Road detection system and method based on highway intelligent gateway system

Abstract

The present invention discloses a road detection system and method based on a highway intelligent gateway system, and relates to the technical field of intelligent traffic control. The system comprises a road detection module and a highway intelligent gateway system; wherein the road detection module is used to execute the road information detection task of the highway; the highway intelligent gateway system is used to dynamically configure resources for the road information detection task; a resource allocation unit is used to allocate the first edge resources of multiple network edge nodes for simple charging tasks in real time; according to the allocation of the first edge resources, a first complex charging task is determined from complex charging tasks, and the second edge resources of multiple network edge nodes are allocated to the first complex charging task in real time; and a second complex charging task other than the first complex charging task in the complex charging tasks is sent to a cloud module; the cloud module is used to allocate cloud resources to the second complex charging task in real time to obtain a third processing result.

Description

Technical Field

[0001] The present application relates to the field of intelligent traffic control technology, and in particular to a road detection system and method based on a highway intelligent gateway system. Background Art

[0002] In the modern highway management system, the road detection system, as an important part of the traffic control system, plays a vital role in ensuring traffic safety and improving traffic efficiency. It needs to process and analyze road information detection tasks. However, the existing road detection system still has some shortcomings when facing the increasingly complex traffic environment and massive data.

[0003] In the prior art, the Chinese invention patent with publication number CN117765734A discloses a road detection method and system, which obtains the traffic status data of each branch road according to the traffic data of each branch road; calculates the congestion impact value according to the traffic status data; judges the traffic congestion connectivity of the intersection based on the congestion impact value, and generates the branch congestion association signal; calculates the accident impact value according to the duration of each branch road accident; judges the traffic accident connectivity of the intersection, and generates the branch road accident association signal; judges the road evacuation capacity according to the branch road congestion association signal and the branch road accident association signal, and obtains the road evacuation signal; sets the impact analysis cycle, obtains the road impact judgment value, and judges the impact between congestion and accidents; the patent not only realizes the judgment of the road evacuation capacity, but also realizes the judgment of the factors affecting the road congestion, so that the evacuation capacity of the road can be effectively judged. The Chinese invention patent with publication number CN114973646A discloses a road detection method, device, system and server. The road can be sensed by perception sensors to collect perception data, and then abnormal road conditions can be detected based on the perception data to obtain abnormal data. Prompt information is generated based on the abnormal data to prompt vehicles on the road so that the vehicles can adjust their operating status based on the prompts. By timely detecting abnormal conditions such as road diseases, damaged or faulty road signs, etc., the impact of the above abnormal conditions on the normal operation of the vehicle can be reduced, and the probability of traffic accidents can also be reduced.

[0004] With the exponential growth of traffic data, existing road detection systems face huge challenges in task processing and analysis. Traditional task processing methods often use centralized processing, transferring all data to a central server for unified processing, which not only increases the burden of data transmission, but also easily leads to delays in data processing.

[0005] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0006] The embodiments of the present application provide a road detection system and method based on a highway intelligent gateway system to solve the above-mentioned technical problems.

[0007] The present application provides a road detection system based on a highway intelligent gateway system, comprising a road detection module and a highway intelligent gateway system; wherein the road detection module is used to perform a highway road information detection task; the highway intelligent gateway system is used to dynamically configure resources for the road information detection task; wherein the highway intelligent gateway system comprises a terminal module, an edge module and a cloud module; the terminal module is used to collect lane-level information of each lane of the highway and send the lane-level information to the edge module; the edge module is composed of a plurality of network edge nodes; the edge module comprises a task classification unit and a resource allocation unit; the task classification unit is used to classify the road information detection task to obtain a simple charging task and a complex charging task; The resource allocation unit is used to allocate the first edge resources of multiple network edge nodes to the simple charging task in real time to obtain the first processing result of the simple charging task; determine the first complex charging task from the complex charging tasks according to the allocation of the first edge resources, and allocate the second edge resources of multiple network edge nodes to the first complex charging task in real time to obtain the second processing result of the first complex charging task; and send the second complex charging task in the complex charging tasks except the first complex charging task to the cloud module; the cloud module is used to allocate cloud resources to the second complex charging task in real time to obtain a third processing result.

