Multi-model collaborative optimization control method and system for urban traffic and medium
By using real-time monitoring and multi-model collaborative decision-making, a road network traffic characteristic vector is constructed and a collaborative optimization scheme is generated. This solves the problem that a single model in the existing technology is unable to cope with the complex changes in urban traffic, and improves the operational efficiency and adaptability of the urban traffic system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-03-27
AI Technical Summary
Existing urban traffic control methods mostly rely on a single model or a single control strategy, making it difficult to simultaneously take into account the traffic operation needs of different regions and time scales. They also lack the ability to adapt to complex traffic scenarios such as sudden congestion and rapid fluctuations in traffic flow.
By monitoring the urban traffic network in real time, a road network traffic characteristic vector is constructed and input into multiple collaborative optimization models (such as the green wave coordination model and the congestion relief model) to generate a traffic optimization collaborative scheme. Combined with safety hazard detection and hazard suppression schemes, multi-model collaborative decision-making is achieved.
It improves the overall efficiency of urban transportation, better copes with the complex and dynamic changes in the urban road network, and enhances the flexibility and adaptability of the transportation system.
Smart Images
Figure CN121747313A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent transportation technology, specifically to a multi-model collaborative optimization control method, system, and medium for urban transportation. Background Technology
[0002] As urban transportation expands and road networks become increasingly complex, traffic flow exhibits highly dynamic and random characteristics. Existing urban traffic control methods often rely on single models or single control strategies, such as traffic signal timing optimization, green wave coordination, or localized congestion relief. These single traffic optimization methods typically struggle to simultaneously address traffic demands across different regions and time scales, and are insufficiently adaptable to complex traffic scenarios such as sudden congestion and rapid fluctuations in traffic flow. Summary of the Invention
[0003] This application provides a multi-model collaborative optimization control method, system, and medium for urban traffic, which is used to address the technical problem that existing single traffic optimization methods cannot effectively cope with the complex dynamic changes of urban road networks.
[0004] In view of the above problems, this application provides a multi-model collaborative optimization control method, system and medium for urban traffic.
[0005] The first aspect of this application provides a multi-model collaborative optimization control method for urban traffic, the method comprising:
[0006] The system monitors the urban traffic network in real time to obtain road network traffic data streams; constructs road network traffic characteristic vectors based on the road network traffic data streams; inputs the road network traffic characteristic vectors into multiple traffic optimization collaborative models to obtain traffic optimization collaborative schemes; and encrypts and distributes the traffic optimization collaborative schemes to the urban traffic network.
[0007] A second aspect of this application provides a multi-model collaborative optimization control system for urban traffic, the system comprising:
[0008] The monitoring module is used to monitor the urban traffic network in real time and obtain road network traffic data streams; the vector construction module is used to construct road network traffic characteristic vectors based on the road network traffic data streams; the scheme acquisition module is used to input the road network traffic characteristic vectors into multiple traffic optimization collaborative models to obtain traffic optimization collaborative schemes; and the sending module is used to encrypt and send the traffic optimization collaborative schemes to the urban traffic network.
[0009] A third aspect of the embodiments of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the multi-model collaborative optimization control method for urban traffic provided in this application.
[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0011] This application monitors the urban traffic network in real time to obtain road network traffic data streams; constructs a road network traffic characteristic vector based on the road network traffic data streams; inputs the road network traffic characteristic vectors into multiple traffic optimization collaborative models to obtain a traffic optimization collaborative scheme; and encrypts and distributes the traffic optimization collaborative schemes to the urban traffic network. This invention solves the technical problem that existing single traffic optimization methods cannot effectively cope with the complex and dynamic changes of urban road networks. By constructing traffic characteristic vectors based on real-time road network data and jointly making decisions using multiple models, it achieves the technical effect of improving the overall operational efficiency of urban traffic. Attached Figure Description
[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 This is a schematic diagram of the multi-model collaborative optimization control method for urban traffic provided in the embodiments of this application;
[0014] Figure 2 This is a schematic diagram of the structure of a multi-model collaborative optimization control system for urban traffic provided in an embodiment of this application.
