and competitive relationships to adapt to macro-level changes. This results in a more resilient and adaptable meso-level path ecosystem—a rule-reconstructed ecosystem.

[0144] In the meso-level path ecosystem, dynamic indicators characterizing changes in regional transportation efficiency and trend evolution data of cross-regional transportation adaptability are monitored and extracted in real time. Through a meso-macro mapping function, these local and regional data at the meso-level are fed back to the macro-level path ecosystem for analysis and reallocation optimization at the global transportation network resource scheduling level. During the optimization process, the distribution ratio, priority, and allocation order of global resources are adjusted based on inter-regional efficiency differences and trend changes, thereby ensuring that the macro-level transportation resource allocation scheme is more in line with the actual operation and development trends of each region, forming a global resource adjustment ecosystem.

[0145] In the meso-level path ecosystem, based on the characteristics of transportation resource fluctuations and traffic flow trends in different regions, a meso-micro mapping function is used to precisely distribute this dynamic data to the micro-level path ecosystem. Upon receiving this data at the micro level, it is applied to the resource scheduling models of specific transportation nodes and local routes, performing fine-grained optimizations on resource allocation ratios, operational sequences, and route switching strategies. This maximizes the operational efficiency and response speed of local transportation units, ensuring that the micro-level maintains a high level of capacity matching and service quality even when facing traffic fluctuations or changes in direction, ultimately forming a local resource optimization ecosystem.

[0146] In the micro-path ecosystem, transportation path patterns exhibiting high fitness levels and node data demonstrating excellent resource utilization efficiency in actual operation are identified. These high-quality patterns and efficiency data are transmitted back to the meso-path ecosystem through the micro-meta-macro mapping function, thereby driving the reallocation of resources and adjustment of competition priorities at the regional node level, and thus generating a node resource adjustment ecosystem.

[0147] By leveraging the macro-micro direct connection mapping function, direct data exchange and synchronous processing can be achieved between the macro-path ecosystem and the micro-path ecosystem, without relying on the meso-level as an intermediary. Macro-level fitness and resource data can be directly distributed to the micro-level, while micro-level operational fitness and resource utilization data are fed back to the macro-level in real time. This creates a high-speed, low-latency bidirectional linkage channel between the global and local levels, enabling rapid response in the event of large-scale changes or local emergencies, maintaining consistency between global strategies and local execution, and thus generating both macro-synchronous ecosystems and micro-synchronous ecosystems.

[0148] The rule-reconstructed ecosystem, global resource adjustment ecosystem, local resource optimization ecosystem, node resource adjustment ecosystem, macro-synchronization ecosystem, and micro-synchronization ecosystem are all input into the conflict coordination and processing module. This module detects and resolves potential resource allocation conflicts, priority conflicts, and path selection conflicts, ultimately resulting in a balanced and stable interactive path ecosystem.

[0149] In this embodiment, by constructing a bidirectional fitness mapping network covering macro, meso, and micro levels, efficient information interaction and dynamic collaborative optimization of the transportation path ecosystem across different scales are achieved. Each mapping function performs bidirectional driven adjustments to key parameters such as global, regional, and local resource density, bottleneck location, risk distribution, transportation efficiency, and resource fluctuations. This not only enables rapid response to the impact of macro-environmental changes on local operations but also provides real-time feedback of micro-operational status to the macro and meso levels for strategic correction. Simultaneously, the direct macro-micro mapping enables direct synchronization across scales, significantly shortening the response link and improving overall adaptability. Finally, through conflict coordination and handling, conflicts in resource allocation and priority settings across multiple scales are effectively eliminated, achieving stable coupling and continuous optimal operation of the global-regional-local multi-layer ecosystem. This maintains the high resilience and efficiency of the path network in complex and ever-changing transportation environments.

[0150] In one exemplary embodiment, such as Figure 6 As shown, resource competition analysis is performed on the ecosystems along each adjustment path to obtain the corresponding ecosystem fitness analysis data, including steps 602 to 608. Wherein:

[0151] Step 602: Construct a multi-scale competition factor matrix corresponding to each adjustment path ecosystem.

[0152] Step 604: Using a multidimensional dynamic game model, based on the multi-scale competition factor matrix and the migration intention perturbation function, calculate the initial resource occupation pattern and the initial resource transfer pattern of each adjustment path ecosystem.

[0153] Step 606: Based on the initial resource occupancy pattern and the initial resource transfer pattern, drive the resource competition relationship to evolve iteratively over time at each scale until each adjustment path ecosystem satisfies the stability criterion or evolution convergence condition, and obtain the full-time domain resource utilization efficiency, conflict dissipation speed, morphological diversity change curves and risk resilience data.

[0154] Step 608: Integrate the data on resource utilization efficiency, conflict dissipation speed, morphological diversity change curves, and risk resilience across the entire time domain to obtain the ecological field fitness analysis data.

