Intelligent road full-time and space instant simulation system

Abstract

A kind of wisdom road full-time space instant simulation system, including data acquisition subsystem, data transmission subsystem, edge computing subsystem, interface display subsystem, situation expansion subsystem;Data acquisition subsystem relies on the data provided by road camera, unmanned aerial vehicle fleet and large map database, data transmission subsystem establishes the two-dimensional transmission network of ground-ground, ground-air, edge computing subsystem provides real-time data processing, calculation and analysis services for road conditions by serving the built-in computer processor in road camera, unmanned aerial vehicle fleet of road, interface display subsystem provides simulation interface, can visually show road real-time situation, situation expansion subsystem provides hierarchical, intelligent interface selection for interface display subsystem.The beneficial effects of the present application are: real-time analysis of road information, through the deployment of road-air computing power, provide full-time space, intelligent, visual wisdom road service.

Description

Technical Field

[0001] This invention relates to the field of intelligent services and data processing, specifically to a real-time simulation system for intelligent roads. Background Technology

[0002] With rapid social development and accelerated urbanization, traffic congestion, road safety, and environmental pollution have become common challenges faced by many cities worldwide. In particular, with the continuous increase in vehicle ownership, traditional road traffic management methods are struggling to meet increasingly complex traffic demands. To address this challenge, Intelligent Transportation Systems (ITS) have emerged. ITS utilizes modern information, communication, and control technologies to achieve real-time monitoring, analysis, and management of traffic, thereby improving traffic efficiency, ensuring traffic safety, and reducing energy consumption. However, despite the application of ITS in many cities and regions, current traffic management systems still have some problems, especially in dealing with complex traffic scenarios, emergencies, and extreme weather conditions, where their intelligence, real-time capabilities, and predictive abilities are still insufficient. Most existing systems adopt a centralized processing approach, relying on a single data source or traditional sensor networks for traffic status monitoring and decision-making. This makes it difficult to achieve comprehensive, accurate, and real-time responses in certain environments and special circumstances.

[0003] While the application scope of intelligent transportation systems is expanding, many limitations remain. Traditional traffic monitoring systems primarily rely on fixed cameras and ground sensors on roads. These devices can only capture a portion of the road surface information, making it difficult to comprehensively perceive the complex conditions of the road. Furthermore, most of these systems are based on centralized processing architectures, with data collection and analysis distributed relatively fixedly. This results in the system's inability to react promptly to changing traffic environments or special scenarios, and its processing capacity is limited. For example, its traffic monitoring and accident handling capabilities are weak in specific environments such as mountain roads, tunnels, and bridges. To compensate for these shortcomings, many systems have begun to incorporate drone technology, providing dynamic monitoring from an aerial perspective. However, traditional drone applications are mostly limited to short-term, localized monitoring, lacking systematic and continuous data collection capabilities. Summary of the Invention

[0004] To address the aforementioned problems, this invention aims to provide a real-time simulation system for intelligent roads, thereby resolving the issues raised in the background section.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] This invention provides a real-time, all-weather simulation system for intelligent roads, comprising a data acquisition subsystem, a data transmission subsystem, an edge computing subsystem, an interface display subsystem, and a situational awareness expansion subsystem. The data acquisition subsystem includes a camera array module, a drone swarm collaboration module, and a large map module. The camera array module consists of cameras deployed on the roads, serving as the first pathway for online data acquisition by the data acquisition subsystem. Cameras are deployed on each road to provide real-time monitoring of road dynamics. All roads within a given area are monitored by the deployed cameras, forming a camera array that provides fundamental support for ground-to-ground monitoring of all roads within the data acquisition subsystem's area. Specifically, camera arrays are deployed along roadsides, intersections, tunnels, and under bridges to collect data under obstructed conditions. The drone swarm collaboration module serves as the second online data acquisition method for the data acquisition subsystem, primarily responsible for real-time ground-to-air monitoring. It can be deployed in blind spots of the camera arrays, enabling mobile deployment, and is particularly useful for special road conditions, mountainous road conditions, and disaster area road conditions. The large map module is the offline data acquisition method for the data acquisition subsystem, serving as its underlying database resource. The large map data is open-source, primarily sourced from KML files downloaded from BIGEMAP, and supplemented by historical road condition statistics, concurrent statistics, and other data. The system calculates the load on a specific road and during a specific time period based on the statistical data collected during the same period. Based on the load, it determines the elevation level for downloading. Higher elevation levels result in more detailed maps but also require more cache resources. The data transmission subsystem provides online data interconnection, including a gateway module and a link module. The gateway module transmits camera array data to the edge computing subsystem via internet protocols and fiber optic transmission. The link module transmits road video data collected by the drone swarm back to the edge computing subsystem via a downlink L-link. After receiving the data from the camera array and drone swarm, the edge computing subsystem... Through the IoT4Edge framework, intelligent data computation and processing at remote edges are achieved between ground and between ground and air. The numerical results calculated by the edge computing subsystem are visualized in the interface display subsystem. The situational awareness extension subsystem enhances the real-time performance, operability, and interactivity of the interface display subsystem. It includes a map elevation module and a map grid module. The map elevation module provides elevation map information to the situational awareness extension subsystem and provides differentiated processing for key monitoring areas, which can reduce loading and unloading of cache. The map grid module divides the map area in a gridded manner and provides users with small-area, dynamic, and high-precision display and interactive services through selection, color blocks, and dynamic simulation.

[0007] Furthermore, the data acquisition subsystem includes a camera array module, a drone swarm collaboration module, and a large map module. The camera array module collects real-time traffic dynamic information on the road through multiple cameras deployed on the road, including multiple high-definition cameras. The cameras are arranged along the road, at intersections, in tunnels, and under bridges to form a complete camera array, providing comprehensive support for real-time road monitoring. Within a region, all roads are covered by cameras, forming a complete ground-to-ground monitoring system. The cameras transmit road traffic status, road surface conditions, and weather information to the data transmission subsystem in real time through video streaming, providing basic data for subsequent data processing and decision-making.