[0008] The present application provides a road detection method based on a highway intelligent gateway system, comprising: executing a highway road information detection task according to a resource allocation strategy of the highway intelligent gateway system; wherein the highway intelligent gateway system comprises a terminal module, an edge module and a cloud module; the terminal module is used to collect lane-level information of each lane of the highway and send the lane-level information to the edge module; wherein the edge module is composed of a plurality of network edge nodes; the edge module comprises a task classification unit and a resource allocation unit; the task classification unit classifies the road information detection task to obtain a simple charging task and a complex charging task; the resource allocation unit allocates a first edge resource of a plurality of network edge nodes to the simple charging task in real time to obtain a first processing result of the simple charging task; according to the allocation of the first edge resource, a first complex charging task is determined from the complex charging tasks, and a second edge resource of a plurality of network edge nodes is allocated to the first complex charging task in real time to obtain a second processing result of the first complex charging task; and a second complex charging task other than the first complex charging task in the complex charging task is sent to the cloud module; the cloud module allocates cloud resources to the second complex charging task in real time to obtain a third processing result.

[0009] Based on the embodiments provided by the present application, the data processing time can be effectively shortened, and real-time information on road conditions can be obtained in a timely manner, thereby improving the overall efficiency of road detection and meeting the needs of modern highway traffic management for efficient detection; the highway intelligent gateway system can dynamically allocate resources for tasks according to the actual needs of road information detection tasks. The task classification unit of the edge module divides the detection tasks into simple toll tasks and complex toll tasks, and the resource allocation unit allocates the corresponding edge resources in real time according to the complexity and urgency of the tasks. This dynamic resource configuration method enables the system to flexibly respond to the detection needs of different sections and different time periods, optimize resource utilization, avoid resource waste, and improve the adaptability and stability of the system; the traditional centralized processing method requires all data to be transmitted to the central server, while the system adopts a distributed architecture, and the lane-level information collected by the terminal module is first sent to the edge module for preliminary processing. The edge module can handle most of the simple toll tasks, and complete part of the complex toll tasks locally, and the other part of the complex tasks are sent to the cloud module. Such a data processing process greatly reduces the amount of data transmission, reduces the burden of data transmission, reduces the delay in the data transmission process, and improves the response speed of the system; the highway intelligent gateway system consists of terminal modules, edge modules and cloud modules, and each module has a clear division of labor and works together. When the scale of the highway expands or the detection demand increases, the corresponding modules or nodes can be flexibly added, such as adding more terminal modules to cover a wider section of the road, or adding network edge nodes of the edge module to improve processing capabilities. This modular design makes the system have good scalability, can adapt to the needs of future traffic development, and provide continuous technical support for highway traffic management. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the embodiments of the present invention and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0011] Figure 1 is a structural diagram of an optional road detection system based on a highway intelligent gateway system according to an embodiment of the present application;

[0012] Figure 2 The present invention is a flowchart of an optional road detection method based on a highway intelligent gateway system according to an embodiment of the present application.

[0013] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] Alternatively, if Figure 1 As shown, the present application provides a road detection system based on a highway intelligent gateway system, including a road detection module 101 and a highway intelligent gateway system 102; wherein:

[0016] The road detection module 101 is used to perform the road information detection task of the expressway;

[0017] The highway intelligent gateway system 102 is used to dynamically configure resources for road information detection tasks; wherein the highway intelligent gateway system 102 includes a terminal module 103, an edge module 104 and a cloud module 105;

[0018] The terminal module 103 is used to collect lane-level information of each lane of the highway and send the lane-level information to the edge module 103;

[0019] The edge module 104 is composed of a plurality of network edge nodes; the edge module includes a task classification unit and a resource allocation unit; the task classification unit is used to classify the road information detection tasks to obtain simple charging tasks and complex charging tasks;

[0020] The resource allocation unit is used to allocate first edge resources of multiple network edge nodes for a simple charging task in real time to obtain a first processing result of the simple charging task; determine a first complex charging task from the complex charging tasks according to the allocation of the first edge resources, allocate second edge resources of multiple network edge nodes for the first complex charging task in real time to obtain a second processing result of the first complex charging task; and send the second complex charging task in the complex charging task except the first complex charging task to the cloud module 105;

[0021] The cloud module 105 is used to allocate cloud resources to the second complex charging task in real time to obtain a third processing result.