[0155] Among them, the multi-scale competition factor matrix is ​​a matrix data structure formed by extracting, quantifying and normalizing multi-level competition factors in the transportation system, such as macro (e.g., global resource allocation and large-scale traffic pressure), meso (e.g., regional network structure and traffic bottlenecks), and micro (e.g., node capacity and local congestion). It is used to comprehensively describe the intensity, direction and correlation of cross-scale resource competition.

[0156] Among them, the multidimensional dynamic game model is an evolutionary analysis tool based on game theory. It regards different paths, nodes or transportation entities as game participants and combines their resource acquisition, occupation and transfer strategies to dynamically evolve in multiple dimensions (time, space and resource type) to simulate the changes in the competitive and cooperative behaviors of each participant under disturbance conditions.

[0157] Among them, the initial resource occupation pattern is the initial allocation and usage status of available transportation resources (such as capacity, right of way, energy, and time window) of each path ecosystem or transportation node under the starting conditions of game analysis, reflecting the resource holding pattern before competition.

[0158] Among them, the initial resource transfer pattern is the proportion and distribution of resources that each participant is willing to release or transfer under specific constraints or incentive mechanisms under the initial conditions of game analysis, reflecting the initial willingness and potential direction of resource flow.

[0159] Among them, resource competition is the state of mutual influence and game formed by different paths, nodes or regions in the transportation network when acquiring limited transportation resources. It is manifested in various interactive modes such as competition, sharing, transfer and coordination, and changes dynamically with time and disturbance conditions.

[0160] Among them, the stability criterion is a quantitative rule used to determine whether the evolution process of resource competition has reached a stable state, such as the rate of change of resource allocation being lower than a set threshold, the frequency of conflict events tending to stabilize, or the fluctuation range of key performance indicators converging within a limited range.

[0161] Among them, the evolution convergence condition refers to the termination condition in which the resource competition relationship reaches equilibrium or no longer changes significantly during the iterative evolution process. It is usually determined by combining stability criteria, time step limit and system energy change trend.

[0162] Among them, the full-time-domain resource utilization efficiency is a comprehensive evaluation index of the degree of occupation and utilization of available resources by each transportation path or node throughout the entire time cycle of resource competition evolution, reflecting the overall optimization level of resource allocation and use.

[0163] Among them, the conflict resolution rate is the rate of time required for a resource conflict to be resolved or eliminated after the intervention of coordination or concession mechanisms during the resource competition process, reflecting the system's conflict handling and recovery capabilities.

[0164] Among them, the morphological diversity change curve refers to the dynamic change trajectory of path structure, network morphology or transportation mode diversity over time during the evolution of resource competition, and is used to measure the structural flexibility and adaptability of the system under different disturbances.

[0165] Risk resilience data is a set of quantitative indicators that reflect the ability of a transportation system to maintain operational stability and restore its original performance when faced with emergencies, resource disruptions, or network failures. It includes data such as recovery time, performance loss magnitude, and adaptive adjustment capabilities.

[0166] Specifically, in each adjusted path ecosystem, key influencing factors are extracted based on resource elements and competitive relationships at different scales, such as macro (e.g., global transportation resource distribution), meso (e.g., regional traffic pressure), and micro (e.g., node capacity). These factors are then normalized and weighted to ultimately construct a multi-scale competitive factor matrix that reflects the cross-scale resource competition situation.

[0167] The multi-scale competition factor matrix and migration intention perturbation function are input into the multi-dimensional dynamic game model. During the game, the resource acquisition and transfer strategies of each participant (such as different transportation paths, nodes or regions) are introduced. The strategy weights are dynamically adjusted according to the real-time perturbation signal, and the initial resource occupation mode and initial resource transfer mode of each adjusted path ecosystem are calculated.

[0168] Using the initial resource occupancy pattern and the initial resource transfer pattern as inputs, the resource competition evolution model runs in parallel at three scales: macro, meso, and micro. It simulates the transfer, conflict, and mitigation processes of resources at different time steps. In each iteration, it calculates resource utilization rate, conflict occurrence and dissipation rate, morphological diversity index, and the resilience of the system under disturbances in real time. When all ecosystems reach the stability criterion or evolution convergence condition, it outputs four types of core dynamic data in the full time domain: full time domain resource utilization efficiency, conflict dissipation rate, morphological diversity change curve, and risk resilience data.

[0169] The data on resource utilization efficiency, conflict dissipation speed, morphological diversity change curves, and risk resilience across the entire time domain are multidimensionally integrated across time, space, and scale. The integration process typically employs weighted aggregation and correlation analysis methods to extract comprehensive fitness indicators, ultimately forming ecofield fitness analysis data that comprehensively reflects the survival and optimization capabilities of the ecofield in dynamic resource competition.

[0170] In this embodiment, by constructing a multi-scale competition factor matrix in each adjustment path ecosystem and introducing a multi-dimensional dynamic game model, the migration intention perturbation function is coupled with the resource competition mechanism at different scales, thus achieving an accurate characterization of the initial resource occupation pattern and transfer pattern. On this basis, through parallel iterative evolution at each scale, the resource competition, coordination, and transfer process is dynamically simulated. This not only captures the resource utilization efficiency and conflict dissipation speed in the entire time domain in real time, but also quantifies the morphological diversity changes and risk resilience level of the system under perturbation. Finally, the ecosystem fitness analysis results obtained by integrating multi-dimensional data can comprehensively reflect the survival and evolution capabilities of the path structure under different environmental and competitive conditions, thereby significantly improving the robustness and global optimization capability of transportation path planning, and possessing dynamic adaptability and prediction accuracy superior to traditional static or single-scale analysis methods.