[0008] Furthermore, the drone swarm collaboration module enables rapid deployment through drone swarms, flexibly responding to various emergencies, disasters, road conditions, and complex mountainous terrain. Drones can quickly enter blind spots of camera arrays and flexibly adjust their flight paths according to changes in road traffic conditions, achieving efficient collection of traffic flow, vehicle speed, and license plate recognition information. The drone swarm collaboration module has multi-drone collaborative combat capabilities, enabling monitoring over a wider area and real-time transmission of road information through coordinated flight, making it particularly suitable for handling large-scale traffic incidents and natural disaster emergencies.

[0009] Furthermore, the large map module relies on open-source data resources, especially the KML files provided by the BIGEMAP platform, and combines big data analytics to provide the system with accurate basic road data based on historical road statistics, concurrent traffic flow data, and road load conditions at different times. Through a high-precision Geographic Information System (GIS), the large map module provides rich spatial information support for edge computing and data processing. The level of map data refinement is dynamically adjusted according to needs. In road sections with high traffic load or during peak hours, the system automatically downloads higher-precision map data to facilitate more accurate road condition assessments in subsequent data processing and calculations. For areas with lower load and lower traffic flow, the map data precision can be appropriately reduced to minimize system resource consumption. The large map module also supports the download and processing of elevation information. Roads in different geographical areas may have significant elevation differences, especially in mountainous areas or urban elevated roads. By supporting multi-level elevation data downloads, the large map module can dynamically select and load different levels of map data based on different road conditions, thereby achieving accurate tracking and intelligent analysis of elevation changes.

[0010] Furthermore, the data transmission subsystem mainly consists of two modules: a gateway module and a link module. The gateway module aggregates, processes, and transmits the data collected by the camera array, and has data aggregation and forwarding functions. The camera array module transmits real-time monitoring data to the gateway module through the fiber optic network. The gateway module integrates and formats the data, and then transmits it to the edge computing subsystem. It supports TCP / IP, HTTP, and MQTT communication protocols. The gateway module enables the system to process data from different devices and ensures their interoperability.

[0011] Furthermore, the link module is responsible for ground-to-air data transmission, mainly handling the return transmission of road video data collected by the UAV swarm. It uses the L-band link in the 960-1215MHz frequency range to receive telemetry data. The L-band link supports high-bandwidth, low-latency transmission, ensuring that the high-definition video stream collected by the UAV is not distorted during the return transmission, while reducing latency and enabling real-time monitoring and analysis. Through optimized frequency selection and signal modulation, the L-band link can ensure the stability of the transmission link. Through frequency switching and intelligent signal adjustment, the L-band link can adapt to complex environments and reduce transmission errors caused by signal interference.

[0012] Furthermore, the edge computing subsystem processes, calculates, and analyzes the data collected by the data acquisition subsystem in real time. Due to the complexity and dynamism of the road traffic system, the edge computing subsystem needs to have high computing power, low latency, high real-time performance, and reliability. Especially in traffic management and disaster early warning scenarios, it needs to respond to system inputs in real time and generate efficient decision support. The edge computing subsystem is mainly composed of a camera array module and a drone swarm collaboration module. Combined with the IoT4Edge framework, it provides efficient and reliable data processing and service support through ground-to-ground and ground-to-air remote intelligent computing.

[0013] Furthermore, the IoT4Edge framework is specifically as follows:

[0014] (1) IoT4Edge framework based on the whole space

[0015] An IoT4Edge framework is built on a complex road (R1*R2). This complex road includes terrain features such as roadside sections, intersections, tunnels, underpasses, and mountainous areas. Roadside sections are primarily long straight lines, representing high-speed traffic (A1). Intersections are characterized by curves and poor road conditions, with cameras serving as the monitoring device, representing typical crossroad features (A2). Tunnels are primarily long straight roads whose external features are obscured by mountains, representing features where the road cannot be observed from the outside (A3), with cameras serving as the monitoring device. Underpasses primarily feature roads with multiple spatial configurations, representing road features in a unified vertical space (A4). Mountainous areas are characterized by rugged roads. The road A5 is characterized by its winding nature and difficulty in deploying cameras and network transmission equipment. The road monitoring deployment device is a drone. All roads within a given area are monitored by deployed cameras, forming a camera array B1. Designated locations in a special spatial area are monitored by deployed drones, forming a drone swarm B2. In the IoT4Edge framework based on full space, the special spatial area refers to a mountainous region, and the designated locations are the areas the user requests to survey. If a complex road of R1*R2 contains |B1| cameras and |B2| drones, with each camera having a uniform deployment height of h1, a uniform field of view of θ1 (where θ1∈[0,360]), a camera pitch angle of α1, and a camera monitoring coverage area of ​​[missing information], then [missing information]. The drone is deployed at an altitude of h2, its downward field of view is α2, and its monitoring range is [missing information]. Each camera can detect a horizontal distance of... In the IoT4Edge framework defined by the entire space, cameras are uniformly distributed, and drones are mobilely deployed. This framework divides roads along routes, intersections, tunnels, under bridges, and in mountainous areas into segments, meaning each segment... After dividing the scene into grids, with the camera's field of view uniformly set to θ1, where θ1∈[0,360], the road width can be ignored. Therefore, we have... The length of the road can be monitored by cameras. The computational offloading strategy of the full-space IoT4Edge framework is partial offloading. This means that based on the working status of occupied and idle cameras, if an idle camera is within the neighborhood of an occupied camera, the occupied camera will offload some tasks to the idle camera to accelerate the overall computing power of the framework. The framework includes two evaluation criteria: cost and computational performance. Cost is reflected in deployment cost, working time cost, and rental cost. Deployment cost is the cost incurred by deploying each camera; working time cost is the product of the unit time working cost and working time of each camera; rental cost is the cost incurred by the framework renting idle resources to accelerate task completion under the partial offloading strategy. Computational performance is reflected in the fact that each camera can be considered an independently computed edge computing node, and the energy consumption of these edge computing nodes is the computational performance. If the deployment cost of each camera is c... 11 Cost per unit of working time c 12 The working time is T1, and the unloading strategy for the i-th occupied camera is y. i ∈{0,1}, where y i =1 indicates that the i-th camera needs to unload its data to an idle node, y i =0 indicates that the i-th camera does not need to unload data to an idle node. Idle cameras in the neighborhood are in competition with each other, meaning that each camera wants to compete for the rental cost provided by camera i when it is idle. An equity incentive mechanism is used to select cameras in the neighborhood. If camera i's TA neighborhood contains S idle cameras, where the TA neighborhood is the neighborhood where camera i can establish communication range, the initial equity of each idle camera is... c rent To cover rental costs, the first round of the equity incentive mechanism involves a competitive selection process using Euclidean distance. Let the distance from the i-th occupied camera to the s-th free camera be denoted as...

