High-precision perception-driven intelligent road network collaborative optimization method and system

Abstract

The invention provides a high-precision perception-driven intelligent road network collaborative optimization method and system, and relates to the technical field of intelligent traffic, and the method comprises the steps: collecting road sensing data through multi-sensor fusion, and obtaining multi-source perception data; performing dynamic traffic flow modeling according to the multi-source sensing data to obtain a traffic state prediction result; performing optimization decision on the traffic state prediction result to obtain an intelligent optimization decision; performing road network level cooperative control on the intelligent optimization decision to obtain a road network level cooperative control result; and performing real-time dynamic adjustment by using the road network level cooperative control result to obtain a road network scheduling result. According to the method and the device, the technical targets of global collaborative scheduling, improvement of traffic flow prediction precision, optimization of a real-time dynamic scheduling scheme and improvement of the intelligent level of urban traffic management can be realized, and the technical effects of reducing traffic congestion, improving road traffic efficiency, optimizing traffic resource allocation and improving emergency response capability are achieved.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technologies, and particularly to an intelligent road network collaborative optimization method and system driven by high-precision perception. Background Art

[0002] In modern urban traffic management, with the acceleration of the urbanization process and the continuous growth of the motor vehicle ownership, the problem of traffic congestion has become increasingly serious, resulting in a decline in traffic efficiency, an increase in energy consumption, and an aggravation of environmental pollution. To solve this problem, intelligent transportation systems have gradually developed, and through advanced perception, communication, and computing technologies, they can achieve efficient monitoring, prediction, and control of traffic flows. However, existing traffic management technologies still have many deficiencies and are difficult to meet the complex and changing urban traffic demands. Therefore, more refined and intelligent optimization methods are needed to improve the overall traffic operation efficiency.

[0003] Currently, urban traffic management mainly relies on means such as fixed traffic signal control, traffic guidance systems, and video surveillance. These traditional methods have obvious limitations. For example, fixed signal control usually adopts a preset cycle or an optimization scheme based on historical statistical data and cannot be dynamically adjusted according to real-time traffic flow changes, making it difficult to quickly respond to sudden changes in traffic flow. In addition, although traffic guidance systems can provide route suggestions, most of them are based on static map information or simple historical traffic flow analysis and cannot fully consider the impact of emergencies (such as accidents, construction) on traffic flow, resulting in overloading of some roads and underutilization of other roads. Although video surveillance technology can monitor traffic conditions in real time, its data processing ability is limited, and it is difficult to efficiently extract key information and quickly form optimization decisions.

[0004] In summary, in the prior art, there are technical problems that due to single traffic perception means, insufficient real-time data processing ability, limited prediction accuracy, and lack of global coordination in optimization scheduling, the traffic flow state is difficult to accurately predict and quickly respond, the optimization scheme is difficult to adapt to the dynamic and complex road conditions, further affecting the intelligent level of urban traffic management, resulting in a reduction in road traffic efficiency, an aggravation of traffic congestion, and uneven resource allocation. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent road network collaborative optimization method and system driven by high-precision perception, so as to solve the technical problems in the prior art that due to single traffic perception means, insufficient real-time data processing ability, limited prediction accuracy, and lack of global coordination in optimization scheduling, the traffic flow state is difficult to accurately predict and quickly respond, the optimization scheme is difficult to adapt to the dynamic and complex road conditions, further affecting the intelligent level of urban traffic management, resulting in a reduction in road traffic efficiency, an aggravation of traffic congestion, and uneven resource allocation.

[0006] In view of the above problems, the present application provides a method and system for collaborative optimization of an intelligent road network driven by high-precision perception.

[0007] In a first aspect, the present application provides a method for collaborative optimization of an intelligent road network driven by high-precision perception, which is implemented through a system for collaborative optimization of an intelligent road network driven by high-precision perception, and includes: collecting road sensing data through multi-sensor fusion to obtain multi-source perception data; performing dynamic traffic flow modeling based on the multi-source perception data to obtain a traffic state prediction result; making an optimized decision on the traffic state prediction result to obtain an intelligent optimized decision; performing road network-level collaborative control on the intelligent optimized decision to obtain a road network-level collaborative control result; and performing real-time dynamic adjustment using the road network-level collaborative control result to obtain a road network scheduling result.

[0008] In a second aspect, the present application further provides a system for collaborative optimization of an intelligent road network driven by high-precision perception, which is used to execute the method for collaborative optimization of an intelligent road network driven by high-precision perception as described in the first aspect, and includes: a multi-source perception data obtaining module, which is used to collect road sensing data through multi-sensor fusion to obtain multi-source perception data; a traffic state prediction result obtaining module, which is used to perform dynamic traffic flow modeling based on the multi-source perception data to obtain a traffic state prediction result; an intelligent optimized decision obtaining module, which is used to make an optimized decision on the traffic state prediction result to obtain an intelligent optimized decision; a road network-level collaborative control result obtaining module, which is used to perform road network-level collaborative control on the intelligent optimized decision to obtain a road network-level collaborative control result; and a road network scheduling result obtaining module, which is used to perform real-time dynamic adjustment using the road network-level collaborative control result to obtain a road network scheduling result.

[0009] The technical solution provided in the present application has at least the following technical effects or advantages: by collecting road sensing data through multi-sensor fusion to obtain multi-source perception data; performing dynamic traffic flow modeling based on the multi-source perception data to obtain a traffic state prediction result; making an optimized decision on the traffic state prediction result to obtain an intelligent optimized decision; performing road network-level collaborative control on the intelligent optimized decision to obtain a road network-level collaborative control result; and performing real-time dynamic adjustment using the road network-level collaborative control result to obtain a road network scheduling result, that is, by realizing global collaborative scheduling based on multi-sensor fusion, V2X technology, edge computing, and intelligent optimization algorithms, the technical goal of improving the accuracy of traffic flow prediction, optimizing the real-time dynamic scheduling scheme, and enhancing the intelligent level of urban traffic management is achieved, and the technical effects of reducing traffic congestion, improving road traffic efficiency, optimizing traffic resource allocation, and enhancing the emergency response ability are achieved.