[0171] In one exemplary embodiment, such as Figure 7 As shown, based on the migration intention perturbation function and real-time environmental perception data, the resource density parameters and resource competition coefficients of each initial path ecosystem are dynamically adjusted to obtain each adjusted path ecosystem, including steps 702 to 708. Wherein:

[0172] Step 702: Generate a multi-scenario evolution prior set based on the time series data of the migration intention perturbation function and the spatial distribution data of the real-time environmental perception data.

[0173] Step 704: Input the multi-scenario evolution prior set into the multi-scale ecological pressure calculation model to generate the dynamic ecological pressure field tensor.

[0174] Step 706: Perform a time-series convolution operation between the dynamic ecological pressure field tensor and the migration intention perturbation function to obtain the predictive perturbation response matrix.

[0175] Step 708: Based on the predictive perturbation response matrix, perform cross-scale dynamic optimization of the resource density parameters and resource competition coefficients of each initial path ecofield to obtain each adjusted path ecofield.

[0176] Time series data is a set of continuous data points collected and arranged in chronological order. It is used to characterize the changing patterns of the migration intention disturbance function at different times, such as the disturbance intensity, frequency, and duration, in order to capture dynamic trends and periodic characteristics.

[0177] Spatial distribution data refers to environmental characteristic data related to geographical location or spatial region, such as temperature distribution, traffic flow density, and location of risk events, which are used to describe the distribution of real-time environmental perception information within different geographical coordinates or partitions.

[0178] Among them, the multi-scenario evolution prior set is a set of potential future evolution scenarios generated by combining different environmental conditions, disturbance patterns and resource allocation schemes based on time series data and spatial distribution data. It is used to provide a variety of possible reference schemes for the adjustment and optimization of the ecosystem.

[0179] Among them, the multi-scale ecological pressure calculation model is a mathematical model that can calculate the degree of impact of environmental disturbance on the ecosystem at different scales such as macro, meso and micro, by comprehensively analyzing resource supply and demand, environmental risks and competitive relationships.

[0180] Among them, the dynamic ecological stress field tensor is a multidimensional array containing three-dimensional information of time, space and resources, used to represent the stress state of the ecosystem over time under different scenarios.

[0181] Temporal convolution is a convolution computation method performed in the time dimension. By sliding the time window, it fuses the dynamic ecological pressure field tensor with the migration intention perturbation function to extract the correlation and temporal dependency between them.

[0182] Among them, the predictive disturbance response matrix is ​​the result of temporal convolution operation, which is used to represent the intensity and direction of the response of the ecosystem to potential disturbances at different time nodes and spatial locations, thereby identifying possible changes in resource competition or areas of risk accumulation in advance.

[0183] Among them, cross-scale dynamic optimization is a parameter adjustment method that is carried out simultaneously at multiple levels, including macro, meso, and micro. By coordinating the resource density and competition coefficient of different scale ecosystems, it enables the whole to maintain high adaptability and stability under future disturbance conditions.

[0184] Specifically, based on the time series data of the migration intention perturbation function, the perturbation intensity and changing trend at different time points are analyzed. At the same time, combined with the spatial distribution characteristics of real-time environmental perception data, macroscopic, mesoscopic, and microscopic multi-scale environmental variables related to path ecology are extracted. These time-space characteristics are then combined to form a variety of possible evolution scenarios, and a multi-scenario evolution prior set is constructed.

[0185] By inputting a multi-scenario evolutionary prior set into a multi-scale ecological pressure calculation model, and calculating the pressure impact of environmental disturbances on the ecological field structure at different spatial scales and with different resource types, a dynamic ecological pressure field tensor that reflects the three-dimensional characteristics of time, space and resources is generated.

[0186] By performing temporal convolution operations on the dynamic ecological pressure field tensor and the migration intention perturbation function, the spatiotemporal distribution characteristics of environmental pressure are integrated with the change pattern of intention perturbation. Potential perturbation propagation paths and intensity evolution trends are extracted to obtain a predictive perturbation response matrix.

[0187] Based on the predictive perturbation response matrix, and considering the resource density parameters and resource competition coefficients of each initial path ecosystem, dynamic cross-scale optimization is performed in conjunction with the interaction patterns between ecosystems at different scales. This enables resource distribution and competition relationships to adapt to future perturbation scenarios in advance, generating adjusted path ecosystems with high adaptability across multiple scenarios at each scale.