[0022] Cloud-edge collaboration is achieved through the Kubernetes control node and the nodes run by KubeEdge, while edge-end collaboration is achieved through KubeEdge and EdgeXFoundry. The cloud module is used to provide policy and algorithm support for the edge module to ensure data consistency and business collaboration of the entire lane toll collection system. The cloud module also includes a smart highway comprehensive management cloud platform. Combining the requirements of smart highway planning and full-network video network monitoring, it uses Internet big data, cloud computing, and artificial intelligence information technology to build a set of information-interconnected, collaborative and efficient highway management and information service platform systems, improve road operation monitoring and command capabilities, and enhance the level of public travel information services on highways.

[0023] Based on the embodiments provided by the present application, the data processing time can be effectively shortened, and real-time information on road conditions can be obtained in a timely manner, thereby improving the overall efficiency of road detection and meeting the needs of modern highway traffic management for efficient detection; the highway intelligent gateway system can dynamically allocate resources for tasks according to the actual needs of road information detection tasks. The task classification unit of the edge module divides the detection tasks into simple toll tasks and complex toll tasks, and the resource allocation unit allocates the corresponding edge resources in real time according to the complexity and urgency of the tasks. This dynamic resource configuration method enables the system to flexibly respond to the detection needs of different sections and different time periods, optimize resource utilization, avoid resource waste, and improve the adaptability and stability of the system; the traditional centralized processing method requires all data to be transmitted to the central server, while the system adopts a distributed architecture, and the lane-level information collected by the terminal module is first sent to the edge module for preliminary processing. The edge module can handle most of the simple toll tasks, and complete part of the complex toll tasks locally, and the other part of the complex tasks are sent to the cloud module. Such a data processing flow greatly reduces the amount of data transmission, reduces the burden of data transmission, reduces the delay in the data transmission process, and improves the response speed of the system; the highway intelligent gateway system consists of terminal modules, edge modules and cloud modules, and each module has a clear division of labor and works together. When the scale of the highway expands or the detection demand increases, the corresponding modules or nodes can be flexibly added, such as adding more terminal modules to cover a wider section of the road, or adding network edge nodes of the edge module to improve processing capabilities. This modular design makes the system have good scalability, can adapt to the needs of future traffic development, and provide continuous technical support for highway traffic management. Furthermore, lane-level information includes vehicle identification information, real-time dynamic information of vehicles and vehicle transaction information; the task classification unit is specifically used to classify the toll tasks of each lane according to computational complexity, data dependency, decision complexity, real-time requirements, security and privacy requirements, and processing requirements, and obtain simple toll tasks and complex toll tasks; the edge module is deployed at the toll station; the edge module is distributed between the terminal module and the cloud module;

[0024] The distinction between simple charging tasks and complex charging tasks can be determined based on factors such as task processing requirements, computational complexity, real-time requirements, and data dependencies. The following are some criteria for distinguishing simple and complex charging tasks:

[0025] Processing requirements: Simple tasks: usually involve standard, predefined processes such as basic vehicle identification, basic toll calculation, simple transaction verification, etc. Complex tasks: may include abnormal transaction processing, fraud detection, conflict resolution of multiple vehicles trading at the same time, application of special preferential policies, etc. These tasks require more in-depth data analysis and decision support.

[0026] Computational complexity: Simple tasks: The computational process is relatively simple, with low demand for computing resources, and can be completed quickly on the edge module. Complex tasks: Require more advanced algorithms and more computing resources, and may involve the application of big data analysis and machine learning models, which are usually more appropriately processed in the cloud.

[0027] Real-time requirements: Simple tasks: have high real-time requirements and require quick response, such as instant toll collection when a vehicle passes through a toll booth. Complex tasks: can tolerate certain delays because they may involve more extensive data collection and analysis.

[0028] Data dependency: Simple tasks usually rely on local data, such as vehicle information in a single lane. Complex tasks may require data integration and analysis across lanes, regions, or even networks.

[0029] Decision complexity: Simple tasks: The decision path is clear and can be easily automated. Complex tasks: Multiple factors need to be considered comprehensively and may require human intervention or advanced decision support systems.

[0030] Security and privacy requirements: Simple tasks: have relatively low security and privacy requirements. Complex tasks: may involve the processing of sensitive data and require more advanced security measures and privacy protection.

