A single-lane two-way traffic distributed control method and system based on real-time dynamic monitoring

Abstract

The application discloses a kind of single lane two-way traffic distributed control method and system based on real-time dynamic monitoring, it is related to intelligent transportation and distributed control technical field.The application is by being laid in the entrance, exit and meeting area of single lane two-way traffic section Sensing device, real-time collection traffic information;Distributed control unit is deployed in each controlled section, and autonomous architecture is used without central server, adjacent unit is coordinated by local communication;Based on real-time traffic state dynamic calculation vehicle emptying time, signal light self-adaptive switching is realized in combination with meeting area capacity, response delay is less than 50 milliseconds;With fault degradation mechanism simultaneously, when node fails, it is taken over by adjacent unit and switched to safety control strategy.The application effectively solves the problem that traditional centralized, timing management and control are poor in real-time, weak in adaptability and insufficient in robustness in long-distance complex scene, significantly reduces vehicle waiting time, improves traffic efficiency and operation safety, and is suitable for mountainous road, tunnel construction, mine transportation and emergency passage and the like scene.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation and distributed control technology, specifically to a distributed control method and system for single-lane two-way traffic based on real-time dynamic monitoring. Background Technology

[0002] In the fields of transportation and road construction, single-lane two-way traffic is a common traffic organization scenario, especially in environments with limited conditions such as mountain roads, long tunnels, construction access roads, mining transport roads, and temporary roads in scenic areas. Because the road width is insufficient to accommodate simultaneous two-way traffic, traffic control measures are typically needed to ensure the orderly alternation of vehicles. Existing technologies commonly employ control methods including manual direction-taking, single-point traffic light control, and timed alternation control based on fixed cycles. These methods can maintain basic traffic order over short distances and with low traffic volume, but their limitations become increasingly apparent when facing long-distance road sections and highly volatile traffic flows.

[0003] Manual traffic control primarily relies on manned checkpoints or flag bearers. This method depends on personnel experience, suffers from significant response delays, and struggles to achieve full coordination over long distances. Traffic light control typically uses alternating signals of fixed durations at both ends of a road segment to allow vehicles to pass in both directions. However, this statically timed approach usually assumes relatively stable vehicle speeds and arrival rates, making it difficult to adapt to random traffic flow and unexpected events. When the road segment is long, vehicles need more time to clear the segment during a single green light, significantly increasing waiting times for oncoming vehicles, and in severe cases, even resulting in both empty-vehicle passage and road congestion. Furthermore, timed schemes often cannot be adjusted according to actual road conditions, leading to decreased traffic efficiency and a poor driving experience.

[0004] In recent years, some studies have attempted to optimize signal control based on traffic flow detection and queue length estimation. For example, by detecting two-way arrival rates and combining this with pre-setting switching cycles based on road segment lengths, so-called "saturation alternation" or "fixed queue length alternation" control can be achieved. However, these methods have two significant problems in practical engineering: First, they rely on accurate traffic flow input and a stable road traffic environment, making it difficult to guarantee the validity of model assumptions in long-distance, curved, sloped, and complex weather conditions. Second, these solutions typically rely on centralized computing and control, requiring a central server to process and distribute global data. The computational and communication delays make it difficult to meet the dynamic requirements of complex scenarios in terms of real-time performance. Especially in mountainous or construction environments, where communication conditions are limited, centralized solutions are at risk of failure.

[0005] Therefore, existing technologies generally face three key challenges in managing long-distance single-lane two-way traffic. First, there is insufficient real-time performance. Existing timed and centralized control methods suffer from lag, making it difficult to update light-color switching logic within milliseconds, leading to significantly increased vehicle waiting times. Second, there is poor adaptability. Existing methods are typically based on static parameter settings, unable to flexibly adjust for nighttime, rainy / foggy weather, or partial vehicle congestion, resulting in large fluctuations in traffic efficiency. Third, there is low robustness. When some sensors or nodes fail, the centralized control architecture often leads to overall control failure, lacking effective local takeover and degradation mechanisms.