[0188] In this embodiment, by combining the time-series features of the migration intention perturbation function with the spatial distribution information of real-time environmental perception, an evolutionary prior set covering multiple possible future scenarios is generated. This prior set is then used in a multi-scale ecological pressure calculation model to achieve dynamic quantification of pressure at the macro, meso, and micro levels, comprehensively characterizing the system's pressure state under different perturbation conditions. A predictive perturbation response matrix is ​​obtained through temporal convolution operations, enabling adjustments to the ecological field to no longer rely on a single static parameter but rather to be based on a forward-looking response to future perturbation trends. This allows for adaptive and coordinated configuration of resource density and competition coefficients during cross-scale dynamic optimization, significantly improving the stability and robustness of transportation routes in complex and volatile environments.

[0189] In one exemplary embodiment, such as Figure 8 As shown, simulation analysis is performed on the path structure family in the current transportation path evolution model environment to obtain transportation adaptability analysis data, including steps 802 to 814. Wherein:

[0190] Step 802: The current transportation path evolution model is spatiotemporally coupled with the real-time meteorological field, real-time traffic flow field and real-time emergency disturbance field to obtain a multi-field dynamic simulation environment.

[0191] Step 804: Drive the path structure family to run heterogeneously in a multi-field dynamic simulation environment to obtain full simulation response data.

[0192] Step 806: The simulation response data is segmented and analyzed according to time series and disturbance event labels to generate a disturbance phase running dataset.

[0193] Step 808: Based on the perturbation phase running dataset, calculate the multidimensional adaptive data of each path structure under different perturbation intensities and types to form a phased adaptive dataset.

[0194] Step 810: Input the phased adaptive dataset into the multidimensional response surface calculation model to obtain the adaptive response surface.

[0195] Step 812: Identify critical failure modes and adaptive evolution trends based on the gradient change information of the adaptive response surface.

[0196] Step 814 involves performing a correlation analysis on the adaptive response surface, critical failure mode, and adaptive evolution trend to obtain transportation adaptability analysis data.

[0197] Among them, the real-time meteorological field is a set of spatial mapping data that provides high precision and dynamic updates of weather conditions within the cargo transportation area. It includes the spatiotemporal distribution of multi-dimensional meteorological elements such as temperature, humidity, wind speed and direction, precipitation, and visibility. It is used to simulate the immediate impact of weather on transportation efficiency and safety in the transportation route evolution model.

[0198] Among them, the real-time traffic flow field is a spatiotemporal mapping of multi-dimensional traffic parameters such as traffic flow, vehicle speed distribution, congestion level and road traffic status collected and dynamically updated within the transportation network. It is used to reflect the real-time transportation capacity and traffic resistance of each road segment.

[0199] Among them, the real-time emergency disturbance field is dynamic scene data that is used to perceive and model the spatiotemporal location, impact range, duration and disturbance intensity of sudden events (such as traffic accidents, road closures, natural disasters, temporary traffic control, etc.) in the transportation environment in real time, and is used to introduce unexpected external shocks in simulation.

[0200] Spatiotemporal synchronous coupling is a process of aligning and merging real-time meteorological fields, real-time traffic flow fields, and real-time sudden event disturbance fields at the same time and in the same spatial coordinate system to ensure that information from multiple fields has a unified time base and spatial reference in the simulation model.

[0201] Among them, the multi-field dynamic simulation environment is a comprehensive simulation environment that can dynamically change in time and space, generated by the spatiotemporal synchronous coupling of real-time meteorological field, real-time traffic flow field and real-time sudden event disturbance field. It is used to reproduce the evolution process of transportation path under complex external conditions with high fidelity.

[0202] Among them, heterogeneous agent operation is to use virtual transportation entities of different types and attributes (such as different vehicle models, load classes, driving strategies, priority rules) as agent objects in a multi-field dynamic simulation environment, and drive them to run in parallel simulation according to a specific path structure in order to evaluate the differences in the performance of the path under various real transportation scenarios.

[0203] Among them, the full-process simulation response data refers to the complete set of output data generated by the heterogeneous agent during the entire process of running from the starting point to the end point in a multi-field dynamic simulation environment, including time series records of indicators such as time consumption, energy consumption, delay, risk event occurrence rate, and path switching frequency.

[0204] The disturbance phase running dataset is a structured dataset that is created by dividing the running data from the full simulation response data into multiple disturbance phases according to the occurrence time, duration and intensity of the disturbance events, and recording the running performance indicators, resource usage and response mode for each phase.

[0205] Among them, disturbance type refers to the category of external disturbances that cause changes in the performance of the transportation system, including weather disturbances, traffic flow disturbances, sudden event disturbances, resource shortage disturbances, etc. Different disturbance types have different impact mechanisms and evaluation indicators in simulation analysis.

[0206] Among them, multidimensional adaptive data is a set of multidimensional indicators calculated based on the operational dataset during the disturbance phase. These indicators reflect the transport adaptability of the route structure under different disturbance intensities, disturbance types, and time scales, including efficiency retention rate, risk control capability, recovery speed, and stability score.

[0207] Among them, the phased adaptive dataset is a structured dataset that groups and organizes multidimensional adaptive data according to the perturbation stage and the corresponding operating conditions. It is used to compare and analyze the adaptive characteristics and changing trends of the path under different stages.