[0031] The resource allocation unit allocates first edge resources of multiple network edge nodes in real time for a simple charging task, and is configured as follows:

[0032] The receiving terminal module collects the vehicle identification information, real-time dynamic information and vehicle transaction information of each lane;

[0033] Determine the network edge node set based on the multiple network edge nodes included in the edge terminal module ; Each network edge node It has corresponding edge computing resources, edge storage resources, edge network resources, and edge task scheduling resources; is the number of network edge nodes;

[0034] Using multi-objective genetic algorithm to identify edge nodes in the network Perform clustering to obtain multiple resource package sets ; Each resource pack yes A subset of represents a group of edge nodes with similar resource characteristics; is the number of resource packages; resource characteristics include geographical location or resource usage;

[0035] The improved ant algorithm is used to optimize the connection between multiple resource packages to form a resource cluster set. ; Each resource cluster is a subset of P, each resource cluster Across different edge nodes, it represents a set of resource packages for collaborative work to achieve resource complementarity and collaborative processing of tasks; is the number of resource clusters; determines the vehicle identification information, vehicle real-time dynamic information and vehicle transaction information corresponding to the simple charging task;

[0036] The formation of resource clusters takes into account the diversity of resources and the distribution characteristics of tasks to improve resource utilization and task processing efficiency. Each resource cluster can provide comprehensive services to meet the needs of different tasks and can quickly adjust resource allocation when task requirements change;

[0037] According to the vehicle identification information, vehicle real-time dynamic information and vehicle transaction information corresponding to the simple charging task, the computational complexity, data dependency, decision complexity and real-time requirements of the simple charging task are updated, and the updated computational complexity, updated data dependency, updated decision complexity and real-time requirements of the simple charging task are obtained;

[0038] Identify the requirements package for each simple billable task ;in, Including the update computation complexity, update data dependency, update decision complexity, update real-time requirements, security and privacy requirements, and processing requirements of a simple charging task; The number of simple fee-based tasks;

[0039] The fitness function of the multi-objective genetic algorithm is used to evaluate each demand package. Resource cluster collection adaptability;

[0040] Through the improved ant algorithm, according to each demand package Resource cluster collection The adaptability assessment results determine the requirements package To the resource cluster collection The mapping relationship of

[0041] Packages on demand To the resource cluster collection The mapping relationship is used to allocate the first edge resources of multiple network edge nodes in real time for simple charging tasks.

[0042] Among them, the multi-objective genetic algorithm, namely MOGA, is the full name of Multi-Objective Genetic Algorithm; the improved ant algorithm, namely IACO, is the full name of Improved Ant Colony Optimization;

[0043] Furthermore, the fitness function of the multi-objective genetic algorithm is used to evaluate the performance of each demand package. Resource cluster collection Adaptability; through the improved ant algorithm, according to each demand package Resource cluster collection The adaptability assessment results determine the requirements package To the resource cluster collection The mapping relationship is configured as:

[0044] In the first stage of processing simple charging tasks, the multi-objective genetic algorithm is used to generate initial pheromones; in the second stage of processing simple charging tasks, the ant colony information is exchanged through the improved ant algorithm, and the positive feedback mechanism is used to find the optimal solution for resource allocation; the second stage is after the first stage;

[0045] Among them, in the first stage of processing simple charging tasks, a fitness function is defined ;

[0046] ;

[0047] in, Demand Package With resource cluster collection The matching degree; Demand Package Real-time requirement function; Demand Package Security requirement function; , , are weight coefficients, such as 0.51, 0.47, and 0.62 respectively;

[0048] Using multi-objective genetic algorithm combined with fitness function , to evaluate each requirement package Resource cluster collection fitness and generate initial pheromones;

[0049] In the second stage of processing simple charging tasks, based on the initial pheromone, an improved ant algorithm is used to determine the demand package. To resource cluster The mapping relationship between ; where the pheromone update rule is defined To represent the dynamic changes of pheromones, as shown below: ;

[0050] in, is the current pheromone concentration; is the attenuation coefficient of the pheromone, which can be 0.12; is the amount of pheromone left by ants on the path. The amount of pheromone left by ants on the path is used to indicate the quality of the path. The path is the demand package. Resource allocation plan from source node to target node; source node is the task initiation point; target node is the task execution point; ants in the improved ant algorithm select resource clusters based on pheromone concentration and heuristic information, and update pheromones to reflect the newly found better path; among them, heuristic information includes the current load of resource clusters, network delay and the urgency of tasks.