[0015] Figure labeling: Monitoring module 11, vector construction module 12, scheme acquisition module 13, sending module 14. Detailed Implementation
[0016] This application provides a multi-model collaborative optimization control method, system, and medium for urban traffic. It addresses the technical problem that single traffic optimization methods in the prior art cannot effectively cope with the complex dynamic changes of urban road networks. By constructing traffic characteristic vectors based on real-time road network data and jointly making decisions with multiple models, it achieves the technical effect of improving the overall operational efficiency of urban traffic.
[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0018] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.
[0019] Example 1, as Figure 1 As shown, this application provides a multi-model cooperative optimization control method for urban traffic, the method comprising:
[0020] Step S100: Monitor the urban traffic network in real time and obtain road network traffic data streams.
[0021] In this embodiment, real-time monitoring of road nodes and road segments in the urban traffic network is performed to continuously collect real-time traffic monitoring data, such as traffic flow, vehicle speed, occupancy rate, and vehicle trajectory, reflecting dynamic information about traffic conditions. The obtained real-time traffic monitoring data is then preprocessed, including cleaning, to form a road network traffic data stream.
[0022] Furthermore, the method provided in the application embodiments, which involves real-time monitoring of the urban traffic network and obtaining road network traffic data streams, further includes:
[0023] Acquire real-time traffic monitoring data of the urban traffic network; perform data cleaning on the real-time traffic monitoring data to obtain the road network traffic data stream.
[0024] In this embodiment of the application, by employing various monitoring methods such as video detectors, geomagnetic sensors, microwave radar, and induction coils at road nodes and sections of the urban traffic network, real-time traffic monitoring data reflecting the traffic operation status is continuously acquired, including dynamic information such as traffic flow, vehicle speed, lane occupancy, queue length, and vehicle trajectory.
[0025] Next, the acquired real-time traffic monitoring data undergoes data cleaning. This involves processes such as outlier removal, missing data imputation, spatiotemporal alignment, and consistency calibration to make the data more accurate, continuous, and usable. The cleaned data is then organized and structured according to road topology and time series, forming a road network traffic data stream that characterizes the operational characteristics of the urban road network.
[0026] Step S200: Construct a road network traffic characteristic vector based on the road network traffic data stream.
[0027] In this embodiment of the application, traffic information reflecting the operating status of road segments is analyzed and processed based on the road network traffic data stream. By extracting, dimensionally calibrating, interval mapping and time series processing of various types of data such as traffic flow, vehicle speed, occupancy rate, queue length, traffic capacity and road segment operating trends, different types of data are kept consistent on the same time and spatial scales.
[0028] Subsequently, the processed data is combined according to a predetermined arrangement based on the road structure relationship to form a numerical expression describing the operating state of the road network at a specific time, that is, to construct the road network traffic characteristic vector.
[0029] Step S300: Input the road network traffic characteristic vector into multiple traffic optimization collaborative models to obtain a traffic optimization collaborative scheme.
[0030] In this embodiment, when inputting the road network traffic characteristic vector into multiple traffic optimization collaborative models, the multiple traffic optimization collaborative models, including a green wave coordination model and a congestion relief model, are first activated. The road network traffic characteristic vector is then input into each model to obtain the green wave coordination scheme and the congestion relief scheme.
[0031] The green wave coordination scheme and congestion relief scheme are then integrated to generate a traffic optimization and coordination scheme.
[0032] Furthermore, in the method provided in the application embodiments, inputting the road network traffic characteristic vector into multiple traffic optimization collaborative models to obtain a traffic optimization collaborative scheme further includes:
[0033] Activate the multiple traffic optimization and coordination models, including a green wave coordination model and a congestion relief model; input the road network traffic characteristic vector into the green wave coordination model to obtain a green wave coordination scheme; input the road network traffic characteristic vector into the congestion relief model to obtain a congestion relief scheme; generate the traffic optimization and coordination scheme based on the green wave coordination scheme and the congestion relief scheme.
[0034] In this embodiment, multiple traffic optimization collaborative models are first activated, including a green wave coordination model and a congestion mitigation model. The green wave coordination model is trained using supervised learning, with historical road network traffic characteristic vectors as input and historical green wave coordination schemes as output. The congestion mitigation model is trained using supervised learning, with historical road network traffic characteristic vectors as input and historical congestion mitigation schemes as output.