[0006] In summary, while existing single-lane two-way traffic control methods can address traffic organization issues to some extent in short-distance or low-complexity scenarios, they still suffer from significant problems in long-distance, multi-interference, and complex road environments, including poor real-time performance, insufficient adaptability, and a lack of robustness. These issues directly lead to a significant increase in average vehicle waiting time and may even pose traffic safety risks. Therefore, there is an urgent need for a traffic control method that does not require a central server, relies on distributed autonomy, can optimize light-color switching logic through real-time dynamic monitoring, and has the capability to degrade operations in the event of node failures, to meet the practical needs of long-distance single-lane two-way traffic. Summary of the Invention

[0007] The purpose of this invention is to address the problems of insufficient real-time performance, excessive waiting time, over-reliance on central servers, and lack of robustness in existing single-lane two-way traffic control methods in long-distance and complex scenarios. This invention provides a distributed control method and system for single-lane two-way traffic based on real-time dynamic monitoring.

[0008] This method deploys vehicle monitoring units at road entrances, exits, and key passing zones, combined with segment-level distributed control units, to dynamically collect bidirectional traffic flow data, calculate signal switching criteria in real time, and adaptively switch traffic light colors. The switching criteria may include, but are not limited to, parameters such as vehicle clearance time, segment occupancy rate, or passing zone capacity, which are dynamically determined by the control unit based on actual traffic data. In this process, the invention emphasizes the rational design and optimization of the light switching logic, minimizing vehicle waiting time when passing through conflict zones while ensuring traffic safety and overall operational efficiency. Compared to existing centralized or timed alternation methods, this invention does not rely on a central server and can complete data processing and decision-making with a latency of less than 50 milliseconds under low computing power conditions, significantly improving traffic control capabilities in long-distance scenarios.

[0009] To achieve the above objectives, the technical solution proposed by this invention includes the following:

[0010] A distributed control method for single-lane two-way traffic based on real-time dynamic monitoring includes the following steps:

[0011] S1. Multi-point sensor deployment: Along a single-lane bidirectional road section, sensor devices are deployed at both entrances and exits to collect real-time data on the number of vehicles entering, exiting, and remaining, as well as their speed. The sensor devices are inductive loops, video recognition equipment, millimeter-wave radar, infrared sensors, or multi-sensor fusion devices, and have the functions of vehicle detection, counting, and speed acquisition.

[0012] S2. Deployment of Distributed Control Units: Distributed control units are set up in each controlled section. Each distributed control unit is connected to the corresponding sensing device. Adjacent distributed control units interact with each other through local communication to form a distributed autonomous management and control architecture. Each distributed control unit has independent data processing, local decision-making and section collaboration functions, without the need for a central server to participate in global calculations.

[0013] S3. Real-time analysis of traffic flow status and dynamic calculation of clearance time: Based on the data collected by the sensing device, the distributed control unit analyzes the vehicle inventory, traffic arrival rate and average driving speed of the current section in real time, and dynamically calculates the clearance time for vehicles in the current direction of travel to completely leave the controlled section by combining the section length and safety redundancy factor.

[0014] In this step, sensors can be used to identify and distinguish between heavy-duty and light-duty vehicles, and automatically configure safety redundancy factors accordingly. Heavy-duty vehicles have larger safety redundancy factor values, thereby accurately matching the driving characteristics of different vehicle types and improving the accuracy of emptying time calculation.

[0015] S4. Dynamic signal light switching control based on collaborative logic: The distributed control unit uses the clearing time as the core criterion and combines it with real-time monitoring data of the capacity of the passing area in adjacent sections. It monitors the traffic status of vehicles in the section in real time through the sensing device. After confirming that all vehicles in the current direction have left the controlled section, it makes a decision on switching the direction of traffic light release. When the vehicles in the current direction have completely left the controlled section within the clearing time and the passing area in the opposite direction has remaining capacity, the signal light color is switched, and the switching response delay is less than 50 milliseconds. Adjacent distributed control units make collaborative predictions in advance to avoid traffic conflicts and congestion in the passing area.