[0051] Further, the resource allocation unit determines the first complex charging task from the complex charging tasks according to the allocation of the first edge resources, and allocates the second edge resources of the plurality of network edge nodes to the first complex charging task in real time, and is configured as follows:

[0052] Real-time monitoring of resource usage of each network edge node included in the edge module;

[0053] Determine the allocation of first edge resources according to resource usage of each network edge node;

[0054] According to the allocation of the first edge resources, determine the resource utilization of each network edge node included in the edge module and the task resource adaptability of the complex charging task ,Based on the resource relative performance quantification algorithm, the relative performance characteristics of each network edge node and each complex charging task are extracted;

[0055] ;

[0056] in, Indicates deadline pressure for the completion of complex charging tasks; is the average time for network edge nodes to process tasks; is the amount of resources required to process complex charging tasks; is the resource utilization of each network edge node;

[0057] Determine a first complex charging task from the complex charging tasks according to the extracted relative performance characteristics of each network edge node and each complex charging task;

[0058] Second edge resources of multiple network edge nodes are allocated in real time for the first complex charging task.

[0059] Based on the embodiments provided in the present application, by real-time monitoring of the resource usage of each network edge node in the edge terminal module, and extracting the relative performance characteristics of the network edge node and the complex charging task based on the resource relative performance quantification algorithm, the resource utilization and task resource adaptability of each node can be quantified more accurately. This quantification method helps to achieve refined management of resources and ensure the rationality and efficiency of resource allocation. The resource relative performance quantification algorithm enables the system to dynamically determine which tasks should be assigned to which nodes based on the performance characteristics and task requirements of each node. This dynamic allocation mechanism can reduce task completion time and improve the efficiency of task execution, especially in the processing of complex charging tasks, which can ensure that tasks are assigned to the most suitable nodes for execution.

[0060] Further, the cloud module is also used to determine the charging decision data and operation monitoring indicators of each lane according to the first processing result of the simple charging task, the second processing result and the third processing result of the first complex charging task; wherein the cloud module determines the charging decision data and operation monitoring indicators of each lane according to the first processing result of the simple charging task, the second processing result and the third processing result of the first complex charging task, and is configured as:

[0061] Receiving a first processing result of a simple charging task and a second processing result of a first complex charging task sent by the edge module;

[0062] Determine charging decision data and operation monitoring indicators of each lane according to the first processing result, the second processing result and the third processing result;

[0063] Among them, the billing decision data and operation monitoring indicators of each lane include: the actuarial results of vehicle tolls; the applicability assessment of vehicle preferential policies and their amounts; the completion status of vehicle transactions, including the specific status of successful transactions, pending or failed transactions; quantitative indicators of lane operation efficiency and real-time congestion analysis; automatic identification of transaction anomalies and risk management recommendations; detailed records of charging objections and tracking of processing progress.

[0064] Furthermore, simple toll collection tasks include toll calculation tasks based on a preset rate table, vehicle pass issuance tasks based on standard transaction procedures, and preliminary verification tasks of vehicle transaction information;

[0065] Complex toll collection tasks include transaction conflict resolution tasks involving multi-lane coordination, abnormal transaction detection tasks based on historical data analysis, applicability analysis tasks of preferential policies, and cross-regional vehicle path tracking and fee accounting tasks.

[0066] Furthermore, the cloud module is also used to feed back the charging decision data and operation monitoring indicators of each lane to the terminal module through the edge module;

[0067] The cloud module is also used to store lane information resources, which include preset rate tables and traffic blacklists.

[0068] Furthermore, vehicle identification information includes license plate number, vehicle model and vehicle characteristics; vehicle real-time dynamic information includes vehicle speed, timestamp and OBU position of the vehicle in the lane; vehicle transaction information includes ETC card information and payment status;

[0069] The terminal modules include self-service card issuance terminal, self-service payment terminal and special situation processing terminal.