[0208] Among them, the multidimensional response surface calculation model is a mathematical modeling method based on multivariate input-output relationships. By fitting the response distribution surface of the staged adaptive dataset under multidimensional perturbation conditions, it is used to reveal the overall trend and sensitivity characteristics of path adaptability as perturbation factors change.

[0209] Among them, the adaptive response surface is a visual mathematical representation output by the multidimensional response surface computation model, which describes the continuous change of path adaptability in the space of disturbance intensity, disturbance type and other key variables.

[0210] Among them, critical failure mode is an operating mode identified in the adaptive response surface that triggers irreversible performance degradation or complete failure under specific perturbation conditions. It is usually accompanied by high risk index value and low recovery capability score.

[0211] Among them, the adaptive evolution trend is the long-term direction and speed of change derived from the adaptive change trajectory under multiple perturbation stages, which is used to predict the adaptive improvement or decline trend of the path under continuous perturbation or future changes in conditions.

[0212] Specifically, by unifying the clock and spatial reference, the structural parameters of the current transportation path evolution model are aligned and fused with real-time collected meteorological, traffic flow and emergency data in time and space to form a multi-source dynamic environment that is highly consistent in time and space, namely a multi-field dynamic simulation environment.

[0213] In the constructed multi-field dynamic simulation environment, multiple types of agents (such as vehicle agents, node agents, and event agents) are used to represent different transportation entities and resources. Asynchronous scheduling and multi-threaded computing drive the path structure group to run in a complex environment, record its state changes and response behavior throughout the process, and generate full-process simulation response data.

[0214] The simulation response data is segmented in chronological order and classified and labeled with pre-defined disturbance event labels (such as rainstorm, traffic congestion, road closure, etc.) to form a running dataset divided by disturbance stage as the disturbance stage running dataset.

[0215] The operational data for each disturbance stage in the disturbance phase dataset is preprocessed, including time alignment, outlier removal, and key indicator extraction. Then, based on different disturbance types (e.g., meteorological, traffic flow, event-related) and disturbance intensity levels, adaptive indicators for the route structure are calculated across multiple dimensions, such as transport timeliness (arrival delay rate, average speed variation), transport economy (fuel consumption rate, additional cost ratio), operational stability (speed fluctuation coefficient, number of route switching), and risk resilience (interruption recovery time, alternative route availability). Multidimensional vectors are constructed based on these indicators and categorized according to disturbance type and intensity to obtain the phased adaptive dataset.

[0216] Using phased adaptive data as input samples, a multidimensional response surface calculation model (e.g., based on radial basis function interpolation, kriging interpolation, or polynomial regression fitting) is used to perform fitting operations in the perturbation type–perturbation intensity–adaptive index space, mapping discrete adaptive sample points to a continuous multidimensional surface, thus obtaining the adaptive response surface.

[0217] The gradient changes of the adaptive response surface in the perturbation space are analyzed to capture the critical points where sharp decreases or discontinuous changes occur. Based on these captured features, the modes that lead to path structure failure are identified as critical failure modes. At the same time, the adaptive evolution trend of the adaptive response with the perturbation is extracted.

[0218] The gradient features, extreme value distributions, and high-risk region information extracted from the adaptive response surface are matched and compared with the characteristic parameters of critical failure modes (such as failure triggering conditions, time windows, and the set of affected path nodes) to determine the distribution location and evolution trajectory of each failure mode in the disturbance space. Subsequently, the evolution trajectories of these failure modes are dynamically correlated with the adaptive evolution trend to analyze the causes and development directions of path adaptive changes under different disturbance scenarios, identifying key influencing factors for stability improvement or decline. Finally, the above analysis results are integrated into structured transportation adaptability analysis data, which includes the adaptability level assessment, failure risk probability distribution, performance degradation rate, recovery capability score, and optimal operating condition recommendations for each path structure under different disturbance conditions.

[0219] In this embodiment, a highly realistic multi-field dynamic simulation environment is constructed by spatiotemporally coupling the current transportation route evolution model with real-time meteorological, traffic flow, and emergency information. Within this environment, a heterogeneous proxy operation mode is used to comprehensively evaluate the entire operational performance of the route family. This allows for the extraction of operational data in stages under various disturbance scenarios and the quantification of multi-dimensional adaptability indicators. By using a multi-dimensional response surface computation model to identify sensitive areas and gradient characteristics of adaptive changes, not only can critical failure modes be accurately located, but also adaptive evolution trends can be predicted. Ultimately, this achieves a precise assessment of the resilience, stability, and optimization potential of transportation routes in complex dynamic environments, providing data-driven decision-making support for route optimization and risk control, and significantly improving the safety and timeliness of transportation tasks.

[0220] In one exemplary embodiment, such as Figure 9 As shown, based on real-time environmental perception data of the cargo transportation task, target migration processing is performed on the initial target intent data to obtain target migration intent data, including steps 902 to 910. Wherein:

[0221] Step 902: Perform multi-dimensional target vector parsing on the initial target intent data to obtain the initial target vector set.