[0016] This step uses real-time monitoring data from sensors to directly verify whether vehicles have completely left, eliminating the risk of vehicle congestion and traffic conflicts caused by relying solely on predictions of clearance times, thus achieving a closed-loop management system.

[0017] S5. Fault Degradation and Adaptive Control: The distributed control unit monitors its own and adjacent nodes and sensing devices in real time. When a node failure, communication interruption, or sensing device failure is detected, the fault degradation mechanism is automatically activated, and the adjacent normal control unit takes over the control authority of the faulty section, switching to a simplified fixed-cycle alternating traffic strategy to ensure basic traffic order on the road section. At the same time, the distributed control unit dynamically adjusts the safety redundancy factor and clearing time parameters according to weather, road conditions, and vehicle type to adapt to different operating environments.

[0018] S6. Digital Twin-Assisted Optimization: Real-world sensing data is synchronized to the digital twin platform to build a virtual simulation model of road traffic flow. The traffic light switching logic is simulated and verified, and the parameters are iteratively optimized. The optimized strategy is then sent to the distributed control unit for execution after simulation verification, thus avoiding on-site deployment risks.

[0019] Furthermore, the capacity of the passing area is monitored in real time. The number of vehicles in the area is collected in real time by the sensing devices deployed in the passing area. The distributed control unit calculates the remaining carrying capacity of the passing area and synchronizes the capacity data to the adjacent control unit. When the remaining capacity of the passing area is lower than a preset threshold, the upstream section suspends the passage of vehicles until the passing area frees up carrying space.

[0020] Furthermore, the distributed control unit adopts an edge computing embedded device to complete data processing and decision-making locally, and local communication adopts short-range wireless or wired communication methods, which is suitable for complex communication environments such as mountainous areas and construction sites.

[0021] Furthermore, the safety redundancy factor is dynamically adjusted according to the operating environment. In low visibility and low-speed driving environments such as rain, fog, and steep slopes, the safety redundancy factor is increased; in sunny and flat road conditions, the safety redundancy factor is decreased, balancing traffic safety and efficiency.

[0022] Furthermore, in the dynamic calculation of clearance time, vehicle type identification and classification are achieved by relying on sensing devices. Larger safety redundancy factors are configured for heavy-duty vehicles such as large trucks and heavy-duty engineering vehicles, while smaller safety redundancy factors are configured for small private cars and light motor vehicles, so as to realize differentiated calculation of clearance time and take into account both traffic safety and overall traffic efficiency.

[0023] The present invention also discloses a distributed control system for single-lane two-way traffic based on real-time dynamic monitoring, including multi-point sensing devices, several distributed control units, traffic light execution units, and digital twin optimization units.

[0024] The multi-point sensing devices are deployed at bidirectional entrances, road segment exits, and key passing area entrances and exits to collect real-time data on the number of vehicles entering, exiting, and remaining, as well as their speed.

[0025] The distributed control unit is connected to the multi-point sensing device in a one-to-one communication manner, and adjacent distributed control units complete data interaction through local communication, forming a distributed autonomous management and control architecture without the dependence on a central server.

[0026] The distributed control unit is used to analyze traffic flow status in real time, dynamically calculate vehicle clearance time, confirm that all vehicles have left through sensor data, and make traffic light release direction decisions based on the capacity of the passing area, and realize fault degradation takeover and environmental adaptive control.

[0027] The signal light execution unit is connected to the distributed control unit and performs light color switching according to control commands.

[0028] The digital twin optimization unit is used to synchronize real-world sensing data to the virtual simulation platform to complete the simulation verification and parameter iterative optimization of the control strategy.

[0029] This system abandons the centralized control architecture and achieves safe, efficient and stable management of long-distance single-lane two-way traffic through distributed autonomy, segment collaboration, closed-loop verification of vehicle departure, fault degradation and environmental adaptive adjustment.