[0070] Furthermore, the edge module also includes a phased array antenna unit, a smart booth, an automatic gate, an island head smart node, and an island tail smart node:

[0071] The phased array antenna unit includes a card reader and an electronic tag; the phased array antenna unit is used to locate the OBU position of the vehicle in the lane to solve the problems of adjacent lane interference and following vehicle interference;

[0072] The OBU location refers to the on-board unit, which is an electronic device installed on the vehicle, mainly used to communicate with the units beside the road to achieve information exchange between the vehicle and the infrastructure. Simply put, the OBU is a device installed on the vehicle for functions such as vehicle identification, electronic toll collection, traffic information collection, and safety warning. In the lane, the OBU location usually refers to the specific location where the OBU is installed on the vehicle. Generally, the OBU device is installed on the front windshield of the car.

[0073] The smart booth integrates self-service intelligent terminals, code scanning terminals, intelligent access control and air conditioning; the duty mode of the smart booth is switched according to the schedule and traffic conditions; the duty mode is manual mode or self-service mode; the automatic gate machine integrates electric barrier fog lights, warning lights and LED display screens; the LED display screen is used to display pass signs and speed limit signs; the island head intelligent node integrates wide-angle cameras, radars and the first display screen; the island tail intelligent node integrates license plate capture, radar, second display screen, vehicle inspection device and high-speed barrier.

[0074] Furthermore, the edge module is also used to notify the cloud module to perform remote maintenance when a lane fails or communication is interrupted.

[0075] Alternatively, if Figure 2 As shown, the present application provides a road detection method based on a highway intelligent gateway system, including:

[0076] S201, executing a highway road information detection task according to a resource allocation strategy of a highway intelligent gateway system; wherein the highway intelligent gateway system includes a terminal module, an edge module, and a cloud module;

[0077] S202, the terminal module collects lane-level information of each lane of the expressway and sends the lane-level information to the edge module; wherein the edge module is composed of a plurality of network edge nodes; the edge module includes a task classification unit and a resource allocation unit; the task classification unit is used to classify the road information detection tasks to obtain simple charging tasks and complex charging tasks;

[0078] S203, the resource allocation unit allocates first edge resources of multiple network edge nodes for the simple charging task in real time to obtain a first processing result of the simple charging task; determines a first complex charging task from the complex charging tasks according to the allocation of the first edge resources, allocates second edge resources of multiple network edge nodes for the first complex charging task in real time to obtain a second processing result of the first complex charging task; and sends the second complex charging task in the complex charging task except the first complex charging task to the cloud module;

[0079] S204, the cloud module allocates cloud resources to the second complex charging task in real time to obtain a third processing result.

[0080] It should be noted that in the present application, the embodiments implemented on the road detection system side based on the highway intelligent gateway system can be referenced with the embodiments implemented on the road detection method side based on the highway intelligent gateway system, and the present application will not repeat them one by one.