[0035] Next, the real-time constructed road network traffic characteristic vector is input into the green wave coordination model for processing. The green wave coordination model generates a green wave coordination scheme based on the traffic flow, speed distribution, intersection load status, and signal operation conditions reflected in the road network traffic characteristic vector. For example, when the characteristic vector shows high traffic speed and concentrated flow in the main direction, the green wave coordination scheme can determine a larger green light ratio in the main direction and adjust the phase difference between intersections so that vehicles passing through multiple intersections at the current speed are all within the green light zone. Simultaneously, the real-time constructed road network traffic characteristic vector is input into the congestion mitigation model for processing to obtain congestion mitigation schemes. For example, when the characteristic vector shows a rapid increase in queues and a significant decrease in speed on a certain road segment, the congestion mitigation scheme can alleviate the queue expansion trend by increasing the green light time of the corresponding phase on that road segment or reducing the inflow at upstream intersections.
[0036] After obtaining the green wave coordination scheme and congestion relief scheme, safety hazard detection is performed based on the road network traffic characteristic vector to determine the traffic safety hazard detection results, and a safety hazard mitigation scheme is generated based on these results. Subsequently, the safety hazard mitigation scheme is integrated with the green wave coordination scheme and the congestion relief scheme to form a traffic optimization and coordination scheme by taking into account safety requirements, traffic efficiency, and congestion mitigation needs.
[0037] Furthermore, in the method provided in the application embodiments, generating the traffic optimization coordination scheme based on the green wave coordination scheme and the congestion relief scheme further includes:
[0038] Safety hazard detection is performed based on the road network traffic characteristic vector to determine the traffic safety hazard detection results; hazard suppression and compensation are performed based on the traffic safety hazard detection results to determine the safety hazard suppression scheme; the safety hazard suppression scheme, the green wave coordination scheme, and the congestion relief scheme are integrated to obtain the traffic optimization and coordination scheme.
[0039] In this embodiment, safety hazard detection is first performed based on the road network traffic characteristic vector. This involves analyzing operational characteristics represented by the road network traffic characteristic vector, such as changes in traffic flow, speed reduction, and queue growth trends, to predict traffic accidents and obtain traffic safety accident prediction results. Then, based on the time-series data corresponding to the road network traffic characteristic vector, the triggering factors leading to the predicted risks are traced. For example, potential conflicts caused by sudden increases in traffic flow, sudden drops in speed, or unreasonable phase release relationships at adjacent intersections are identified, thereby generating traffic safety hazard detection results.
[0040] After obtaining the results of traffic safety hazard detection, the relevant signal control parameters are adjusted to mitigate and compensate for the identified risks. This is achieved by specifically adjusting the green light ratio, green time, or phase sequence for the corresponding direction to reduce the impact of potential risks. For example, if the detection results indicate a conflict risk in a certain left-turn direction, the green light time for the left-turn phase can be appropriately extended, or a protective phase can be added, while ensuring safety constraints, to form a safety hazard mitigation plan.
[0041] Finally, the safety hazard mitigation scheme, green wave coordination scheme, and congestion relief scheme are integrated. During this process, control parameters such as signal cycle, green light time, release sequence, and phase difference involved in the three schemes are compared sequentially to check for overlapping release times, phase sequence conflicts, or inconsistent cycles. Any parameter conflicts discovered are adjusted according to actual road operating conditions to ensure that all parameters simultaneously meet the execution requirements within the same signal cycle. Subsequently, the coordinated parameters are recombined to form a unified signal control configuration, thereby generating a traffic optimization and coordination scheme.
[0042] Furthermore, in the method provided in the application embodiments, the process of detecting safety hazards based on the road network traffic characteristic vector and determining the traffic safety hazard detection result further includes:
[0043] Traffic safety accident prediction is performed based on the road network traffic characteristic vector to obtain traffic safety accident prediction results; triggering factors are traced based on the road network traffic characteristic vector to generate traffic safety hazard detection results.
[0044] In this embodiment, when predicting traffic safety accidents based on the road network traffic characteristic vector, a threshold judgment method is used to analyze the traffic flow, vehicle speed, and queue length in the road network traffic characteristic vector. Specifically, the current traffic flow is compared with a preset traffic flow threshold; if the traffic flow exceeds the threshold, a traffic flow anomaly is identified. The current vehicle speed is compared with a minimum safe speed threshold; if the vehicle speed is lower than the threshold, a speed anomaly is identified. The queue length is compared with a queue length threshold; if the queue length exceeds the threshold, a queue anomaly is identified. When any anomaly is triggered, it is determined that there is a potential accident risk on that road segment or intersection, thereby obtaining the traffic safety accident prediction result.