[0010] The above description is only an overview of the technical solution of the present application. In order to better understand the technical means of the present application, it can be implemented according to the content of the specification. In order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the specific embodiments of the present application are given below. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description of the specification. Description of the Drawings

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the implementation examples or the description of the prior art. Obviously, the drawings described below are only exemplary. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0012] Figure 1 It is a schematic flow chart of the intelligent road network collaborative optimization method driven by high-precision perception of the present application;

[0013] Figure 2 It is a schematic structural diagram of the intelligent road network collaborative optimization system driven by high-precision perception of the present application.

[0014] Description of the reference numerals: The multi-source perception data acquisition module 11, the traffic state prediction result acquisition module 12, the intelligent optimization decision-making acquisition module 13, the road network-level collaborative control result acquisition module 14, and the road network scheduling result acquisition module 15. Detailed Embodiments

[0015] By providing the intelligent road network collaborative optimization method and system driven by high-precision perception, the present application solves the technical problems in the prior art, such as single traffic perception means, insufficient real-time data processing ability, limited prediction accuracy, and lack of global coordination in optimization scheduling, resulting in difficult accurate prediction and rapid response of traffic flow states, difficult adaptation of optimization schemes to dynamic and complex road conditions, further affecting the intelligent level of urban traffic management, leading to reduced road traffic efficiency, increased traffic congestion, and uneven resource allocation. The technical goal of realizing global collaborative scheduling based on multi-sensor fusion, V2X technology, edge computing, and intelligent optimization algorithms, improving the accuracy of traffic flow prediction, optimizing real-time dynamic scheduling schemes, and enhancing the intelligent level of urban traffic management is achieved, and the technical effect of reducing traffic congestion, improving road traffic efficiency, optimizing traffic resource allocation, and enhancing the emergency response ability is achieved.

[0016] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the accompanying drawings rather than all of them.

[0017] Example 1. Please refer to the attached Figure 1 drawings. The present application provides an intelligent road network collaborative optimization method driven by high-precision perception, which is applied to an intelligent road network collaborative optimization system driven by high-precision perception, and specifically includes the following steps:

[0018] S1: Collect road sensing data through multi-sensor fusion to obtain multi-source perception data.

[0019] Specifically, collect road sensing data through multi-sensor fusion to obtain multi-source perception data. "Through" indicates using a certain method or technology. "Multi-sensor" refers to using multiple different types of sensing devices. "Fusion" means integrating the data of multiple sensors to improve the accuracy and integrity of the data. "Collect" represents the process of obtaining external environmental information, and "road sensing data" refers to various information from the road environment, such as vehicle flow, road conditions, signal light status, obstacle positions, etc. "Multi-source perception data" represents a more accurate and reliable information set obtained after fusing data from different sensors. Therefore, through multi-sensor fusion, the coverage range, perception accuracy of road data, and the robustness of the system can be effectively improved.

[0020] For example, on urban roads, cameras can obtain traffic signal light status and lane line information, but may be affected by lighting; millimeter-wave radars can detect the speed and distance of vehicles, but may have misjudgments in complex scenarios; lidars can provide high-precision three-dimensional environmental information, but are costly. Through multi-sensor fusion, different sensors can complement each other. For example, when a camera detects an obstacle ahead but is unsure of its specific distance, millimeter-wave radars and lidars can provide more accurate position information, thus enhancing the overall perception ability of the system.

[0021] S2: Perform dynamic traffic flow modeling based on the multi-source perception data to obtain traffic state prediction results.

[0022] Specifically, dynamic traffic flow modeling is carried out based on multi-source perception data to obtain traffic state prediction results. The multi-source perception data comes from different types of sensors, including lidar, millimeter-wave radar, cameras, and vehicle-to-everything (V2X) systems. This data covers information such as vehicle position, speed, acceleration, road environment, and traffic signals. After fusing these data, the current traffic conditions can be more comprehensively reflected, providing a reliable data basis for modeling.

[0023] Dynamic traffic flow modeling describes the evolution of traffic flow in time and space through mathematical or computer models and can be updated in real time to adapt to the changing traffic environment. This modeling method can use microscopic modeling to simulate the behavior of individual vehicles, such as lane changing, car following, and overtaking, or macroscopic modeling that treats the entire traffic flow as a fluid and describes the changes in traffic flow density, flow velocity, and flow rate through fluid dynamics equations. In addition, mesoscopic modeling methods are in between the two, analyzing the overall behavior of vehicle platoons or groups. Combining these modeling methods can accurately characterize the dynamic characteristics of traffic flow.

[0024] The traffic state prediction results are the future traffic conditions obtained based on dynamic modeling, including information such as road congestion level, average vehicle speed, traffic flow density, and queue length. By using statistical analysis, machine learning, or fluid mechanics methods, the traffic state of a specific road section within a certain time period can be predicted and an optimization plan can be provided. For example, during the morning rush hour, the system analyzes that the current traffic flow is 1,000 vehicles per hour and the average vehicle speed is 40 kilometers per hour. It is predicted that within the next ten minutes, due to unreasonable signal control, the vehicle speed may drop to 20 kilometers per hour and the queue length may increase to 500 meters. Based on this prediction, the traffic management system can dynamically adjust the signal timing or issue navigation suggestions to guide vehicles to detour and reduce congestion.

[0025] The fusion of multi-source perception data improves the perception accuracy of traffic conditions, while dynamic traffic flow modeling provides a scientific way to analyze the evolution of traffic flow. By predicting the traffic state, decision-making support can be provided for signal optimization, route planning, and traffic scheduling, thereby improving road traffic efficiency, reducing congestion, and contributing to the development of intelligent transportation systems.

[0026] S3: Optimize and make decisions on the traffic state prediction results to obtain intelligent optimization decisions.

[0027] Specifically, optimize and make decisions on the traffic state prediction results to obtain intelligent optimization decisions. The traffic state prediction results are the future traffic conditions obtained based on multi-source perception data and dynamic traffic flow modeling, including key indicators such as road congestion level, traffic flow density, average vehicle speed, and queue length. These predicted data provide a basis for traffic optimization, enabling the system to identify potential traffic problems in advance and take corresponding optimization measures.