[0081] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. The road detection system based on the highway intelligent gateway system is characterized by: It includes a road detection module and a highway intelligent gateway system; wherein: The road detection module is used to perform the road information detection task of the expressway; The highway intelligent gateway system is used to dynamically configure resources for road information detection tasks; The highway intelligent gateway system includes a terminal module, an edge module, and a cloud module; The terminal module is used to collect lane-level information of each lane of the highway and send the lane-level information to the edge module; The edge module is composed of multiple network edge nodes; the edge module includes a task classification unit and a resource allocation unit; the task classification unit is used to classify road information detection tasks to obtain simple charging tasks and complex charging tasks; The resource allocation unit is used to allocate first edge resources of multiple network edge nodes for a simple charging task in real time to obtain a first processing result of the simple charging task; determine a first complex charging task from the complex charging tasks according to the allocation of the first edge resources, allocate second edge resources of multiple network edge nodes for the first complex charging task in real time to obtain a second processing result of the first complex charging task; and send a second complex charging task other than the first complex charging task in the complex charging tasks to the cloud module; The cloud module is used to allocate cloud resources to the second complex charging task in real time to obtain a third processing result; The lane-level information includes vehicle identification information, vehicle real-time dynamic information and vehicle transaction information; the task classification unit is specifically used to classify the toll collection tasks of each lane according to computational complexity, data dependency, decision complexity, real-time requirements, security and privacy requirements and processing requirements, and obtain the simple toll collection tasks and the complex toll collection tasks; the edge terminal module is deployed at the toll station; the edge terminal module is distributed between the terminal module and the cloud module; The resource allocation unit allocates first edge resources of multiple network edge nodes in real time for the simple charging task, and is configured as follows: Receiving vehicle identification information, vehicle real-time dynamic information and vehicle transaction information of each lane collected by the terminal module; Determine a network edge node set according to the plurality of network edge nodes included in the edge terminal module ; Each network edge node It has corresponding edge computing resources, edge storage resources, edge network resources, and edge task scheduling resources; is the number of network edge nodes; A multi-objective genetic algorithm is used to Perform clustering to obtain multiple resource package sets ; Each resource pack yes A subset of represents a group of edge nodes with similar resource characteristics; is the number of resource packages; the resource characteristics include geographical location or resource usage; The improved ant algorithm is used to optimize the connection between multiple resource packages to form a resource cluster set. ; Each resource cluster is a subset of P, each resource cluster Across different edge nodes, it represents a set of resource packages for collaborative work; is the number of resource clusters; Determine the vehicle identification information, vehicle real-time dynamic information and vehicle transaction information corresponding to the simple charging task; According to the vehicle identification information, vehicle real-time dynamic information and vehicle transaction information corresponding to the simple charging task, the computational complexity, data dependency, decision complexity and real-time requirements of the simple charging task are updated, and the updated computational complexity, updated data dependency and updated decision complexity of the simple charging task are obtained to meet the updating and real-time requirements; Identify the requirements package for each simple billable task ;in, Including the update computation complexity, update data dependency, update decision complexity, update real-time requirements, security and privacy requirements, and processing requirements of a simple charging task; The number of simple fee-charging tasks; The fitness function of the multi-objective genetic algorithm is used to evaluate each demand package. Resource cluster collection adaptability; Through the improved ant algorithm, according to each demand package Resource cluster collection The adaptability assessment results determine the requirements package To the resource cluster collection The mapping relationship of Packages on demand To the resource cluster collection A mapping relationship is established to allocate first edge resources of multiple network edge nodes in real time for the simple charging task.

2. The road detection system based on the highway intelligent gateway system according to claim 1 is characterized in that: The fitness function of the multi-objective genetic algorithm is used to evaluate each demand package. Resource cluster collection Adaptability; through the improved ant algorithm, according to each demand package Resource cluster collection The adaptability assessment results determine the requirements package To the resource cluster collection The mapping relationship is configured as: In the first stage of processing simple charging tasks, the multi-objective genetic algorithm is used to generate initial pheromones; in the second stage of processing simple charging tasks, the ant colony information is exchanged through the improved ant algorithm, and the positive feedback mechanism is used to find the optimal solution for resource allocation; the second stage is after the first stage; Among them, in the first stage of processing simple charging tasks, a fitness function is defined ; ； in, Demand Package With resource cluster collection The matching degree; Demand Package Real-time requirement function; Demand Package Security requirement function; , , is the weight coefficient; Using multi-objective genetic algorithm combined with fitness function , to evaluate each requirement package Resource cluster collection fitness and generate initial pheromones; In the second stage of processing simple charging tasks, based on the initial pheromone, an improved ant algorithm is used to determine the demand package. To resource cluster The mapping relationship between ; where the pheromone update rule is defined To represent the dynamic changes of pheromones, as shown below: ; in, is the current pheromone concentration; is the attenuation coefficient of the pheromone; is the amount of pheromone left by ants on the path. The amount of pheromone left by ants on the path is used to indicate the quality of the path. The path is the demand package. Resource allocation plan from source node to target node; source node is the task initiation point; target node is the task execution point; ants in the improved ant algorithm select resource clusters based on pheromone concentration and heuristic information, and update pheromones to reflect the newly found better path; among them, heuristic information includes the current load of resource clusters, network delay and the urgency of tasks.

3. The road detection system based on the highway intelligent gateway system according to claim 2 is characterized in that: The resource allocation unit determines the first complex charging task from the complex charging tasks according to the allocation of the first edge resources, and allocates the second edge resources of multiple network edge nodes to the first complex charging task in real time, and is configured as follows: Real-time monitoring of resource usage of each network edge node included in the edge module; Determine the allocation of first edge resources according to resource usage of each network edge node; According to the allocation of the first edge resources, determine the resource utilization of each network edge node included in the edge module and the task resource adaptability of the complex charging task ,Based on the resource relative performance quantification algorithm, the relative performance characteristics of each network edge node and each complex charging task are extracted; ； in, Indicates deadline pressure for the completion of complex charging tasks; is the average time for network edge nodes to process tasks; is the amount of resources required to process complex charging tasks; is the resource utilization of each network edge node; Determine a first complex charging task from the complex charging tasks according to the extracted relative performance characteristics of each network edge node and each complex charging task; Second edge resources of multiple network edge nodes are allocated in real time for the first complex charging task.