[0045] Subsequently, when tracing the triggering factors of traffic safety accident prediction results based on the road network traffic characteristic vector, a time-series backtracking analysis method is used to locate the source of the characteristic vector corresponding to the traffic safety accident prediction results. Specifically, based on the road network traffic characteristic vector used to generate the traffic safety accident prediction results, features such as traffic flow, vehicle speed, and queue length are analyzed item by item. By comparing the direction and degree of deviation of these features from the thresholds, the main operational changes that cause the anomalies are determined. For example, when the accident prediction result is triggered by speed anomalies, the change causing the speed anomaly is identified as the triggering factor by examining the magnitude and rate of decrease of speed relative to the speed threshold. Through the above analysis steps, the traffic safety hazard detection results are finally generated.
[0046] Furthermore, the method provided in the application embodiments also includes:
[0047] Based on the traffic safety hazard detection results, a safety hazard alarm is generated.
[0048] In this embodiment, abnormal operational characteristics identified in the traffic safety hazard detection results are judged, including features representing potential risks such as abnormal speed, abnormal traffic flow, and abnormal queuing. These features are compared with preset safety level thresholds to determine their risk level. When the risk level reaches the alarm triggering condition, a safety hazard alarm is generated based on the corresponding risk level.
[0049] Step S400: The traffic optimization and coordination scheme is encrypted and distributed to the urban traffic network.
[0050] In this embodiment, when the traffic optimization coordination scheme is encrypted and distributed to the urban traffic network, the control parameters in the scheme are first encrypted. These control parameters include information such as signal cycle, green light time, phase difference, and release order. These parameters are then converted into a secure, transmittable format using encryption. Subsequently, the encrypted control parameters are sent to the corresponding traffic signal control equipment via communication, enabling the equipment to execute the traffic optimization coordination scheme according to the control parameters after receiving and decrypting them.
[0051] In summary, the embodiments of this application have at least the following technical effects:
[0052] This application monitors the urban traffic network in real time to obtain road network traffic data streams; constructs a road network traffic characteristic vector based on the road network traffic data streams; inputs the road network traffic characteristic vectors into multiple traffic optimization collaborative models to obtain a traffic optimization collaborative scheme; and encrypts and distributes the traffic optimization collaborative schemes to the urban traffic network. This invention solves the technical problem that existing single traffic optimization methods cannot effectively cope with the complex and dynamic changes of urban road networks. By constructing traffic characteristic vectors based on real-time road network data and jointly making decisions using multiple models, it achieves the technical effect of improving the overall operational efficiency of urban traffic.
[0053] Example 2, based on the same inventive concept as the multi-model cooperative optimization control method for urban traffic in the previous examples, such as... Figure 2 As shown, this application provides a multi-model cooperative optimization control system for urban traffic. The system and method embodiments in this application are based on the same inventive concept. The system includes:
[0054] The monitoring module 11 is used to monitor the urban traffic network in real time and obtain the road network traffic data stream; the vector construction module 12 is used to construct the road network traffic characteristic vector based on the road network traffic data stream; the scheme acquisition module 13 is used to input the road network traffic characteristic vector into multiple traffic optimization collaborative models to obtain traffic optimization collaborative schemes; and the sending module 14 is used to encrypt and send the traffic optimization collaborative schemes to the urban traffic network.
[0055] Furthermore, the system is also used to implement the following functions:
[0056] Acquire real-time traffic monitoring data of the urban traffic network; perform data cleaning on the real-time traffic monitoring data to obtain the road network traffic data stream.
[0057] Furthermore, the system is also used to implement the following functions:
[0058] Activate the multiple traffic optimization and coordination models, including a green wave coordination model and a congestion relief model; input the road network traffic characteristic vector into the green wave coordination model to obtain a green wave coordination scheme; input the road network traffic characteristic vector into the congestion relief model to obtain a congestion relief scheme; generate the traffic optimization and coordination scheme based on the green wave coordination scheme and the congestion relief scheme.