[0028] Optimized decision-making refers to formulating reasonable traffic management and scheduling strategies based on the results of traffic state prediction to improve the overall traffic efficiency and reduce traffic congestion. This process can adopt artificial intelligence optimization algorithms (such as genetic algorithms, reinforcement learning, etc.), combined with optimization means such as adaptive signal control, intelligent route planning, and dynamic lane adjustment, to optimize signal timing, vehicle flow guidance plans, road resource allocation, etc. For example, if the prediction result shows that a certain main road will be severely congested in the next fifteen minutes, the optimized decision-making can include extending the green light time, guiding some vehicles to detour, or dynamically adjusting the direction of variable lanes.

[0029] Intelligent optimized decision-making is based on optimized decision-making and combines intelligent connected vehicle technology, autonomous driving strategies, and real-time traffic scheduling systems to make the optimization plan more accurate and efficient. For example, intelligent connected vehicles can adjust their driving routes according to the optimized route planning, and the traffic signal control system can adjust the signal timing in real time to adapt to the changing vehicle flow conditions. In addition, autonomous vehicles can drive according to the intelligent optimized decision-making, following the optimal acceleration, lane-changing, and car-following strategies, thereby reducing traffic fluctuations and improving the overall traffic efficiency.

[0030] For example, during the morning rush hour in a certain city, the traffic state prediction system detects that the traffic flow on a certain expressway is about to reach its maximum carrying capacity and predicts that the vehicle speed will drop to 20 kilometers per hour in the next ten minutes. The optimized decision-making system adopts an intelligent signal control strategy to adjust the green light duration of multiple adjacent intersections and sends a detour suggestion to intelligent connected vehicles to guide some vehicles into parallel roads. At the same time, autonomous vehicles adjust their driving speeds according to the optimized strategy to ensure a smoother traffic flow. After intelligent optimized decision-making, the average vehicle speed on this section finally stabilizes at 35 kilometers per hour, and the traffic efficiency is increased by 20%.

[0031] The results of traffic state prediction provide key information on future traffic trends, and optimized decision-making can formulate reasonable traffic management plans based on this data. Intelligent optimized decision-making further combines technologies such as intelligent connection and autonomous driving to make the optimization plan more intelligent and precise. Ultimately, this method can effectively reduce traffic congestion, improve road utilization, and achieve more efficient traffic management and scheduling.

[0032] S4: Perform road network-level collaborative control on the intelligent optimized decision-making to obtain the road network-level collaborative control result.

[0033] Specifically, the intelligent optimization decision is used for network-level collaborative control to obtain the network-level collaborative control result. The intelligent optimization decision is based on the traffic state prediction result and improves the operation efficiency of local traffic by optimizing signal control, path planning, traffic dispatching, etc. However, single-point or local optimization may lead to imbalance in the overall traffic network. For example, the optimization of a certain intersection may increase the pressure on adjacent road sections. Therefore, it is necessary to further conduct collaborative control at the level of the entire road network to make the optimization strategy play a role in a larger range and ensure the coordinated and efficient operation of the overall traffic flow.

[0034] Network-level collaborative control refers to the collaborative scheduling of multiple road nodes, signal control systems, intelligent connected vehicles, and traffic management centers within the entire traffic network to achieve global optimization. This method involves various means such as global timing optimization of signal lights, traffic flow guidance, variable lane management, and dynamic speed limit adjustment. For example, on the main road of a city, if the traffic flow at a certain key intersection speeds up after optimization, but there is still a bottleneck at the downstream intersection, the network-level collaborative control will further adjust the signal light duration of the downstream intersection to match the change in the overall traffic flow and avoid the generation of new congestion points.

[0035] The network-level collaborative control result is a traffic dispatching plan after global optimization, ensuring that the operation efficiency of the entire traffic network reaches the optimal. For example, during the peak period, the intelligent optimization decision may recommend adjusting the signal light timing of a certain main road to improve the traffic capacity. However, if the burden on adjacent roads increases, the network-level collaborative control system will synchronously adjust the signal lights of the surrounding roads and guide some traffic flows into secondary roads or detour routes to make the traffic flow distribution in the whole city more balanced.

[0036] For example, in the traffic management of a large city, the intelligent optimization decision predicts that a certain expressway will be severely congested in the next ten minutes and formulates an optimization plan, such as adjusting signal light timing and optimizing path planning. However, if only this expressway is optimized, it may lead to an increase in the load on surrounding roads. Therefore, the network-level collaborative control system will further optimize the signal light strategy of the entire area, guide some vehicles into secondary roads, and adjust the signal light cycle of adjacent highway exits to relieve the pressure on the main road. After optimization, the average traffic speed in this area has increased by 20%, the overall traffic flow is more balanced, and the possibility of secondary congestion is reduced.

[0037] The intelligent optimization decision focuses on local optimization, while the network-level collaborative control starts from a global perspective and makes the traffic flow in the whole city or region more efficient through multi-level and multi-node coordination. Ultimately, the network-level collaborative control result can improve the overall traffic capacity of the road network, reduce the negative impact brought by local optimization, and achieve more intelligent traffic management and dispatching.

[0038] S5: Use the road network-level collaborative control result for real-time dynamic adjustment to obtain the road network scheduling result.

[0039] Specifically, use the road network-level collaborative control result for real-time dynamic adjustment to obtain the road network scheduling result. The road network-level collaborative control result is a traffic scheduling plan formed based on a global optimization strategy, including measures such as optimized signal control, route planning, and traffic flow guidance. However, traffic flow is dynamically changing, and the real-time traffic flow situation may deviate from the prediction. Therefore, continuous dynamic adjustment is required to adapt to the actual traffic conditions and ensure that the optimization plan remains effective under different time periods and different traffic loads.