[0016] The equity allocated in the first round of the Euclidean distance screening competition is now updated as follows: in Let d be the distance matrix. i1 Let d be the distance from the i-th occupied camera to the first free camera. i2 Let d be the distance from the i-th occupied camera to the second free camera. is Let x be the distance from the i-th occupied camera to the s-th idle camera. i Let x1 be the x-coordinate of the i-th occupied camera, x2 be the x-coordinate of the first idle camera, and x3 be the x-coordinate of the second idle camera. s Let y be the x-coordinate of the s-th idle camera. iLet y1 be the ordinate of the i-th occupied camera, y2 be the ordinate of the first idle camera, and y3 be the ordinate of the second idle camera. s Let be the ordinate of the s-th idle camera, SORT be an ascending function sorted from smallest to largest, and · be the dot product. The second round involves cost bidding, with the principle that idle cameras further away, due to higher transmission latency, will sacrifice cost to obtain tasks and compete for rental costs. If the rental cost is c... rent When an idle node is far from an occupied camera, meaning there is a probability that it will have a cheaper cost, the unloading probability p1 of the s-th idle camera is... The cost-based bidding competition has a discount factor γ, a decay factor of ε in each round, and the bottom-line cost of the s-th idle camera is... If the initial cost is r0, then the cost after the g-th round of bidding is... Lower cost cameras are more likely to be selected for unloading by the framework, so the probability p2 of the s-th idle camera being selected based on cost is:

[0017]

[0018] The overall unloading probability of the s-th idle camera is p1*p2, and the updated equity obtained through the equity incentive mechanism is as follows:

[0019]

[0020] in<R→s> This indicates that under the equity incentive mechanism, idle node s is selected to unload the camera, and the corresponding equity is R. The total cost of the camera C1 is C1 = |B1|(c 11 +c 12 T1)+∑ i y i <R→s> The camera's computing performance is Where, k i f is a constant of the camera chip architecture. i D represents the CPU cycle of the i-th camera, measured in revolutions per second. i Let a be the amount of data generated by the i-th camera. i The number of CPU cycles required to unload 1 bit of data;

[0021] (2) IoT4Edge framework based on all time

[0022] The IoT4Edge framework based on all-time data is characterized by: calculating road traffic conditions based on historical statistics over time, using a 24-hour standard, and then performing road condition statistics on the segmented roads with an hourly precision, recording each... Traffic flow over a period of 1 hour, with traffic flow thresholds set as δ1 and δ2, yields: Smooth roads refer to roads where the traffic flow through is less than δ1, i.e. On roads of considerable length, cameras remain idle for a 1-hour statistical period. Typical roads refer to roads where the traffic flow through thresholds are between δ1 and δ2. On roads of a certain length, the amount of data collected by cameras within a one-hour statistical period does not exceed the data processing capacity of the cameras within that one-hour statistical period. Congested roads refer to roads where the traffic flow through δ2 is higher than δ2. On a road of considerable length, if a camera remains occupied for 1 hour, and the amount of data collected by the camera exceeds its data processing capacity within that 1 hour, the task needs to be offloaded to another idle camera in the vicinity to complete the task.

[0023] The IoT4Edge framework, based on the entire time, stipulates that road congestion is counted every period. If a road is continuously counted as congested within a period, the IoT4Edge framework will suggest adding cameras to speed up the task processing capacity of that road and reduce rental costs.

[0024] The IoT4Edge framework, based on all-time conditions, stipulates that if the variance of road congestion exceeds the expected threshold during the same period within a cycle, indicating an irregular or special event that has caused a change in road conditions, the road conditions at the beginning of the cycle (congestion = 1, normal = 0.5, smooth = 0) and at the end of the cycle (congestion = 1, normal = 0.5, smooth = 0) must be recorded. If the values ​​in the road conditions are increasing, it indicates that the road conditions are trending towards congestion, and additional drones need to be deployed to assist the system in road monitoring.

[0025] Furthermore, the data visualization module transforms massive amounts of data obtained through cameras and drones into visual charts, maps, and dynamic animations to help users understand complex road information and make quick decisions. The system displays real-time road traffic conditions through dynamic maps, dynamically showing the real-time status of roads by refreshing the map in real time. The data displayed includes traffic flow, vehicle speed, road conditions, and weather warnings. On the map interface, users can select different view levels to view real-time data for the entire city or a specific area. Based on historical data, the system can predict the future traffic flow of a certain road, helping traffic managers to prepare contingency plans in advance.

[0026] Furthermore, the situational awareness enhancement subsystem is responsible for further processing and enhancing real-time road information based on the data display. It provides hierarchical and intelligent interface selection and operation to support more detailed and accurate decision-making and interactive operations. The situational awareness enhancement subsystem mainly includes a map elevation module and a map grid module, which respectively provide differentiated processing of elevation information and fine-grained division of map regions, enhancing the hierarchy, intelligence, and operability of road information. The map elevation module provides elevation information for the road simulation system, helping users conduct differentiated analysis in complex road environments. Through the processing of the elevation module, the system can accurately reflect the road elevations of different areas, providing more... To enhance refined and dynamic traffic management support, the map raster module divides road map areas into multiple grids, further refining the display of road information and providing users with high-precision, dynamic traffic information. Rasterization transforms continuous geographic data into discrete grid units, each representing a specific geographic area. Through the raster module, the system can dynamically select specific areas and information levels as needed to meet the needs of different users. The raster display can dynamically adjust the display area based on real-time data updates. When a traffic accident occurs, the raster data of the accident area is updated in real time, changing its color or status indicator and immediately reflected on the map. The system can also automatically expand or shrink the raster area according to actual needs to focus on displaying the traffic conditions of a specific hotspot. In the event of a major traffic incident, the raster module can zoom in on specific areas to provide more detailed information, facilitating emergency response by management personnel.