[0030] The beneficial effects of this invention are as follows:

[0031] (1) In terms of computing power requirements, the distributed control unit of this invention only needs to perform simple calculations based on vehicle counting and speed, with a response time of less than 50 milliseconds, which is far lower than the second-level latency commonly found in centralized methods. This low computing power requirement enables this invention to be deployed in embedded devices or edge computing platforms without relying on a high-performance central server, thereby reducing system construction and maintenance costs and improving the feasibility of engineering applications.

[0032] (2) Regarding traffic efficiency, field tests and simulations show that this invention can significantly reduce the average waiting time for vehicles. In typical long-distance single-lane two-way traffic scenarios, compared with the traditional timed alternation method, the average waiting time is reduced by 40% to 60%, and the vehicle throughput is significantly improved. This effect is achieved thanks to the real-time dynamic monitoring-based light color switching logic proposed in this invention. This logic not only considers the entry and exit of vehicles, but also combines dynamic feedback of clearing time and meeting zone capacity, thereby maximizing road utilization efficiency while ensuring safety.

[0033] (3) In terms of solving technical problems, the present invention effectively overcomes the real-time bottleneck in long-distance scenarios. Existing timed control methods cause oncoming vehicles to wait for a long time due to excessively long cycle settings, while centralized optimization methods cannot meet the needs of dynamic scenarios due to data transmission and calculation delays. However, the present invention achieves high real-time performance and high reliability in traffic organization in long-distance road sections through a distributed architecture and fast-response light color switching logic. Attached Figure Description

[0034] Figure 1 These are the parameters and solution results of this invention in numerical experiments;

[0035] Figure 2 This invention provides an optimized signal timing scheme and a vehicle spatiotemporal trajectory diagram.

[0036] Figure 3 This is a sensitivity analysis diagram of the average velocity based on the present invention;

[0037] Figure 4 This is a sensitivity analysis chart of the present invention regarding the proportion of large trucks;

[0038] Figure 5 This is a sensitivity analysis chart based on the minimum green light duration of this invention. Detailed Implementation

[0039] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to examples. The following content is merely illustrative and explanatory of the concept of the present invention. Those skilled in the art can make various modifications or additions to the described specific embodiments or use similar methods to replace them, as long as they do not deviate from the inventive concept, all of which should fall within the protection scope of the present invention. The following description, in conjunction with the appendix... Figure 1-5 The specific embodiments of the present invention will be described in further detail below.

[0040] This embodiment addresses the control problem of long-distance single-lane two-way traffic by providing system modeling and engineering verification. In the numerical analysis, a single-lane two-way traffic route exceeding 500.0 m in total length with two meeting points was constructed, and signal control units and vehicle monitoring units were deployed at both ends of the road and at the entrances and exits of the meeting points. To ensure the universality of the method, traffic flow was set between 100.0 and 1000.0 veh / h, signal cycle between 120.0 and 300.0 s, and the green light lower limit was 60.0 s. Based on these conditions, the optimization model was run, and the results are as follows: Figure 1 As shown.

[0041] Figure 1This presents the parameters and solution results of the present invention in numerical experiments, demonstrating the average solution time under different traffic flow and planning time conditions. Figure 1 As can be seen, under low traffic conditions, the model can output the optimal solution within 70.0 s with an optimal gap value of 0.00%, demonstrating significant real-time performance. Under high-load conditions with a traffic flow of 1000.0 veh / h and a cycle of 0.5 h, the average solution time is 706.12 s with a gap value of 5.00%. Although the computation time increases, it is still within an acceptable engineering range. These results prove that the distributed optimization method proposed in this invention can maintain stable computational performance and feasibility under various traffic intensity conditions.