[0040] Real-time dynamic adjustment refers to the immediate optimization of the road network-level collaborative control plan based on the latest traffic flow data so that it can adapt to the changing traffic environment. This process relies on various real-time data collection means such as traffic sensors, intelligent connected vehicles, video surveillance, and drone patrols, combined with artificial intelligence algorithms for rapid calculation to dynamically adjust signal timings, variable lane directions, speed limit strategies, etc. For example, if the traffic volume at a certain intersection suddenly increases, the system will automatically shorten the green light time of the lateral road to improve the traffic capacity of the main road. At the same time, if there is an unexpected traffic volume on a certain detour road, the system will adjust the guidance strategy accordingly to avoid new congestion.

[0041] The road network scheduling result is an optimized scheduling plan that finally applies to the current traffic state after real-time adjustment. It not only combines historical data and prediction results but also can perform adaptive optimization based on real-time traffic flow data. For example, during the evening rush hour in a large city, a certain expressway was originally operating according to the collaborative control plan. However, due to a sudden accident, traffic congestion occurred. The system real-time monitored the decrease in vehicle speed in this area and automatically adjusted the signal duration of adjacent roads. At the same time, it pushed a new detour route to intelligent connected vehicles. After optimization and adjustment, the traffic flow in this area returned to the normal level, and the average vehicle waiting time decreased by 30%.

[0042] Real-time dynamic adjustment enables the road network scheduling plan to adapt to sudden situations at any time and ensures that the optimization measures are always effective. Through this mechanism, the traffic management system can more accurately match the actual traffic demand, reduce congestion, improve road traffic efficiency, and achieve more intelligent and flexible traffic scheduling.

[0043] The intelligent road network collaborative optimization method driven by high-precision perception is applied to the intelligent road network collaborative optimization system driven by high-precision perception, which can achieve the technical goals of global collaborative scheduling based on multi-sensor fusion, V2X technology, edge computing, and intelligent optimization algorithms, improve the accuracy of traffic flow prediction, optimize the real-time dynamic scheduling scheme, and enhance the intelligent level of urban traffic management, and achieve the technical effects of reducing traffic congestion, improving road traffic efficiency, optimizing traffic resource allocation, and enhancing the emergency response ability.

[0044] Furthermore, this application also includes: collecting road environment, traffic flow, and vehicle status based on multi-sensors, where the multi-sensors include lidar, millimeter-wave radar, and cameras; using Kalman filtering to fuse the road environment, traffic flow, and vehicle status to obtain the multi-source perception data.

[0045] Specifically, the road environment, traffic flow, and vehicle status are collected based on multi-sensors, where the multi-sensors include lidar, millimeter-wave radar, and cameras. "Based on" indicates using a certain technology or method as the foundation. Multi-sensors mean using multiple different types of sensing devices to obtain information. Collection refers to the process of obtaining data from the external environment. The road environment includes information such as the width, slope, road surface condition, and obstacles of the road. Traffic flow refers to the set of vehicles traveling on the road within a certain period of time, including parameters such as traffic volume, vehicle speed, and headway. Vehicle status describes the driving information of a single vehicle, such as position, speed, and acceleration. Lidar measures the distance and shape of objects by emitting laser light and is suitable for high-precision 3D modeling. Millimeter-wave radar detects the speed and distance of objects using electromagnetic waves with millimeter-level wavelengths and is particularly suitable for harsh weather conditions. Cameras obtain information such as road signs, lane lines, and traffic signals by capturing visible light images. Therefore, by combining lidar, millimeter-wave radar, and cameras, the road and vehicle conditions can be comprehensively perceived in different environments.

[0046] Using Kalman filtering to fuse the road environment, traffic flow, and vehicle status to obtain the multi-source perception data. "Using" means using a certain method or technology. Kalman filtering is an algorithm widely used in signal processing and state estimation. Its core idea is to predict the true state of the system based on the current observation value and historical estimation value, and dynamically correct the error using a mathematical model to make the data more accurate. Fusion refers to integrating the information of multiple data sources to improve the reliability and accuracy of the overall data. Here, Kalman filtering can comprehensively process the data of lidar, millimeter-wave radar, and cameras, eliminate their respective measurement errors, and make the obtained road environment, traffic flow, and vehicle status information more accurate and stable. Therefore, by fusing data with Kalman filtering, the measurement bias that may be brought by a single sensor can be eliminated, and the accuracy of multi-source perception data can be improved.

[0047] Furthermore, this application also includes: collecting traffic flow information based on the vehicle networking to construct a dynamic traffic flow model; combining traffic simulation to analyze the vehicle behavior patterns of the dynamic traffic flow model to obtain the traffic state prediction result.

[0048] Specifically, traffic flow information is collected based on the vehicle networking to construct a dynamic traffic flow model. "Based on" indicates relying on a certain technology or method. The vehicle networking is the V2X (Vehicle-to-Everything) technology, which refers to the wireless communication network between vehicles and vehicles (V2V), vehicles and infrastructure (V2I), vehicles and pedestrians (V2P), and vehicles and the cloud (V2N). It can obtain information such as the speed, position, acceleration, and driving direction of vehicles in real time and interact with infrastructure such as traffic lights and road monitoring devices. Traffic flow information collection refers to collecting the operating status of vehicles and road traffic data through vehicle networking technology, such as traffic flow, vehicle speed, and lane occupancy rate. "Construct" means to establish or build a certain system or model. The dynamic traffic flow model refers to a traffic flow simulation model that can be updated in real time as time changes. It can adjust model parameters according to factors such as the real-time behavior of vehicles, road conditions, and traffic signal changes to make it closer to the actual traffic conditions. Therefore, using vehicle networking to collect traffic flow information can more efficiently and accurately construct a traffic flow model that changes in real time, providing data support for subsequent traffic optimization.