[0027] This invention proposes two IoT4Edge frameworks: one based on full space and the other on full time. These frameworks encompass terrain features along roads, at intersections, in tunnels, under bridges, and in mountainous areas. In the full-space IoT4Edge framework, frame segmentation enhances the statistical granularity of the entire system. It also considers camera deployment schemes and edge collaboration among idle nodes, using cost and computational efficiency as evaluation metrics. An equity incentive mechanism is employed to competitively select cameras within the neighborhood, thereby increasing the utilization rate of idle resources and mobilizing the activity of the entire system's spatial computing resources. The incentive mechanism involved two rounds of competition. The first round used Euclidean distance for selection, prioritizing the closest nodes within the communication range. The second round involved cost bidding, where idle cameras further away would sacrifice cost to secure tasks and lower rental costs due to higher transmission latency. Ultimately, the equity incentive mechanism considered both factors and selected a reasonably priced node that was not too far away as the offloading node. This reduced the load on the occupied camera resources and lowered the overall rental cost of the system, resulting in cheaper and more reliable service. In the all-time-based IoT4Edge framework, road traffic conditions are statistically analyzed based on historical data over a 24-hour period. The segmented roads are then statistically analyzed with an hourly precision. Road congestion is assessed every cycle. If a road is consistently classified as congested within a cycle, the all-time-based IoT4Edge framework recommends adding cameras to accelerate processing and reduce rental costs. If the variance of road congestion exceeds a desired threshold within a cycle, indicating an irregular or unusual event causing a change in road conditions, the road conditions at the start of the cycle must be recorded.

[0028] (Congestion = 1, Normal = 0.5, Smooth = 0) and the road conditions at the end of the cycle (Congestion = 1, Normal = 0.5, Smooth = 0). If the value in the road conditions is increasing, it means that the road conditions are trending towards congestion, and it is necessary to deploy more drones to assist the system in road monitoring. Based on the all-time IoT4Edge framework, by statistically analyzing historical data and the variance of statistics within the cycle, it intelligently evaluates whether the camera resources in the original deployment scenario are deployed reasonably, whether it is necessary to increase camera resources, or to flexibly dispatch drone resources for short-term road reconnaissance, bringing intelligent choices for flexible road deployment.

[0029] The system's innovations in visualization are also highly valuable. Through an integrated interface display subsystem, the system can present massive amounts of traffic information and data to users in an intuitive and easy-to-use interface. Users can view real-time traffic flow, vehicle speed, and road condition information for the entire city or specific roads. Simultaneously, the system's interactivity allows users to query, analyze, and manipulate information as needed. This visualization not only enhances users' intuitive understanding of road conditions but also enables traffic management personnel to make decisions and manage more efficiently. Especially in emergencies, operators can quickly understand and manage road conditions, avoiding human error and ensuring timely emergency response. The situational awareness extension subsystem of this invention provides users with more accurate and dynamic traffic information displays by introducing a map elevation module and a map grid module. The elevation module combines road topographic relief data, enabling differentiated processing of road conditions in different geographical environments. Particularly in complex mountainous areas, tunnels, and bridge areas, the system can provide more accurate road condition information, helping traffic management personnel identify potential risk points and adjust traffic strategies in a timely manner. The map raster module, by dividing the map area into several smaller blocks, provides users with more refined local area data displays. Especially in key areas with high traffic volume and frequent accidents, it can accurately display road conditions through rasterization and provide dynamic analysis and predictions based on the data of that area. This hierarchical and detailed display method enhances the interactivity and intelligence of the system, enabling users to react quickly when faced with complex road conditions. Attached Figure Description

[0030] The invention will be further illustrated with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the invention. For those skilled in the art, other drawings can be obtained based on the following drawings without any creative effort.

[0031] Figure 1 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] Please see Figure 1 The present invention will be further described in conjunction with the following examples.

[0034] See Figure 1This invention aims to provide a real-time, all-weather simulation system for intelligent roads, comprising a data acquisition subsystem, a data transmission subsystem, an edge computing subsystem, an interface display subsystem, and a situational awareness expansion subsystem. The data acquisition subsystem includes a camera array module, a drone swarm collaboration module, and a large map module. The camera array module consists of cameras deployed on the roads, serving as the first pathway for online data acquisition by the data acquisition subsystem. Cameras are deployed on each road to provide real-time monitoring of road dynamics. All roads within a given area are monitored by the deployed cameras, forming a camera array that provides basic ground-to-ground monitoring of all roads within the data acquisition subsystem's area. In particular, camera arrays are deployed along roadsides, intersections, tunnels, and under bridges to collect data under obstructed conditions. The drone swarm collaboration module is the second online data acquisition method for the data acquisition subsystem, primarily responsible for real-time ground-to-air monitoring. It can be deployed in blind spots of the camera arrays, enabling mobile deployment, and is particularly useful for special road conditions, mountainous road conditions, and disaster area road conditions. The large map module is the offline data acquisition method for the data acquisition subsystem. It is the underlying database resource of the data acquisition subsystem. The large map data is open-source, primarily sourced from KML files downloaded from BIGEMAP, and supplemented with historical and concurrent statistics on road conditions. Based on the statistical data of the same period, the load volume of a specific road and time period is determined. The download level for elevation information is then decided based on the load volume. Higher elevation levels result in more detailed maps but also require more cache resources. The data transmission subsystem provides online data interconnection, including a gateway module and a link module. The gateway module transmits camera array data to the edge computing subsystem via internet protocols and fiber optic transmission. The link module transmits road video data collected by the drone swarm back to the edge computing subsystem via downlink L-link transmission. After receiving the data from the camera array and drone swarm, the edge computing subsystem... Through the IoT4Edge framework, intelligent data computation and processing at remote edges are achieved between ground and between ground and air. The numerical results calculated by the edge computing subsystem are visualized in the interface display subsystem. The situational awareness extension subsystem enhances the real-time performance, operability, and interactivity of the interface display subsystem. It includes a map elevation module and a map grid module. The map elevation module provides elevation map information to the situational awareness extension subsystem and provides differentiated processing for key monitoring areas, which can reduce loading and unloading of cache. The map grid module divides the map area in a gridded manner and provides users with small-area, dynamic, and high-precision display and interactive services through selection, color blocks, and dynamic simulation.