[0222] Step 904: The real-time environmental perception data is processed in layers according to spatial resolution and temporal granularity to obtain the environment-target matching matrix.

[0223] Step 906: Calculate the rate of change of environmental fitness at the macro, meso, and micro levels based on the environment-target matching matrix to obtain migration trigger signals and trigger dimension data.

[0224] Step 908: If the migration trigger signal is not null, perform multi-scenario prediction correction based on the trigger dimension data to obtain a migration candidate intent vector set.

[0225] Step 910: Select the highest priority vector from the migration candidate intent vector set as the target migration intent data.

[0226] Among them, multidimensional target vector analysis decomposes and quantifies the comprehensive target information in cargo transportation tasks (such as timeliness requirements, transportation costs, risk tolerance, service level, etc.) and transforms it into a multidimensional vector form that can be expressed in mathematical space, so that each dimension corresponds to a specific task indicator.

[0227] The initial target vector set is a set of vectors obtained after multidimensional target vector parsing. It represents the expected target state in each dimension at the beginning of the task. As the baseline intent model of the task, it is used to measure the degree of impact of environmental changes on the task objective when matching and comparing with real-time environmental data.

[0228] Spatial resolution refers to the degree of subdivision of environmental perception data in spatial dimensions. For example, it describes the environmental state at different scales such as cities, regions, road networks, and nodes. The higher the spatial resolution, the more refined the capture of environmental changes, which helps to plan and adjust paths at different levels.

[0229] Among them, time granularity refers to the sampling interval and fineness of environmental perception data in the time dimension, such as dividing it into time units such as minutes, hours, and days. The smaller the time granularity, the more timely the system can capture dynamic changes, but the amount of computation and data processing will also increase accordingly.

[0230] Among them, the environment-target matching matrix is ​​a multi-dimensional comparison data structure. It calculates the fit between the features of real-time environment perception data at different spatial resolutions and temporal granularities and the various dimensions of the initial target vector set, and records the matching results in the form of a matrix, which facilitates the subsequent calculation of fitness change rate and migration trigger determination.

[0231] Among them, the rate of change of environmental adaptability measures the rate at which the adaptability of a certain task objective dimension changes over time under specific environmental conditions. It is used to reflect the degree of influence of environmental changes on the possibility of achieving the objective. The larger the rate of change, the more significant the environmental impact on the objective dimension.

[0232] Among them, the migration trigger signal is a signal generated when the rate of change of a certain dimension exceeds a preset threshold after calculating the rate of change of environmental fitness. It indicates that the current target intent needs to be adjusted according to the environmental changes in order to avoid task execution risks or performance degradation.

[0233] Among them, the trigger dimension data is the labeled information generated along with the migration trigger signal, which clearly indicates which one or more target dimensions triggered the migration condition. It serves as the input basis for subsequent multi-scenario prediction correction, ensuring that the adjustment only applies to the affected dimensions.

[0234] Among them, multi-scenario prediction correction generates different possible future states for the trigger dimension by combining historical data, trend prediction models and hypothetical scenarios when the migration trigger conditions are met, and corrects the target vector based on these predictions to obtain migration candidate solutions with high feasibility in different scenarios.

[0235] Among them, the migration candidate intent vector set is a set of alternative target vectors generated by multi-scenario prediction and correction. Each vector represents a possible migration adjustment scheme. Subsequently, these candidate schemes will be sorted according to priority, and the optimal solution will be selected as the target migration intent data.

[0236] Specifically, the initial target intent data of the cargo transportation task is analyzed, and the multi-dimensional demand factors involved in the task (such as time priority, transportation cost constraints, risk tolerance, service level requirements, etc.) are transformed into structured multi-dimensional target vectors. Each vector component represents the quantified target value under that dimension, thus forming a complete set of initial target vectors.

[0237] The real-time environmental perception data is processed in a multi-scale hierarchical manner, and is structured into groups according to different spatial resolutions and temporal granularities, such as macro (regional traffic flow and weather trends), meso (local regional event impact), and micro (node ​​or road segment status). Then, these grouped environmental features are matched with the initial target vector set, the fit between each target component and the environmental features is calculated, and a multi-dimensional environment-target matching matrix is ​​constructed.

[0238] Using the environment-target matching matrix, the fitness change rate of the target under the three environmental dimensions of macro, meso and micro are calculated and compared with the preset migration trigger threshold. When the change rate of one or more dimensions exceeds the threshold, a migration trigger signal is generated and the specific target dimension of the trigger is recorded as the trigger dimension data.

[0239] When the system detects a migration trigger signal (i.e., when the migration trigger signal is not null), it will generate predicted values ​​for the target component corresponding to the trigger dimension data under different scenarios, while keeping other non-triggered dimensions stable, by combining historical task execution data, real-time trend prediction models, and multiple hypothetical scenarios. Based on these prediction results, the target vector will be corrected, thereby forming a set of multiple possible migration candidate intent vectors.