[0049] Combining traffic simulation to analyze the vehicle behavior patterns of the dynamic traffic flow model to obtain the traffic state prediction result. "Combining" means using different methods or technologies in coordination. Traffic simulation refers to using a computer to simulate the operation of a real traffic system. Traffic simulation software such as SUMO and VISSIM can simulate the vehicle flow under different road conditions based on the collected data. "Analyze" means to process, compare, and study data to draw meaningful conclusions. The dynamic traffic flow model has been defined previously, which is a mathematical or computer model that describes the changes in traffic flow. Vehicle behavior patterns refer to the driving rules of vehicles in different traffic environments, such as accelerating, decelerating, changing lanes, following, and overtaking. These behaviors can be classified and modeled through historical data and machine learning methods. "Obtain" means to get a certain result. The traffic state prediction result refers to the estimation of the traffic conditions within a certain future time period, such as predicting the congestion degree, average vehicle speed, and queue length of a certain road within the next ten minutes. Therefore, through traffic simulation, the dynamic traffic flow model can be analyzed in depth, the vehicle behavior patterns can be extracted, and the future traffic state can be predicted based on this, providing decision-making support for intelligent traffic management.

[0050] For example, on the main roads of a city, the vehicle networking system collects the speed, acceleration, and position information of each vehicle in real time and transmits it to the cloud for processing. Based on this data, a dynamic traffic flow model is constructed. If the current traffic volume on a certain road is 300 vehicles per hour, and historical data indicates that traffic congestion occurs when the traffic volume reaches 400 vehicles per hour, then through simulation analysis, it can be predicted whether traffic congestion is likely to occur on this road within the next 20 minutes. If the prediction result shows that the average vehicle speed on this road will drop to 20 kilometers per hour, the traffic management system can adjust the signal light duration in advance or guide some vehicles to choose other routes to relieve the possible congestion.

[0051] Furthermore, this application also includes: obtaining vehicle status, road conditions, and traffic signals in real time through the vehicle networking, and collecting traffic data using in-vehicle terminals; performing dynamic traffic flow modeling on the vehicle status, the road conditions, the traffic signals, and the traffic data based on fluid dynamics to obtain the dynamic traffic flow model.

[0052] Specifically, vehicle status, road conditions, and traffic signals are obtained in real time through the vehicle networking, and traffic data is collected using in-vehicle terminals. "Through" indicates the use of a certain technology or means. The vehicle networking, namely V2X (Vehicle-to-Everything), is a wireless communication-based technology that enables vehicles to exchange data with other vehicles (V2V), infrastructure (V2I), pedestrians (V2P), and the cloud (V2N). "Real time" means that data acquisition and transmission are completed in an extremely short time, ensuring the timeliness and accuracy of information. "Obtain" means to collect information from the external environment, and vehicle status refers to the current operating parameters of the vehicle, such as speed, acceleration, driving direction, fuel consumption, etc. Road conditions describe the physical attributes and traffic environment of the current road, such as the wetness of the road surface, slope, lane width, construction area, etc. Traffic signals refer to the status and change cycle of traffic lights, as well as the electronic indication signs on the road. "Adopt" means to use a certain method, and in-vehicle terminal refers to the intelligent device installed on the vehicle, such as OBU (On-Board Unit), which can collect, process, and transmit traffic data. "Collect" means the process of obtaining and recording data, and traffic data refers to all information related to road traffic, including traffic volume, vehicle speed, lane occupancy rate, etc. Therefore, the vehicle networking can obtain key traffic information in real time and collect traffic data through in-vehicle terminals, providing basic data support for subsequent traffic flow modeling.

[0053] Based on hydrodynamics, dynamic traffic flow modeling is carried out on the vehicle state, the road condition, the traffic signal and the traffic data to obtain the dynamic traffic flow model. "Based on" indicates that it is founded on a certain theory or method. Hydrodynamics is a scientific theory that studies the motion of fluids (such as air or water). In traffic flow modeling, vehicles are regarded as fluid particles, and the traffic flow on the road is like the flow of fluid in a pipeline, affected by factors such as density, pressure and speed. "On" indicates the object of action, including the vehicle state, the road condition, the traffic signal and the traffic data, and these factors jointly determine the change of traffic flow. "Carry out" indicates the execution of a certain operation. Dynamic traffic flow modeling refers to simulating and analyzing the characteristics of traffic flow changing over time through a mathematical model. "Obtain" indicates getting a certain result. The dynamic traffic flow model is a mathematical description of the actual traffic flow, which can reflect the vehicle flow characteristics under different road conditions and traffic control conditions. Therefore, using hydrodynamics for traffic flow modeling can more accurately simulate the evolution of traffic flow and provide a scientific basis for traffic optimization.

[0054] Furthermore, this application also includes: optimizing the traffic state prediction result by using a genetic algorithm to obtain an optimized control decision, where the optimized control decision includes optimized signal control, optimized path planning and optimized traffic scheduling; combining driving strategies and carrying out collaborative optimization according to the optimized control decision to obtain the intelligent optimization decision, where the driving strategies include an autonomous driving strategy and a manual driving strategy.

[0055] Specifically, the traffic state prediction result is optimized by using a genetic algorithm to obtain an optimized control decision. The genetic algorithm is an optimization algorithm based on the principles of natural selection and biological evolution, which can find the optimal solution in complex and high-dimensional optimization problems. This algorithm continuously optimizes the candidate solutions through operations such as selection, crossover and mutation, so that the final optimization result can better meet the target requirements. The traffic state prediction result provides the change trend of future traffic conditions, such as information on road congestion degree, traffic flow density, average vehicle speed, etc. In the optimization process of the genetic algorithm, these prediction data will be used as input variables to optimize the traffic management strategy and make the traffic flow smoother.

[0056] The optimized control decisions include optimized signal control, optimized path planning, and optimized traffic scheduling. Optimized signal control refers to maximizing the traffic efficiency at different intersections by adjusting the signal timing plan of traffic lights and reducing the waiting time of vehicles. For example, during peak hours, genetic algorithms can optimize the signal cycle to extend the green light time on the main road, thereby improving the traffic capacity. Optimized path planning means providing the optimal driving route for vehicles based on real-time traffic data to reduce congestion and travel time. For instance, if the system predicts that a certain road will become congested within the next ten minutes, the optimization algorithm can recommend a detour plan for the vehicle to avoid entering the congested section. Optimized traffic scheduling involves the intelligent allocation of public transportation, freight logistics, etc. For example, dynamically adjusting the departure interval of buses or guiding freight vehicles to travel during off-peak hours to reduce the traffic load.