[0035] Specifically, the data acquisition subsystem includes a camera array module, a drone swarm collaboration module, and a large map module. The camera array module collects real-time traffic dynamic information on the road through multiple cameras deployed on the road, including multiple high-definition cameras. The cameras are arranged along the road, at intersections, in tunnels, and under bridges to form a complete camera array, providing comprehensive support for real-time road monitoring. Within a region, all roads are covered by cameras, forming a complete ground-to-ground monitoring system. The cameras transmit road traffic status, road surface conditions, and weather information to the data transmission subsystem in real time through video streaming, providing basic data for subsequent data processing and decision-making.

[0036] Specifically, the drone swarm collaboration module enables rapid deployment through drone swarms, flexibly responding to various emergencies, disasters, road conditions, and complex mountainous terrain. Drones can quickly enter blind spots of camera arrays and flexibly adjust their flight paths according to changes in road traffic conditions, achieving efficient collection of traffic flow, vehicle speed, and license plate recognition information. The drone swarm collaboration module has multi-drone collaborative combat capabilities, enabling monitoring over a wider area and real-time transmission of road information through coordinated flight. It is particularly suitable for handling large-scale traffic incidents and natural disaster emergencies.

[0037] Specifically, the large map module relies on open-source data resources, especially the KML files provided by the BIGEMAP platform. Combined with big data analytics, it provides the system with accurate basic road data based on historical road statistics, concurrent traffic flow data, and road load conditions at different times. Through a high-precision Geographic Information System (GIS), the large map module provides rich spatial information support for edge computing and data processing. The level of map data refinement is dynamically adjusted according to needs. In areas with high traffic load or during peak hours, the system automatically downloads higher-precision map data to facilitate more accurate road condition assessments in subsequent data processing and calculations. For areas with lower load and lower traffic flow, the map data precision can be appropriately reduced to minimize system resource consumption. The large map module also supports the download and processing of elevation information. Roads in different geographical areas may have significant elevation differences, especially in mountainous areas or urban elevated roads. By supporting multi-level elevation data downloads, the large map module can dynamically select and load different levels of map data based on different road conditions, thereby achieving accurate tracking and intelligent analysis of elevation changes.

[0038] Specifically, the data transmission subsystem mainly consists of two modules: a gateway module and a link module. The gateway module aggregates, processes, and transmits the data collected by the camera array, and has data aggregation and forwarding functions. The camera array module transmits real-time monitoring data to the gateway module through a fiber optic network. The gateway module integrates and formats the data, and then transmits it to the edge computing subsystem. It supports TCP / IP, HTTP, and MQTT communication protocols. The gateway module enables the system to process data from different devices and ensures their interoperability.

[0039] Specifically, the link module is responsible for ground-to-air data transmission, mainly handling the return transmission of road video data collected by the UAV swarm. It uses the L-band link in the 960-1215MHz frequency range to receive telemetry data. The L-band link supports high-bandwidth, low-latency transmission, ensuring that the high-definition video stream collected by the UAV is not distorted during the return transmission, while reducing latency and enabling real-time monitoring and analysis. Through optimized frequency selection and signal modulation, the L-band link can ensure the stability of the transmission link. Through frequency switching and intelligent signal adjustment, the L-band link can adapt to complex environments and reduce transmission errors caused by signal interference.

[0040] Specifically, the edge computing subsystem processes, calculates, and analyzes the data collected by the data acquisition subsystem in real time. Due to the complexity and dynamism of the road traffic system, the edge computing subsystem needs to have high computing power, low latency, high real-time performance, and high reliability. Especially in traffic management and disaster early warning scenarios, it needs to respond to system inputs in real time and generate efficient decision support. The edge computing subsystem is mainly composed of a camera array module and a drone swarm collaboration module. Combined with the IoT4Edge framework, it provides efficient and reliable data processing and service support through ground-to-ground and ground-to-air remote intelligent computing.

[0041] Specifically, the IoT4Edge framework is as follows:

[0042] (1) IoT4Edge framework based on the whole space

[0043] An IoT4Edge framework is built on a complex road (R1*R2). This complex road includes terrain features such as roadside sections, intersections, tunnels, underpasses, and mountainous areas. Roadside sections are primarily long straight lines, representing high-speed traffic (A1). Intersections are characterized by curves and poor road conditions, with cameras serving as the monitoring device, representing typical crossroad features (A2). Tunnels are primarily long straight roads whose external features are obscured by mountains, representing features where the road cannot be observed from the outside (A3), with cameras serving as the monitoring device. Underpasses primarily feature roads with multiple spatial configurations, representing road features in a unified vertical space (A4). Mountainous areas are characterized by rugged roads. The road A5 is characterized by its winding nature and difficulty in deploying cameras and network transmission equipment. The road monitoring deployment device is a drone. All roads within a given area are monitored by deployed cameras, forming a camera array B1. Designated locations in a special spatial area are monitored by deployed drones, forming a drone swarm B2. In the IoT4Edge framework based on full space, the special spatial area refers to a mountainous region, and the designated locations are the areas the user requests to survey. If a complex road of R1*R2 contains |B1| cameras and |B2| drones, with each camera having a uniform deployment height of h1, a uniform field of view of θ1 (where θ1∈[0,360]), a camera pitch angle of α1, and a camera monitoring coverage area of ​​[missing information], then [missing information]. The drone is deployed at an altitude of h2, its downward field of view is α2, and its monitoring range is [missing information]. Each camera can detect a horizontal distance of... In the IoT4Edge framework defined by the entire space, cameras are uniformly distributed, and drones are mobilely deployed. This framework divides roads along routes, intersections, tunnels, under bridges, and in mountainous areas into segments, meaning each segment... After dividing the scene into grids, with the camera's field of view uniformly set to θ1, where θ1∈[0,360], the road width can be ignored. Therefore, we have... The length of the road can be monitored by cameras. The computational offloading strategy of the full-space IoT4Edge framework is partial offloading. This means that based on the working status of occupied and idle cameras, if an idle camera is within the neighborhood of an occupied camera, the occupied camera will offload some tasks to the idle camera to accelerate the overall computing power of the framework. The framework includes two evaluation criteria: cost and computational performance. Cost is reflected in deployment cost, working time cost, and rental cost. Deployment cost is the cost incurred by deploying each camera; working time cost is the product of the unit time working cost and working time of each camera; rental cost is the cost incurred by the framework renting idle resources to accelerate task completion under the partial offloading strategy. Computational performance is reflected in the fact that each camera can be considered an independently computed edge computing node, and the energy consumption of these edge computing nodes is the computational performance. If the deployment cost of each camera is c... 11 Cost per unit of working time c 12 The working time is T1, and the unloading strategy for the i-th occupied camera is y. i ∈{0,1}, where y i =1 indicates that the i-th camera needs to unload its data to an idle node, y i =0 indicates that the i-th camera does not need to unload data to an idle node. Idle cameras in the neighborhood are in competition with each other, meaning that each camera wants to compete for the rental cost provided by camera i when it is idle. An equity incentive mechanism is used to select cameras in the neighborhood. If camera i's TA neighborhood contains S idle cameras, where the TA neighborhood is the neighborhood where camera i can establish communication range, the initial equity of each idle camera is... c rent To cover rental costs, the first round of the equity incentive mechanism involves a competitive selection process using Euclidean distance. Let the distance from the i-th occupied camera to the s-th free camera be denoted as...