[0240] Based on the task priority rules (which may take into account factors such as task urgency, cost-effectiveness, risk exposure, and service level requirements), the multiple possible migration candidate intent vectors in the migration candidate intent vector set are prioritized and ranked, and the intent vector with the highest score is selected as the final target migration intent data.

[0241] In this embodiment, by parsing the initial target intent data into a multi-dimensional target vector and combining it with the hierarchical processing of real-time environmental perception data at different spatial resolutions and temporal granularities, precise matching between environmental states and task objectives is achieved, thereby enabling the quantification of environmental adaptability changes at macro, meso, and micro scales. Through a migration trigger signal and trigger dimension data determination mechanism, the affected target dimension can be quickly located when environmental shocks are significant. Multiple alternative migration schemes are generated using multi-scenario prediction and correction, and finally, the optimal target migration intent data is selected based on priority. This not only significantly improves the adaptability of transportation route planning to dynamic environments but also ensures decision stability and task execution continuity while reducing computational burden, thereby effectively improving the safety, timeliness, and resource utilization efficiency of cargo transportation.

[0242] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0243] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for cargo transportation route planning and tracking, characterized in that, The method includes: Acquire initial target intent data, transportation environment status data, and original transportation path evolution model for cargo transportation tasks; Based on the initial target intent data and the transportation environment status data, an initial transportation planning path for the cargo transportation task is constructed; Based on the initial transportation planning path, the initialization parameters of the original transportation path evolution model are set to obtain the current transportation path evolution model; Based on the real-time environmental perception data of the cargo transportation task, the initial target intent data is processed to obtain target migration intent data. The target migration intention data is input into the current transportation path evolution model to obtain an updated transportation planning path; The execution status of the cargo transportation task is tracked based on the updated transportation planning route.

2. The method according to claim 1, characterized in that, The step of inputting the target migration intention data into the current transportation path evolution model to obtain an updated transportation planning path includes: The target migration intention data is mapped to the multidimensional path evolution topology space of the cargo transportation task to construct a migration intention perturbation function. Based on the migration intention perturbation function, the path structure in the current transportation path evolution model is driven to undergo multi-scale self-fractal evolution to generate a path structure family. The path structure family is simulated and analyzed in the current transportation path evolution model environment to obtain transportation adaptability analysis data; Based on the transportation adaptability analysis data, the trend path structure is extracted from the evolutionary stable region of the current transportation path evolution model to obtain the updated transportation planning path.

3. The method according to claim 2, characterized in that, The step of driving the path structure in the current transportation path evolution model to undergo multi-scale self-fractal evolution based on the migration intention perturbation function to generate a path structure family includes: Based on the current transportation path evolution model, initial path ecosystems are constructed under different scale environments; Based on the migration intention perturbation function and the real-time environmental perception data, the resource density parameters and resource competition coefficients of each initial path ecosystem are dynamically adjusted to obtain each adjusted path ecosystem. Resource competition analysis was performed on each of the aforementioned adjustment path ecofields to obtain ecofield fitness analysis data corresponding to each of the aforementioned adjustment path ecofields; Based on the fitness analysis data of each of the aforementioned ecological fields, cross-scale information interaction processing is performed on each of the aforementioned adjustment path ecological fields to obtain each interaction path ecological field. If any of the interaction path ecosystems has not reached a stable state, return to the step of performing resource competition analysis on each of the adjustment path ecosystems to obtain the ecosystem fitness analysis data corresponding to each of the adjustment path ecosystems, until each of the interaction path ecosystems reaches the stable state and obtains each stable path ecosystem. Paths whose morphological and ecological fitness differences meet preset thresholds are identified from each stable path ecosystem and are thus identified as the path structure family.

4. The method according to claim 3, characterized in that, The adjusted path ecosystem includes a macroscopic path ecosystem, a mesoscopic path ecosystem, and a microscopic path ecosystem; the cross-scale information interaction processing of each adjusted path ecosystem, based on fitness analysis data of each ecosystem, yields each interactive path ecosystem, including: Based on the fitness analysis data of each ecological field, a three-layer bidirectional fitness mapping network is constructed; the three-layer bidirectional fitness mapping network includes macro-macro mapping function, mid-macro mapping function, mid-micro mapping function, micro-macro mapping function, and macro-micro direct connection mapping function; Based on the macro-macro mapping function, the resource density and competition coefficient of the meso-path ecosystem are adjusted using the resource density, global bottleneck, and risk distribution of the macro-path ecosystem to obtain a rule-reconstructed ecosystem. Based on the aforementioned meso-macro mapping function, the global resource allocation of the macro-path ecosystem is adjusted using the transportation efficiency pattern and fitness trend of the meso-path ecosystem to obtain a global resource adjustment ecosystem. Based on the aforementioned meso-micro mapping function, the resource fluctuations and flow changes of the meso-path ecosystem are used to adjust the local resource utilization strategy of the micro-path ecosystem, thereby obtaining a locally optimized ecosystem. Based on the micro-middle mapping function, the fitness path pattern and resource utilization efficiency of the micro-path ecosystem are used to adjust the node resources and competition priorities of the meso-path ecosystem to obtain the node resource adjustment ecosystem. Based on the macro-micro direct connection mapping function, the fitness data and resource data of the macro-path ecosystem and the micro-path ecosystem are used for synchronization processing to obtain the macro-synchronous ecosystem and the micro-synchronous ecosystem. The conflict coordination process is performed on the rule-reconstructed ecosystem, the global resource adjustment ecosystem, the local resource optimization ecosystem, the node resource adjustment ecosystem, the macro-synchronization ecosystem, and the micro-synchronization ecosystem to obtain the ecosystems of each interaction path.