[0057] Combined with driving strategies, collaborative optimization is performed according to the optimized control decisions to obtain the intelligent optimization decision. Collaborative optimization means that, based on the optimized control decisions, further combined with the characteristics of different driving modes, the traffic management strategy is adjusted to adapt to the actual driving behavior. Driving strategies include autonomous driving strategies and manual driving strategies. The autonomous driving strategy is for connected and autonomous vehicles and can dynamically adjust the driving speed, following distance, and lane-changing behavior based on real-time traffic data. For example, based on the optimized signal plan, autonomous vehicles can adjust their vehicle speeds according to the predicted green light time to reduce stop-and-wait. The manual driving strategy is for human-driven vehicles, and the optimization plan needs to consider the driver's behavior patterns. For example, through information prompts, dynamic speed limits, variable lane guidance, etc., the driver is guided to drive along the optimized route to improve the overall traffic efficiency.

[0058] The intelligent optimization decision is the final result of the combination of genetic algorithm optimization, traffic control decisions, and driving strategies. For example, during the peak hours in a city, the system predicts that a certain main road will be severely congested within the next fifteen minutes. After optimizing the signal control, the green light time is adjusted to 120% of the original, and at the same time, the path planning is optimized to guide some vehicles to detour through side roads. In addition, after receiving the optimized signal timing data, autonomous vehicles adjust their vehicle speeds in advance to reduce the number of stops, while human-driven vehicles receive the recommended route and lane-changing suggestions through the navigation system. This collaborative optimization method can effectively relieve congestion and improve the road traffic capacity.

[0059] Genetic algorithms can find the optimal solution in complex traffic optimization problems, and the optimized control decisions cover signal control, path planning, and traffic scheduling, improving the traffic management ability from different levels. Combining autonomous driving and manual driving strategies for collaborative optimization can further enhance the applicability of the optimization plan, making the intelligent optimization decision more accurate and efficient. Ultimately, this method can significantly improve the road traffic efficiency, reduce traffic congestion, and provide an efficient optimization plan for the intelligent transportation system.

[0060] Further, this application also includes: performing collaborative scheduling on the intelligent optimization decision through an intelligent connected vehicle and infrastructure to obtain a global traffic scheduling optimization result; and optimizing the global traffic scheduling optimization result by using adaptive signal control to obtain the road network-level collaborative control result.

[0061] Specifically, performing collaborative scheduling on the intelligent optimization decision through an intelligent connected vehicle and infrastructure to obtain a global traffic scheduling optimization result. An intelligent connected vehicle (ICV) is a vehicle equipped with intelligent perception, computing, and communication capabilities, capable of data interaction with other vehicles (V2V), road infrastructure (V2I), pedestrians (V2P), and the cloud (V2N). Infrastructure includes traffic lights, road cameras, radar sensors, electronic information screens, etc., which can provide key information such as road conditions, traffic flow, and signal control strategies. Collaborative scheduling means that, based on the intelligent optimization decision, the intelligent connected vehicle and infrastructure work together to make the scheduling of the entire traffic system more reasonable. For example, when the system calculates that a certain road section is about to become congested, the infrastructure can send a detour suggestion to the intelligent connected vehicle and at the same time adjust the signal duration to guide the traffic flow to a smoother road, thereby achieving global optimization.

[0062] The global traffic scheduling optimization result is an optimized traffic flow organization plan for the entire traffic network under the action of collaborative scheduling. It not only focuses on the optimization of a single intersection or a single road, but also considers the overall traffic conditions of the entire city or region. For example, if the traffic flow on a certain main road is too large, the intelligent scheduling system can coordinate the traffic lights at multiple intersections to make the traffic flow distribution more balanced, avoiding excessive congestion in a certain area. At the same time, the intelligent connected vehicle can drive along the optimal path, reducing intersection conflicts and improving road utilization.

[0063] Optimizing the global traffic scheduling optimization result by using adaptive signal control to obtain the road network-level collaborative control result. Adaptive signal control is a control method that dynamically adjusts the traffic light duration based on real-time traffic data. It can optimize the signal timing according to the actual traffic flow situation, rather than controlling according to a fixed cycle. For example, if it detects an increase in the traffic flow in a certain direction, the adaptive signal control system can automatically extend the green light time to improve the traffic efficiency and reduce the queue length.

[0064] The road network-level collaborative control result is a global optimization solution for the traffic signals of the entire city. Through the collaborative effect among the signal lights, it enables the traffic efficiency of the entire road network to reach the optimal level. For example, during the peak hours, if a certain main road needs to improve its traffic capacity, the adaptive signal control can adjust the green light time of multiple intersections along the line, enabling the vehicle flow to seamlessly pass through multiple intersections, reducing the vehicle waiting time. At the same time, the signal lights on the branch roads will also be adjusted accordingly to ensure that the optimization of the main road does not cause severe congestion in other areas.

[0065] For example, during the morning peak period in a large city, intelligent connected vehicles upload their driving status in real time through the vehicle networking system. The infrastructure monitors that the traffic volume on a certain main road exceeds 3,000 vehicles per hour and predicts that congestion may occur within the next ten minutes. The collaborative scheduling system immediately adjusts the signal control, extends the green light duration of the main road, and at the same time sends a detour suggestion to the intelligent connected vehicles to guide some vehicles into the parallel roads. Subsequently, the adaptive signal control system further optimizes the signal timing of adjacent intersections to ensure the improvement of the traffic efficiency of the entire road network. Finally, this road network-level collaborative control solution reduces the average vehicle waiting time and improves the traffic capacity.

[0066] The collaborative scheduling between intelligent connected vehicles and the infrastructure extends traffic optimization from single-point control to global optimization, while the adaptive signal control can further dynamically adjust the signal lights to achieve more refined management. Through the combination of the two, the road network-level collaborative control solution can significantly improve the road traffic efficiency, reduce congestion, and make the traffic system of the entire city more intelligent and operate more efficiently.