[0044] The equity allocated in the first round of the Euclidean distance screening competition is now updated as follows: in Let d be the distance matrix. i1 Let d be the distance from the i-th occupied camera to the first free camera. i2 Let d be the distance from the i-th occupied camera to the second free camera. is Let x be the distance from the i-th occupied camera to the s-th idle camera. i Let x1 be the x-coordinate of the i-th occupied camera, x2 be the x-coordinate of the first idle camera, and x3 be the x-coordinate of the second idle camera. s Let y be the x-coordinate of the s-th idle camera. iLet y1 be the ordinate of the i-th occupied camera, y2 be the ordinate of the first idle camera, and y3 be the ordinate of the second idle camera. s Let be the ordinate of the s-th idle camera, SORT be an ascending function sorted from smallest to largest, and · be the dot product. The second round involves cost bidding, with the principle that idle cameras further away, due to higher transmission latency, will sacrifice cost to obtain tasks and compete for rental costs. If the rental cost is c... rent When an idle node is far from an occupied camera, meaning there is a probability that it will have a cheaper cost, the unloading probability p1 of the s-th idle camera is... The cost-based bidding competition has a discount factor γ, a decay factor of ε in each round, and the bottom-line cost of the s-th idle camera is... If the initial cost is r0, then the cost after the g-th round of bidding is... Lower cost cameras are more likely to be selected for unloading by the framework, so the probability p2 of the s-th idle camera being selected based on cost is:

[0045] The overall unloading probability of the s-th idle camera is p1*p2, and the updated equity obtained through the equity incentive mechanism is as follows:

[0046]

[0047] in<R→s> This indicates that under the equity incentive mechanism, idle node s is selected to unload the camera, and the corresponding equity is R. The total cost of the camera C1 is C1 = |B1|(c 11 +c 12 T1)+∑ i y i <R→s> The camera's computing performance is Where, k i f is a constant of the camera chip architecture. i D represents the CPU cycle of the i-th camera, measured in revolutions per second. i Let a be the amount of data generated by the i-th camera. i The number of CPU cycles required to unload 1 bit of data;

[0048] (2) IoT4Edge framework based on all time

[0049] The IoT4Edge framework based on all-time data is characterized by: calculating road traffic conditions based on historical statistics over time, using a 24-hour standard, and then performing road condition statistics on the segmented roads with an hourly precision, recording each... Traffic flow over a period of 1 hour, with traffic flow thresholds set as δ1 and δ2, yields: Smooth roads refer to roads where the traffic flow through is less than δ1, i.e. On roads of considerable length, cameras remain idle for a 1-hour statistical period. Typical roads refer to roads where the traffic flow through thresholds are between δ1 and δ2. On roads of a certain length, the amount of data collected by cameras within a one-hour statistical period does not exceed the data processing capacity of the cameras within that one-hour statistical period. Congested roads refer to roads where the traffic flow through δ2 is higher than δ2. On a road of considerable length, if a camera remains occupied for 1 hour, and the amount of data collected by the camera exceeds its data processing capacity within that 1 hour, the task needs to be offloaded to another idle camera in the vicinity to complete the task.

[0050] The IoT4Edge framework, based on the entire time, stipulates that road congestion is counted every period. If a road is continuously counted as congested within a period, the IoT4Edge framework will suggest adding cameras to speed up the task processing capacity of that road and reduce rental costs.

[0051] The IoT4Edge framework, based on all-time conditions, stipulates that if the variance of road congestion exceeds the expected threshold during the same period within a cycle, indicating an irregular or special event that has caused a change in road conditions, the road conditions at the beginning of the cycle (congestion = 1, normal = 0.5, smooth = 0) and at the end of the cycle (congestion = 1, normal = 0.5, smooth = 0) must be recorded. If the values ​​in the road conditions are increasing, it indicates that the road conditions are trending towards congestion, and additional drones need to be deployed to assist the system in road monitoring.

[0052] Specifically, the data visualization module transforms massive amounts of data obtained from cameras and drones into visual charts, maps, and dynamic animations to help users understand complex road information and make quick decisions. The system displays real-time traffic conditions through dynamic maps, dynamically showing the real-time status of roads by updating the map in real time. The data displayed includes traffic flow, vehicle speed, road conditions, and weather warnings. On the map interface, users can select different view levels to view real-time data for the entire city or a specific area. Based on historical data, the system can predict future traffic flow on a particular road, helping traffic managers to prepare contingency plans in advance.

[0053] Specifically, the situational awareness enhancement subsystem is responsible for further processing and enhancing real-time road information based on the data display. It provides hierarchical and intelligent interface selection and operation to support more detailed and accurate decision-making and interactive operations. The situational awareness enhancement subsystem mainly includes a map elevation module and a map grid module, which respectively provide differentiated processing of elevation information and fine-grained division of map regions, enhancing the hierarchy, intelligence, and operability of road information. The map elevation module provides elevation information for the road simulation system, helping users conduct differentiated analysis in complex road environments. Through the processing of the elevation module, the system can accurately reflect the road elevations of different areas, providing more... The map raster module provides refined and dynamic traffic management support, dividing the road map area into multiple grids to further refine the display of road information and provide users with high-precision, dynamic traffic information. Rasterization transforms continuous geographic data into discrete grid units, each representing a specific geographic area. Through the raster module, the system can dynamically select specific areas and information levels as needed to meet the needs of different users. The raster display can be dynamically adjusted based on real-time data updates. When a traffic accident occurs, the raster data of the accident area is updated in real time, changing its color or status indicator and immediately reflected on the map. The system can also automatically expand or shrink the raster area according to actual needs to focus on displaying the traffic conditions of a specific hotspot. In the event of a major traffic incident, the raster module can zoom in on specific areas to provide more detailed information, facilitating emergency response by management personnel.