[0042] In the engineering application section, this embodiment selects a tunnel as a case study. The tunnel is approximately 1100.0 m long, with two one-way passing platforms on both sides and three two-way passing points. The platform lengths are 330.0 m, 165.0 m, 99.0 m, 99.0 m, and 495.0 m, respectively. Ten sets of traffic lights are installed inside the tunnel, including at the two entrances and the entrances / exits of each platform. Traffic characteristics include a peak two-way flow of 873.0 veh / h, a large vehicle proportion of 9.8%, and an average speed of 80.0 km / h. Radar-visual fusion sensors are installed at both ends of the tunnel, and the prediction error for vehicle arrival time is no greater than 5.0 s.

[0043] Under these conditions, the system architecture and execution flow for signal timing optimization using the method of this invention are as follows: Figure 2 As shown.

[0044] Figure 2 This diagram illustrates the distributed control system architecture and process flow of this embodiment, showcasing the information interaction and decision-making logic between each monitoring node and the control unit. The system acquires real-time vehicle traffic data through vehicle monitoring units deployed at road entrances / exits and meeting points, and transmits the observed information to the distributed control unit. The control unit internally includes an information interaction module, a local computing module, and an anomaly detection module. The information interaction module is responsible for communication and collaborative optimization between adjacent sections; the local computing module calculates the optimal traffic time slot based on real-time data; and the anomaly detection module triggers a degradation mechanism when it detects equipment failure or communication interruption, switching to fixed-cycle alternating control. Finally, the light color switching logic integrates the outputs of all modules and sends instructions to the signal execution unit to achieve precise control of the on-site traffic lights. This architecture ensures high reliability and real-time response capabilities in long-distance, multi-node scenarios.

[0045] Simulation results show that under the optimized scheme, the total delay for 163 vehicles is 11611.0 seconds, with an average delay of 71.23 seconds per vehicle. In contrast, the traditional scheme with signals at only two ends results in a total delay of 30737.0 seconds. The method of this invention reduces delay by 62.2%, significantly improving traffic efficiency while maintaining safety throughout the process.

[0046] To further examine the adaptability and robustness of the method, this embodiment combines... Figures 3-5 Multi-parameter sensitivity analysis was performed.

[0047] Figure 3 This is a sensitivity analysis chart of the present invention based on average speed. When the speed increases from 40.0 km / h to 60.0 km / h, the average delay decreases by more than 35%, while when it continues to increase from 80.0 km / h to 120.0 km / h, the delay decreases by less than 11%. It is evident that increasing speed has a more significant effect on improving efficiency in the low-speed range, while in the high-speed range, the system bottleneck mainly comes from the signal cycle. The method of this invention, through distributed coordination and dynamic adjustment, can maintain stable operating performance under different speed conditions.

[0048] Figure 4 This is a sensitivity analysis chart based on the proportion of heavy trucks in this invention. When the proportion of heavy trucks increases from 10% to 30%, the average delay increases significantly to 104.56 s / veh. When the proportion further increases to 50%, the delay reaches 180.31 s / veh, and the system performance deteriorates significantly. The method of this invention can appropriately extend the clearing time and adjust the timing based on the detected vehicle type and structure, thereby mitigating the adverse effects of the increased proportion of heavy trucks to some extent.

[0049] Figure 5 This is a sensitivity analysis chart based on the minimum green light duration of this invention. When the minimum green light duration is increased from 40.0 s to 60.0 s, the increase in average delay is limited. However, when it is further extended to 100.0 s, the delay increase reaches 24.24 s / veh, and the system flexibility decreases significantly. The method of this invention effectively avoids the reduction in operating efficiency caused by improper setting of a single parameter by introducing a dual boundary of green light duration and clearing duration during the modeling stage.

[0050] In summary, this embodiment, through the combination of numerical analysis and a tunnel case study, verifies the engineering value of the single-lane bidirectional traffic distributed control method based on real-time dynamic monitoring proposed in this invention. Figures 1-5The results show that the method of the present invention can significantly reduce delays and improve efficiency under long-distance single-lane two-way traffic conditions, while maintaining safety and stability under different traffic demands and operating environments. Compared with existing "two-end control" methods, the present invention can not only quickly output the optimal solution under low-flow conditions, but also provide a near-optimal solution within a reasonable time under high-flow conditions; at the same time, it has strong adaptability to changes in vehicle type ratio, speed conditions, and signal cycle. This embodiment fully demonstrates that the present invention has innovation, practicality, and promotion potential in both theoretical modeling and engineering practice.