[0067] Furthermore, this application also includes: constructing a virtual simulation environment for the road network-level collaborative control result in combination with digital twin, and conducting real-time simulation and effectiveness evaluation to obtain the effectiveness evaluation result; continuously optimizing the effectiveness evaluation result through closed-loop control to obtain the road network scheduling result.

[0068] Specifically, construct a virtual simulation environment for the road network-level collaborative control result in combination with digital twin, and conduct real-time simulation and effectiveness evaluation to obtain the effectiveness evaluation result. Digital twin is a technology that constructs a virtual model based on real-time data. It can truly reproduce the operating state of the physical world in the digital space and conduct dynamic simulation and optimization. In the traffic system, digital twin technology can construct a high-precision virtual road network, digitize elements such as roads, vehicles, and traffic signals, and keep them synchronized with the actual data, thereby realizing the integrated management and optimization of the virtual and the real.

[0069] During the construction of the virtual simulation environment, the road network-level collaborative control results are mapped into the digital twin system, including the optimized signal timing plan, the scheduling strategy of intelligent connected vehicles, the adjustment of path planning, etc. These data are input into the simulation environment, enabling the virtual traffic system to operate according to the optimized plan, thereby testing and verifying the effectiveness of the optimization strategy. Real-time simulation refers to the dynamic demonstration of the evolution process of traffic flow in the simulation system, such as how vehicles respond to the new signal control plan, how the traffic flow density changes, whether congestion is alleviated, etc. The effectiveness evaluation is to conduct a quantitative analysis of the simulation results, such as calculating key indicators such as average vehicle speed, travel time, vehicle queue length, etc., to measure the actual effect of the optimization plan.

[0070] The effectiveness evaluation results are continuously optimized through closed-loop control to obtain the road network scheduling results. Closed-loop control is a feedback regulation mechanism, that is, the system continuously collects evaluation results and conducts continuous optimization according to the actual operation effect to ensure that the final traffic scheduling plan reaches the optimal state. For example, if the simulation evaluation results show that the optimization plan for a certain intersection improves the traffic efficiency of the main road but causes greater congestion on the branch road, the closed-loop control system will automatically adjust the signal timing plan to make the overall traffic flow more balanced.

[0071] The road network scheduling results are the final traffic scheduling plan optimized by closed-loop control. After continuous adjustment and optimization, it can adapt to different traffic environments and maximize traffic efficiency. For example, in the traffic management of a large city, the digital twin system constructs the virtual traffic network of the entire city and conducts simulation tests based on historical data and real-time monitoring data. The preliminary optimization plan shows in the simulation that after the signal is adjusted, the vehicle speed on the main road increases by 20%, but the waiting time on some branch roads increases by 30%. The closed-loop control system automatically adjusts the signal timing according to the evaluation results. The optimized plan increases the overall average vehicle speed by 15%, and at the same time controls the waiting time on the branch roads within an acceptable range, finally forming the optimal road network scheduling plan.

[0072] Digital twin technology enables traffic optimization plans to be tested with high precision in a virtual environment, avoiding the risks and costs brought by direct experiments on actual roads. Closed-loop control ensures that the optimization plan can be continuously adjusted according to real-time evaluation results, making the final road network scheduling results more intelligent and efficient. Through this method, the traffic management system can dynamically adapt to complex and changeable traffic conditions, improve road traffic efficiency, reduce congestion, and achieve more efficient urban traffic scheduling.

[0073] In summary, the high-precision perception-driven intelligent road network collaborative optimization method provided by this application has the following technical effects: By realizing global collaborative scheduling based on multi-sensor fusion, V2X technology, edge computing, and intelligent optimization algorithms, the technical goal of improving traffic flow prediction accuracy, optimizing real-time dynamic scheduling schemes, and enhancing the intelligent level of urban traffic management is achieved, and the technical effects of reducing traffic congestion, improving road traffic efficiency, optimizing traffic resource allocation, and enhancing the emergency response ability are achieved.

[0074] Embodiment 2. Based on the same inventive concept as the high-precision perception-driven intelligent road network collaborative optimization method in the foregoing embodiment, this application also provides a high-precision perception-driven intelligent road network collaborative optimization system. Please refer to the appendix Figure 2 , including: a multi-source perception data obtaining module 11, which is used to collect road sensing data by multi-sensor fusion to obtain multi-source perception data; a traffic state prediction result obtaining module 12, which is used to perform dynamic traffic flow modeling based on the multi-source perception data to obtain a traffic state prediction result; an intelligent optimization decision-making obtaining module 13, which is used to perform optimization decision-making on the traffic state prediction result to obtain an intelligent optimization decision; a road network-level collaborative control result obtaining module 14, which is used to perform road network-level collaborative control on the intelligent optimization decision to obtain a road network-level collaborative control result; a road network scheduling result obtaining module 15, which is used to perform real-time dynamic adjustment by using the road network-level collaborative control result to obtain a road network scheduling result.

[0075] Furthermore, the high-precision perception-driven intelligent road network collaborative optimization system is also used for: collecting road environment, traffic flow, and vehicle status based on multi-sensors, where the multi-sensors include lidar, millimeter-wave radar, and cameras; using Kalman filtering to fuse the road environment, traffic flow, and vehicle status to obtain the multi-source perception data.

[0076] Furthermore, the high-precision perception-driven intelligent road network collaborative optimization system is also used for: collecting traffic flow information based on the vehicle network to construct a dynamic traffic flow model; analyzing the vehicle behavior patterns of the dynamic traffic flow model in combination with traffic simulation to obtain the traffic state prediction result.

[0077] Furthermore, the high-precision perception-driven intelligent road network collaborative optimization system is also used for: obtaining vehicle status, road conditions, and traffic signals in real time through the vehicle network, and collecting traffic data by using an on-vehicle terminal; performing dynamic traffic flow modeling on the vehicle status, the road conditions, the traffic signals, and the traffic data based on fluid dynamics to obtain the dynamic traffic flow model.