[0054] The beneficial effects of this invention are as follows: This invention provides all-weather, intelligent traffic management. Unlike traditional intelligent transportation systems that mostly rely on fixed ground sensors and cameras, this invention overcomes the limitations of traditional monitoring methods by combining ground camera arrays with aerial monitoring by drone swarms, achieving all-weather data collection and road condition monitoring. Especially in complex road environments, such as mountainous areas, bridges, and tunnels, traditional equipment often has blind spots and signal dead zones, while drone swarms can flexibly and mobilely cover these hard-to-reach areas, ensuring comprehensive real-time road information collection. This combined air-ground data collection method provides unprecedented accuracy and flexibility for traffic management, enabling the system to cope with more complex and changing traffic conditions.

[0055] Secondly, the system's dual-dimensional data transmission network design significantly improves data transmission efficiency and real-time performance. In traditional systems, data transmission typically relies on a single terrestrial communication network. This invention, however, combines L-links and gateway modules to employ a dual-dimensional ground-to-ground and ground-to-air transmission approach, making data transmission from various sensors to edge computing nodes more stable and faster. Especially when UAV swarms transmit data back, the L-link's frequency range (960-1215MHz) ensures strong anti-interference capabilities and greater bandwidth, providing stable signal transmission in various environments. This data transmission method not only guarantees the system's real-time performance and efficiency but also provides strong support for network scalability during large-scale deployments.

[0056] This invention proposes two IoT4Edge frameworks: one based on full space and the other on full time. These frameworks encompass terrain features along roads, at intersections, in tunnels, under bridges, and in mountainous areas. In the full-space IoT4Edge framework, frame segmentation enhances the statistical granularity of the entire system. It also considers camera deployment schemes and edge collaboration among idle nodes, using cost and computational efficiency as evaluation metrics. An equity incentive mechanism is employed to competitively select cameras within the neighborhood, thereby increasing the utilization rate of idle resources and mobilizing the activity of the entire system's spatial computing resources. The incentive mechanism involved two rounds of competition. The first round used Euclidean distance for selection, prioritizing the closest nodes within the communication range. The second round involved cost bidding, where idle cameras further away would sacrifice cost to secure tasks and lower rental costs due to higher transmission latency. Ultimately, the equity incentive mechanism considered both factors and selected a reasonably priced node that was not too far away as the offloading node. This reduced the load on the occupied camera resources and lowered the overall rental cost of the system, resulting in cheaper and more reliable service. In the all-time-based IoT4Edge framework, road traffic conditions are statistically analyzed based on historical data over a 24-hour period. The segmented roads are then statistically analyzed with an hourly precision. Road congestion is assessed every cycle. If a road is consistently classified as congested within a cycle, the all-time-based IoT4Edge framework recommends adding cameras to accelerate processing and reduce rental costs. If the variance of road congestion exceeds a desired threshold within a cycle, indicating an irregular or unusual event causing a change in road conditions, the road conditions at the start of the cycle must be recorded.

[0057] (Congestion = 1, Normal = 0.5, Smooth = 0) and the road conditions at the end of the cycle (Congestion = 1, Normal = 0.5, Smooth = 0). If the value in the road conditions is increasing, it means that the road conditions are trending towards congestion, and it is necessary to deploy more drones to assist the system in road monitoring. Based on the all-time IoT4Edge framework, by statistically analyzing historical data and the variance of statistics within the cycle, it intelligently evaluates whether the camera resources in the original deployment scenario are deployed reasonably, whether it is necessary to increase camera resources, or to flexibly dispatch drone resources for short-term road reconnaissance, bringing intelligent choices for flexible road deployment.

[0058] The system's innovations in visualization are also highly valuable. Through an integrated interface display subsystem, the system can present massive amounts of traffic information and data to users in an intuitive and easy-to-use interface. Users can view real-time traffic flow, vehicle speed, and road condition information for the entire city or specific roads. Simultaneously, the system's interactivity allows users to query, analyze, and manipulate information as needed. This visualization not only enhances users' intuitive understanding of road conditions but also enables traffic management personnel to make decisions and manage more efficiently. Especially in emergencies, operators can quickly understand and manage road conditions, avoiding human error and ensuring timely emergency response. The situational awareness extension subsystem of this invention provides users with more accurate and dynamic traffic information displays by introducing a map elevation module and a map grid module. The elevation module combines road topographic relief data, enabling differentiated processing of road conditions in different geographical environments. Particularly in complex mountainous areas, tunnels, and bridge areas, the system can provide more accurate road condition information, helping traffic management personnel identify potential risk points and adjust traffic strategies in a timely manner. The map raster module, by dividing the map area into several smaller blocks, provides users with more refined local area data displays. Especially in key areas with high traffic volume and frequent accidents, it can accurately display road conditions through rasterization and provide dynamic analysis and predictions based on the data of that area. This hierarchical and detailed display method enhances the interactivity and intelligence of the system, enabling users to react quickly when faced with complex road conditions.