4. The road detection system based on the highway intelligent gateway system according to claim 3 is characterized in that: The cloud module is also used to determine the charging decision data and operation monitoring indicators of each lane according to the first processing result of the simple charging task, the second processing result and the third processing result of the first complex charging task; wherein the cloud module determines the charging decision data and operation monitoring indicators of each lane according to the first processing result of the simple charging task, the second processing result and the third processing result of the first complex charging task, and is configured as follows: Receive a first processing result of a simple charging task and a second processing result of a first complex charging task sent by the edge module; Determine charging decision data and operation monitoring indicators for each lane according to the first processing result, the second processing result and the third processing result; Among them, the billing decision data and operation monitoring indicators of each lane include: the actuarial results of vehicle tolls; the applicability assessment of vehicle preferential policies and their amounts; the completion status of vehicle transactions, including the specific status of successful transactions, pending or failed transactions; quantitative indicators of lane operation efficiency and real-time congestion analysis; automatic identification of transaction anomalies and risk management recommendations; detailed records of charging objections and tracking of processing progress.

5. The road detection system based on the highway intelligent gateway system according to claim 4 is characterized in that: Vehicle identification information includes license plate number, vehicle model and vehicle features; vehicle real-time dynamic information includes vehicle speed, timestamp and OBU position of the vehicle in the lane; vehicle transaction information includes ETC card information and payment status; The terminal modules include self-service card issuance terminal, self-service payment terminal and special situation processing terminal.

6. The road detection system based on the highway intelligent gateway system according to claim 5 is characterized in that: The cloud module is also used to feed back the charging decision data and operation monitoring indicators of each lane to the terminal module through the edge module; The cloud module is also used to store lane information resources, which include preset rate tables and blacklists.

7. The road detection system based on the highway intelligent gateway system according to claim 6 is characterized in that: The edge module also includes phased array antenna units, smart booths, automatic gate machines, island head smart nodes, and island tail smart nodes: Among them, the phased array antenna unit includes a card reader and an electronic tag; the phased array antenna unit is used to locate the OBU position of the vehicle in the lane; the smart booth integrates self-service intelligent terminals, code scanning terminals, intelligent access control and air conditioning; the duty mode of the smart booth is switched according to the schedule and traffic conditions; the duty mode is manual mode or self-service mode; the automatic gate machine integrates electric barrier fog lights, warning lights and LED display screens; the LED display screen is used to display pass signs and speed limit signs; the island head intelligent node integrates wide-angle cameras, radars and the first display screen; the island tail intelligent node integrates license plate capture, radar, the second display screen, vehicle inspection device and high-speed barrier.

8. The road detection system based on the highway intelligent gateway system according to claim 7 is characterized in that: The edge module is also used to notify the cloud module for remote maintenance when a lane fails or communication is interrupted.

9. A road detection method based on a highway intelligent gateway system, the method being implemented based on the system as claimed in claim 8, characterized in that: include: According to the resource allocation strategy of the highway intelligent gateway system, the highway road information detection task is performed; wherein the highway intelligent gateway system includes a terminal module, an edge module and a cloud module; The terminal module collects lane-level information of each lane of the highway and sends the lane-level information to the edge module; the edge module is composed of multiple network edge nodes; the edge module includes a task classification unit and a resource allocation unit; the task classification unit is used to classify the road information detection tasks to obtain simple toll tasks and complex toll tasks; The resource allocation unit allocates first edge resources of multiple network edge nodes in real time for a simple charging task to obtain a first processing result of the simple charging task; determines a first complex charging task from the complex charging tasks according to the allocation of the first edge resources, allocates second edge resources of multiple network edge nodes in real time for the first complex charging task to obtain a second processing result of the first complex charging task; and sends a second complex charging task other than the first complex charging task in the complex charging tasks to the cloud module; The cloud module allocates cloud resources to the second complex charging task in real time to obtain a third processing result.

Images