[0078] Furthermore, the intelligent road network collaborative optimization system driven by high-precision perception is also used for: optimizing the traffic state prediction result by using a genetic algorithm to obtain an optimized control decision, where the optimized control decision includes optimized signal control, optimized path planning, and optimized traffic scheduling; combining driving strategies, and performing collaborative optimization according to the optimized control decision to obtain the intelligent optimization decision, where the driving strategies include an autonomous driving strategy and a manual driving strategy.

[0079] Furthermore, the intelligent road network collaborative optimization system driven by high-precision perception is also used for: performing collaborative scheduling on the intelligent optimization decision through intelligent connected vehicles and infrastructure to obtain a global traffic scheduling optimization result; optimizing the global traffic scheduling optimization result by using adaptive signal control to obtain the road network-level collaborative control result.

[0080] Furthermore, the intelligent road network collaborative optimization system driven by high-precision perception is also used for: combining digital twins to construct a virtual simulation environment for the road network-level collaborative control result, and performing real-time simulation and effectiveness evaluation to obtain an effectiveness evaluation result; continuously optimizing the effectiveness evaluation result through closed-loop control to obtain the road network scheduling result.

[0081] The various embodiments in this specification are described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The high-precision perception-driven intelligent road network collaborative optimization method and specific examples in the foregoing Embodiment 1 are equally applicable to the high-precision perception-driven intelligent road network collaborative optimization system in this embodiment. Through the foregoing detailed description of the high-precision perception-driven intelligent road network collaborative optimization method, those skilled in the art can clearly understand the high-precision perception-driven intelligent road network collaborative optimization system in this embodiment. Therefore, for the sake of brevity of the specification, it will not be elaborated herein.

[0082] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0083] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. A high-precision perception-driven intelligent road network collaborative optimization method, characterized in that: include: Use multi-sensor fusion to collect road sensor data and obtain multi-source perception data; Perform dynamic traffic flow modeling based on the multi-source perception data to obtain traffic state prediction results; Performing optimization decision on the traffic state prediction result to obtain intelligent optimization decision; Performing road network-level collaborative control on the intelligent optimization decision to obtain a road network-level collaborative control result; The road network-level collaborative control result is used to perform real-time dynamic adjustments to obtain a road network scheduling result.

2. The high-precision perception-driven intelligent road network collaborative optimization method according to claim 1, characterized in that: Multi-sensor fusion is used to collect road sensor data to obtain multi-source perception data, including: Obtaining road environment, traffic flow and vehicle status based on multi-sensor acquisition, wherein the multi-sensor includes a laser radar, a millimeter-wave radar and a camera; Kalman filtering is used to fuse the road environment, traffic flow and vehicle status to obtain the multi-source perception data.

3. The high-precision perception-driven intelligent road network collaborative optimization method according to claim 1, characterized in that: Dynamic traffic flow modeling is performed based on the multi-source perception data to obtain traffic state prediction results, including: Collect traffic flow information based on the Internet of Vehicles and build a dynamic traffic flow model; The vehicle behavior pattern of the dynamic traffic flow model is analyzed in combination with traffic simulation to obtain the traffic state prediction result.

4. The high-precision perception-driven intelligent road network collaborative optimization method according to claim 3, characterized in that: Traffic flow information is collected based on the Internet of Vehicles, and a dynamic traffic flow model is constructed, including: Obtain vehicle status, road conditions and traffic signals in real time through the Internet of Vehicles, and collect traffic data using vehicle-mounted terminals; Dynamic traffic flow modeling is performed on the vehicle state, the road condition, the traffic signal and the traffic data based on fluid dynamics to obtain the dynamic traffic flow model.

5. The high-precision perception-driven intelligent road network collaborative optimization method according to claim 1, characterized in that: Optimizing the traffic state prediction result to obtain an intelligent optimization decision, including: Using a genetic algorithm to optimize the traffic state prediction result to obtain an optimized control decision, wherein the optimized control decision includes optimized signal control, optimized path planning and optimized traffic scheduling; In combination with the driving strategy, collaborative optimization is performed according to the optimization control decision to obtain the intelligent optimization decision, wherein the driving strategy includes an automatic driving strategy and a manual driving strategy.

6. The high-precision perception-driven intelligent road network collaborative optimization method according to claim 1, characterized in that: The intelligent optimization decision is subjected to a road network level collaborative control to obtain a road network level collaborative control result, including: The intelligent optimization decision is coordinated and dispatched by intelligent networked vehicles and infrastructure to obtain a global traffic dispatch optimization result; The global traffic scheduling optimization result is optimized using adaptive signal control to obtain the road network-level collaborative control result.

7. The high-precision perception-driven intelligent road network collaborative optimization method according to claim 1, characterized in that: The road network level collaborative control result is used to perform real-time dynamic adjustment to obtain a road network dispatch result, including: Combined with the digital twin, a virtual simulation environment is constructed for the road network-level collaborative control results, and real-time simulation and effectiveness evaluation are performed to obtain effectiveness evaluation results; The effectiveness evaluation result is continuously optimized through closed-loop control to obtain the road network scheduling result.

8. High-precision perception-driven intelligent road network collaborative optimization system, characterized by: The steps for implementing the high-precision perception-driven intelligent road network collaborative optimization method according to any one of claims 1 to 7 include: A multi-source sensing data acquisition module, wherein the multi-source sensing data acquisition module is used to collect road sensing data by adopting multi-sensor fusion to obtain multi-source sensing data; A traffic state prediction result acquisition module, wherein the traffic state prediction result acquisition module is used to perform dynamic traffic flow modeling according to the multi-source perception data to obtain a traffic state prediction result; An intelligent optimization decision obtaining module, wherein the intelligent optimization decision obtaining module is used to optimize the traffic state prediction result to obtain an intelligent optimization decision; A road network level collaborative control result obtaining module, wherein the road network level collaborative control result obtaining module is used to perform road network level collaborative control on the intelligent optimization decision to obtain a road network level collaborative control result; A road network scheduling result obtaining module is used to use the road network-level collaborative control results to perform real-time dynamic adjustments to obtain road network scheduling results.

Images