[0059] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make substitutions for some of the technical features. Any modifications, substitutions, or improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1.A smart road full-time and space instant simulation system, comprising a data acquisition subsystem, a data transmission subsystem, an edge computing subsystem, an interface display subsystem, and a situation expansion subsystem; The data acquisition subsystem comprises a camera array module, a UAV fleet cooperation module, and a large map module. Cameras are deployed on each road to provide real-time dynamic monitoring services. All roads in a region are monitored by the deployed cameras, forming a camera array. The camera array is deployed along the road, at intersections, in tunnels, and under bridges to collect data under obstructed conditions. The UAV fleet cooperation module is responsible for real-time ground-air monitoring and is deployed in the monitoring dead zone of the camera array to achieve mobile deployment and be applied to road emergencies, mountainous road conditions, and disaster areas. The large map module is an offline data acquisition approach for the data acquisition subsystem. The data sources for the large map data include KML files downloaded from BIGEMAP. According to the statistical data of previous years, the same period, and the same period, the load of a certain road in a certain period is arranged. According to the size of the load, the download of the elevation level is decided. The data transmission subsystem comprises a gateway module and a link module. The gateway module transmits camera array data to the edge computing subsystem through the Internet protocol. The transmission mode is optical fiber transmission. The link module transmits road video data collected by the UAV fleet to the edge computing subsystem through the L chain in the downlink mode. After receiving the data transmitted by the camera array and the UAV fleet, the edge computing subsystem realizes remote edge intelligent data calculation and processing through the IoT4Edge framework. The numerical results calculated by the edge computing subsystem are visualized and displayed on the interface display subsystem. The situation expansion subsystem comprises a map elevation module and a map grid module. The map elevation module provides elevation map information for the situation expansion subsystem. The map grid module separates the map area in a rasterization manner and provides interactive display services for users in a selection, color block, and dynamic deduction manner. The IoT4Edge framework includes a full-space-based IoT4Edge framework and a full-time-based IoT4Edge framework, and the calculation offloading strategy of the full-space-based IoT4Edge framework is partial offloading, that is, according to the working conditions of the occupied camera and the idle camera, if the idle camera is in the neighborhood of the occupied camera, the occupied camera will offload part of the task to the idle camera to speed up the calculation processing capacity of the entire framework, wherein, The IoT4Edge framework based on full space includes two evaluation criteria, namely cost and computing efficiency. The cost includes deployment cost, working time cost, and rental cost. The IoT4Edge framework based on full time stipulates that road congestion is counted every period. If a road is continuously counted as congested within a period, more cameras are added to speed up the processing capacity of the road. If the variance of road congestion in the same period within a period exceeds the expected threshold, the road conditions at the beginning and end of the period are recorded. If the numerical value in the road condition increases, it represents a trend of road congestion, and more UAV fleets are dispatched to assist in road monitoring. 2.The intelligent road full-time and space instant simulation system according to claim 1, wherein, The camera array module collects real-time traffic dynamic information on the road through multiple cameras deployed on the road, including multiple high-definition cameras. The cameras are arranged along the road, intersections, tunnels, and under bridges to form a complete camera array, providing comprehensive support for real-time monitoring of the road. In a region, all roads are covered by cameras, forming a complete ground-ground monitoring system. The cameras transmit real-time video streams to the data transmission subsystem, conveying road traffic status, road conditions, and weather information to provide basic data for subsequent data processing and decision-making. 3.The intelligent road full-time and space instant simulation system according to claim 2, characterized in that, The UAV fleet coordination module enables rapid deployment through a fleet of UAVs that can enter the blind spots of the camera array monitoring system. Based on changes in road traffic conditions, the UAVs adjust their flight paths to collect traffic flow, speed, and license plate recognition information. The UAV fleet coordination module has multi-vehicle cooperative combat capability and transmits real-time road information. 4.The intelligent road full-time and space instant simulation system according to claim 2, wherein, The large map module relies on open-source data resources, including KML files provided by the BIGEMAP platform, and combines big data analysis techniques to provide road base data based on historical statistical data, current traffic flow data, and road load data at different times. The large map module provides spatial information support for edge computing and data processing through high-precision GIS. The level of detail of the map data is dynamically adjusted according to demand. The large map module also supports elevation information download and processing. By supporting multi-level elevation data download, the large map module can dynamically select and load different levels of map data based on different road conditions. 5.The intelligent road full-time and space instant simulation system according to claim 1, wherein, The data transmission subsystem consists of two modules: the gateway module and the link module. The gateway module collects, processes, and transmits data collected by the camera array, with data aggregation and forwarding functions. The camera array module transmits real-time monitoring data to the gateway module through a fiber optic network. The gateway module integrates and formats the data and then transmits it to the edge computing subsystem, supporting TCP / IP, HTTP, and MQTT communication protocols. The gateway module enables the system to process data from different devices, ensuring interoperability. 6.The intelligent road full-time and space instant simulation system according to claim 1, wherein, The link module is responsible for ground-air data transmission and handles the return of road video data collected by the UAV fleet. It uses L-band links with a frequency range of 960-1215MHz to receive telemetry data. L-band links support high-bandwidth, low-latency transmission. 7.The intelligent road full-time and space instant simulation system according to claim 1, wherein, The edge computing subsystem processes, calculates, and analyzes data collected by the data acquisition subsystem in real time. The edge computing subsystem includes the camera array module and the UAV fleet coordination module, and combines the IoT4Edge framework for ground-ground and ground-air remote intelligent computing. 8.The intelligent road full-time and space instant simulation system according to claim 1, wherein, The data visualization module converts the data obtained by the camera and the unmanned aerial vehicle into visual charts, maps, and dynamic animations. The system dynamically displays the real-time state of the road by refreshing the map in real time. The data display includes traffic flow, vehicle speed, road conditions, and weather warnings. In the map interface, users can select different view levels to view real-time data in the entire city or a local area. The system can predict future traffic on a certain road based on historical data. 9.The intelligent road full-time and space instant simulation system according to claim 1, wherein, The situation expansion subsystem is responsible for further processing of real-time road information based on data display, providing hierarchical and intelligent interface selection and operation. The situation expansion subsystem includes a map elevation module and a map grid module, which provide differentiated processing of elevation information and fine division of map areas, respectively. The map elevation module provides elevation information for the road simulation system, and the map grid module is responsible for dividing the road map area into multiple grids to refine the display of road information. Grid processing converts continuous geographic information data into discrete grid cells, with each grid representing a specific geographic area. Through the grid module, the system can dynamically select specific areas and information levels as needed. The grid display can be updated based on real-time data and dynamically adjusted to display the area. When a traffic accident occurs, the grid data of the accident area will be updated in real-time, changing its color or status identifier, which will be immediately reflected on the map. The system can also automatically expand or reduce the range of the grid area based on actual needs to focus on displaying the traffic conditions of a hot area. In the event of a large-scale traffic incident, the grid module can zoom in on a specific area.

Images