Claims

1. A distributed control method for single-lane two-way traffic based on real-time dynamic monitoring, characterized in that, Includes the following steps: S1. Along a single-lane two-way road section, sensors are installed at both entrances and exits to collect real-time data on the number of vehicles entering, exiting, and remaining, as well as their speed. S2. Deploy distributed control units in each controlled section. Each distributed control unit is connected to the local sensing device. Adjacent distributed control units can exchange data through local communication to build a distributed autonomous control architecture. S3. The distributed control unit analyzes the traffic flow status of the section in real time based on the data collected by the sensing device, and dynamically calculates the clearance time for vehicles in the current direction of travel to completely leave the controlled section by combining the section length and safety redundancy factor. S4. Based on the clearing time as the core judgment criterion, combined with real-time monitoring data of the meeting area capacity, the distributed control unit monitors and confirms in real time through the sensing device that all vehicles in the current direction of travel have left the controlled section, and then autonomously completes the decision to switch the direction of traffic light release. S5. Monitor the operating status of nodes and equipment in real time. When a fault is detected, activate the degradation mechanism and let the adjacent normal control unit take over the control of the faulty section to ensure basic traffic flow. At the same time, dynamically adjust the control parameters according to the operating environment to achieve adaptive control.

2. The control method according to claim 1, characterized in that, The sensing device mentioned in S1 is an inductive loop, video recognition equipment, millimeter-wave radar, infrared sensor, or multi-sensor fusion device, which has the functions of vehicle detection, real-time counting, and driving speed acquisition.

3. The control method according to claim 1, characterized in that, The distributed control unit described in S2 uses an embedded edge computing device, eliminating the need for a central server to participate in global computing, and local communication uses short-range wireless or wired communication methods.

4. The control method according to claim 1, characterized in that, The response delay of the distributed control unit in S2 in making the decision to switch the traffic light release direction is less than 50 milliseconds.

5. The control method according to claim 1, characterized in that, During the dynamic calculation of the emptying time in S3, the vehicle type is identified and automatically matched with the corresponding safety redundancy factor based on the sensor device. The safety redundancy factor value matched for heavy-duty vehicles is greater than that matched for light-duty vehicles.

6. The control method according to claim 1, characterized in that, The switching decision described in S4 requires two conditions to be met simultaneously: vehicles in the current direction of traffic must completely leave the controlled section within the clearing time, and the remaining carrying capacity of the oncoming traffic area must be higher than a preset threshold. Adjacent distributed control units can coordinate and predict traffic flow in advance to avoid traffic conflicts and congestion in passing areas.

7. The control method according to claim 1, characterized in that, The fault degradation mechanism described in S5 is as follows: when a node fault, communication interruption, or sensor failure is detected, the adjacent normal control unit automatically takes over the control authority of the faulty section and switches to a fixed-period alternating passage strategy.

8. The control method according to claim 1, characterized in that, In addition to implementing the S1 to S5 control processes, a digital twin-assisted optimization step is added: real-world sensing data is synchronized to the digital twin platform to build a virtual simulation model of road traffic flow. After completing the simulation verification and parameter optimization of the traffic light switching logic, the data is then sent to the distributed control unit for execution.

9. A distributed control system for single-lane two-way traffic based on real-time dynamic monitoring, characterized in that, For performing the method according to any one of claims 1-8, comprising: Sensing device, distributed control unit, traffic light execution unit; The sensors are installed at the two-way entrances, road exits, and passing areas; The distributed control unit is communicatively connected to the sensing device and connected to the traffic light execution unit to realize distributed autonomous control and fault degradation operation.

10. The system according to claim 9, characterized in that, It also includes a digital twin optimization unit for simulating and verifying control strategies.

Images