5. The method according to claim 3, characterized in that, The resource competition analysis of each of the adjusted path ecosystems is performed to obtain the ecosystem fitness analysis data corresponding to each of the adjusted path ecosystems, including: Construct a multi-scale competition factor matrix corresponding to each of the aforementioned adjustment path ecosystems; Using a multidimensional dynamic game model, based on the multi-scale competition factor matrix and the migration intention perturbation function, the initial resource occupation mode and the initial resource transfer mode of each adjustment path ecosystem are calculated. Based on the initial resource occupancy mode and the initial resource transfer mode, the resource competition relationship is driven to evolve iteratively over time at each scale until each of the adjusted path ecosystems meets the stability criterion or evolution convergence condition, thereby obtaining the full-time domain resource utilization efficiency, conflict dissipation speed, morphological diversity change curves and risk resilience data. By integrating the full-time-domain resource utilization efficiency, the conflict dissipation rate, the morphological diversity change curve, and the risk resilience data, the ecofield fitness analysis data is obtained.

6. The method according to claim 3, characterized in that, The step of dynamically adjusting the resource density parameters and resource competition coefficients of each initial path ecosystem based on the migration intention perturbation function and the real-time environmental perception data to obtain each adjusted path ecosystem includes: Based on the time series data of the migration intention perturbation function and the spatial distribution data of the real-time environmental perception data, a multi-scenario evolution prior set is generated. The multi-scenario evolution prior set is input into the multi-scale ecological pressure calculation model to generate a dynamic ecological pressure field tensor. The dynamic ecological pressure field tensor is convolved with the migration intention perturbation function in a time series to obtain the predictive perturbation response matrix. Based on the predictive perturbation response matrix, the resource density parameters and resource competition coefficients of each initial path ecofield are dynamically optimized across scales to obtain each adjusted path ecofield.

7. The method according to claim 2, characterized in that, The simulation analysis of the path structure family in the current transportation path evolution model environment yields transportation adaptability analysis data, including: The current transportation path evolution model is spatiotemporally coupled with the real-time meteorological field, real-time traffic flow field, and real-time emergency disturbance field to obtain a multi-field dynamic simulation environment. In the multi-field dynamic simulation environment, the path structure family is driven to perform heterogeneous proxy operation to obtain full-process simulation response data; The full-process simulation response data is segmented and analyzed according to time series and disturbance event labels to generate a disturbance phase running dataset. Based on the perturbation phase running dataset, multidimensional adaptive data of each path structure under different perturbation intensities and types are calculated to form a phased adaptive dataset; The staged adaptive dataset is input into the multidimensional response surface calculation model to obtain the adaptive response surface; Based on the gradient change information of the adaptive response surface, the critical failure mode and adaptive evolution trend are identified; The adaptive response surface, the critical failure mode, and the adaptive evolution trend are correlated to obtain the transportation adaptability analysis data.

8. The method according to claim 1, characterized in that, The step of performing target migration processing on the initial target intent data based on the real-time environmental perception data of the cargo transportation task to obtain target migration intent data includes: The initial target intent data is parsed using multi-dimensional target vector analysis to obtain an initial target vector set; The real-time environmental perception data is processed in layers according to spatial resolution and temporal granularity to obtain the environment-target matching matrix. Based on the environment-target matching matrix, the rate of change of environmental fitness at the macro, meso, and micro levels is calculated to obtain migration trigger signals and trigger dimension data. If the migration trigger signal is not null, multi-scenario prediction and correction are performed based on the trigger dimension data to obtain a set of migration candidate intent vectors. The highest priority vector is selected from the set of migration candidate intent vectors as the target migration intent data.

9. A cargo transportation route planning and tracking device, characterized in that, The device includes: The transportation data acquisition module is used to acquire the initial target intent data, transportation environment status data, and original transportation path evolution model of the cargo transportation task. The transportation route planning module is used to construct the initial transportation planning route for the cargo transportation task based on the initial target intent data and the transportation environment status data. The path model initialization module is used to set the initialization parameters of the original transportation path evolution model according to the initial transportation planning path, so as to obtain the current transportation path evolution model. The target migration processing module is used to perform target migration processing on the initial target intent data based on the real-time environmental perception data of the cargo transportation task, so as to obtain target migration intent data. The transportation route update module is used to input the target migration intention data into the current transportation route evolution model to obtain an updated transportation planning route; The transportation task tracking module is used to track the execution status of the cargo transportation task based on the updated transportation planning route.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images