[0059] Furthermore, the system is also used to implement the following functions:
[0060] Safety hazard detection is performed based on the road network traffic characteristic vector to determine the traffic safety hazard detection results; hazard suppression and compensation are performed based on the traffic safety hazard detection results to determine the safety hazard suppression scheme; the safety hazard suppression scheme, the green wave coordination scheme, and the congestion relief scheme are integrated to obtain the traffic optimization and coordination scheme.
[0061] Furthermore, the system is also used to implement the following functions:
[0062] Traffic safety accident prediction is performed based on the road network traffic characteristic vector to obtain traffic safety accident prediction results; triggering factors are traced based on the road network traffic characteristic vector to generate traffic safety hazard detection results.
[0063] Furthermore, the system is also used to implement the following functions:
[0064] Based on the traffic safety hazard detection results, a safety hazard alarm is generated.
[0065] In Example 3, based on the multi-model collaborative optimization control method for urban traffic in the foregoing embodiments and with the same inventive concept, this application also provides a computer-readable storage medium storing a computer program, which, when executed, implements the steps of any one of the methods in Example 1 above.
[0066] It should be noted that the order of the embodiments described above is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.
[0067] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.
Claims
1. A multi-model collaborative optimization control method for urban traffic, characterized in that, The method includes: Real-time monitoring of urban traffic networks to obtain road network traffic data streams; Based on the road network traffic data flow, construct a road network traffic characteristic vector; The road network traffic characteristic vector is input into multiple traffic optimization collaborative models to obtain a traffic optimization collaborative scheme. The traffic optimization and coordination scheme is encrypted and distributed to the urban traffic network.
2. The multi-model collaborative optimization control method for urban traffic as described in claim 1, characterized in that, Real-time monitoring of urban traffic networks to obtain road network traffic data streams, including: Obtain real-time traffic monitoring data of the urban traffic network; The real-time traffic monitoring data is cleaned to obtain the road network traffic data stream.
3. The multi-model collaborative optimization control method for urban traffic as described in claim 1, characterized in that, The road network traffic characteristic vector is input into multiple traffic optimization collaborative models to obtain traffic optimization collaborative schemes, including: Activate the multiple traffic optimization coordination models, which include a green wave coordination model and a congestion relief model; The road network traffic characteristic vector is input into the green wave coordination model to obtain the green wave coordination scheme; Input the road network traffic characteristic vector into the congestion relief model to obtain a congestion relief scheme; Based on the green wave coordination scheme and the congestion relief scheme, the traffic optimization coordination scheme is generated.
4. The multi-model collaborative optimization control method for urban traffic as described in claim 3, characterized in that, Based on the green wave coordination scheme and the congestion relief scheme, the traffic optimization coordination scheme is generated, including: Safety hazard detection is performed based on the road network traffic characteristic vector to determine the traffic safety hazard detection results; Based on the traffic safety hazard detection results, hazard mitigation and compensation will be carried out to determine the hazard mitigation plan; By integrating the aforementioned safety hazard mitigation scheme, the aforementioned green wave coordination scheme, and the aforementioned congestion relief scheme, the traffic optimization and coordination scheme is obtained.
5. The multi-model collaborative optimization control method for urban traffic as described in claim 4, characterized in that, Based on the road network traffic characteristic vector, safety hazard detection is performed to determine the traffic safety hazard detection results, including: Based on the road network traffic characteristic vector, traffic safety accident prediction results are obtained. Based on the road network traffic characteristic vector, the triggering factors of the traffic safety accident prediction results are traced to generate the traffic safety hazard detection results.
6. The multi-model collaborative optimization control method for urban traffic as described in claim 4, characterized in that, Based on the traffic safety hazard detection results, a safety hazard alarm is generated.
7. A multi-model collaborative optimization control system for urban traffic, characterized in that, The system is used to execute the multi-model cooperative optimization control method for urban traffic as described in any one of claims 1-6, and the system includes: The monitoring module is used to monitor the urban traffic network in real time and obtain road network traffic data streams; The vector construction module is used to construct road network traffic characteristic vectors based on the road network traffic data stream; The scheme acquisition module is used to input the road network traffic characteristic vector into multiple traffic optimization collaborative models to obtain traffic optimization collaborative schemes. The sending module is used to encrypt and send the traffic optimization coordination scheme to the urban traffic network.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the multi-model collaborative optimization control method for urban traffic as described in any one of claims 1-6.