Intersection signal control method and system

Abstract

The invention relates to the technical field of traffic signal control, and provides an intersection signal control method and system, and the method comprises the steps: obtaining the abnormal information of rail transit operation; sensing and analyzing according to the abnormal information, and judging the type of the ground traffic retention state; executing a corresponding signal scheduling strategy according to the type of the ground traffic retention state; the signal scheduling strategy comprises the following steps: if the type of the ground traffic detention state is a bus detention state, executing the scheduling of dredging a downstream channel and intensively releasing buses; and if the type of the ground traffic retention state is the temporary parking interference state of the connection vehicle, executing the scheduling of compressing the green light of the interfered lane and prolonging the green light of the non-interfered key lane. The method has the advantages that the problem of ground traffic detention caused by abnormal operation of rail transit can be effectively solved, and the intersection passing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the technical field of traffic signal control, and specifically to a method and system for controlling intersection signals. Background Art

[0002] In urban traffic management, intersections near rail transit stations often bear enormous traffic pressure. These intersections are not only key nodes on urban arterial roads, but also important hubs for rail transit passengers to enter and exit stations and transfer to surface transportation. Their design typically takes into account the connection needs of various transportation modes such as pedestrians, surface buses, taxis, and ride-hailing services. For example, wide pedestrian crossings are set up in all directions of the intersection, and multiple surface bus stops are set up in the downstream exit lanes, especially the east-west exit lanes. At the same time, dedicated temporary parking areas for ride-hailing services and taxis are planned in the surrounding area to meet the diverse connection needs of passengers.

[0003] However, when the rail transit system experiences sudden operational anomalies, such as power outages causing train stoppages or large passenger congestion, existing signaling methods become problematic. These anomalies rapidly alter passenger travel strategies, forcing those planning to continue using rail transit to switch to surface public transportation, ride-hailing services, or taxis. This unexpected surge in passengers, far exceeding the scale and speed that signal control systems can predict based on conventional data, creates a chain reaction on the surrounding surface traffic environment. Ultimately, localized congestion at intersections quickly escalates into widespread traffic inefficiency, with some directions even experiencing complete traffic standstill, causing severe and lasting impacts on the entire urban road network.

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application provides an intersection signal control method and system, which has the advantages of effectively addressing ground traffic congestion caused by abnormal operation of rail transit and improving intersection traffic efficiency.

[0006] This application provides a method for controlling intersection signals, including: Obtain abnormal information about rail transit operations; Based on the abnormal information, the type of ground traffic congestion is determined; Based on the type of ground traffic congestion, corresponding signal dispatch strategies are implemented. These strategies include: if the ground traffic congestion is a bus congestion, downstream channel diversion and centralized release of buses are implemented; and if the ground traffic congestion is a temporary stop interference of shuttle vehicles, the green light of the affected lanes is shortened and the green light of the unaffected key lanes is extended.

[0007] The above solution can intelligently identify the types of ground traffic congestion based on abnormal information from rail transit and execute targeted scheduling, effectively alleviating traffic congestion.

[0008] To further address the issue, this application also proposes that, when the type of ground traffic congestion is a bus congestion, the steps for implementing downstream channel diversion and centralized release of buses include: When the type of ground traffic congestion is bus congestion, the downstream exit lane diversion phase is activated. During the downstream exit lane traffic control phase, green light pulses and red light buffers are periodically executed; During the red light buffer period, monitor the length of the vehicle queue in the U-turn lane area; The green light pulse is adjusted according to the length of the vehicle queue, thus regulating the release rhythm of downstream vehicles. When the downstream exit main lane is cleared and the vehicle queue length has not reached the preset overflow warning threshold, the downstream exit lane diversion phase ends. After the downstream exit lane traffic control phase ends, the centralized release phase for buses will begin.

[0009] The above solutions effectively alleviate congestion at bus stop areas and improve bus traffic efficiency by diverting traffic from downstream areas and releasing buses in a concentrated manner when buses are stuck.

[0010] To improve the solution, this application also proposes steps for monitoring the vehicle queue length in the U-turn lane area during the red light buffer period, including: Acquire video image information of the U-turn lane area; Acquire millimeter-wave radar detection information in the U-turn lane area; Based on video image information, identify vehicle targets and calculate the length of the first vehicle queue; Based on millimeter-wave radar detection information, vehicle targets are identified and the length of the second vehicle platoon is calculated; Based on video image information and millimeter-wave radar detection information, the effectiveness of the video image information and millimeter-wave radar detection information is evaluated, and the effectiveness evaluation results are obtained. Based on the effectiveness assessment results, the vehicle queue length of the U-turn lane is obtained by merging the first vehicle queue length and the second vehicle queue length.

[0011] The above scheme uses multi-source data fusion to accurately monitor the length of vehicle queues in U-turn lanes, improving data accuracy and reliability.

[0012] To improve the solution, this application also proposes a step for evaluating the effectiveness of video image information and millimeter-wave radar detection information based on the latter, and obtaining the effectiveness evaluation result, including: Image quality analysis is performed on video image information to obtain image quality analysis results; Signal feature analysis is performed on the millimeter-wave radar detection information to obtain the signal feature analysis results; By comparing the consistency between the vehicle target in the video image information and the vehicle target in the millimeter-wave radar detection information, a consistency result is obtained; Based on the image quality analysis results, signal feature analysis results, and consistency results, the validity of video image information and millimeter-wave radar detection information is judged, and the validity assessment results are obtained.

[0013] The above scheme, through a multi-dimensional evaluation mechanism, ensures the validity of video images and millimeter-wave radar detection information, providing reliable data for queue length calculation.

[0014] To improve the solution, this application also proposes a step to obtain the vehicle queue length of the U-turn lane by integrating the first vehicle queue length and the second vehicle queue length based on the effectiveness assessment results, including: Based on the effectiveness assessment results, obtain the historical effectiveness assessment results of video image information and millimeter-wave radar detection information within a preset time period; Based on the effectiveness assessment results and historical effectiveness assessment results, the dynamic weights of video image information and millimeter-wave radar detection information are determined. Based on dynamic weights, the lengths of the first and second vehicle queues are weighted and fused to obtain a preliminary fused queue length. The initial fusion queue length is smoothed over time to obtain the vehicle queue length of the U-turn lane.

[0015] The above scheme dynamically integrates multi-source data and performs smoothing processing, which improves the accuracy and stability of queue length estimation.

[0016] To improve the solution, this application also proposes a step for determining the dynamic weights of video image information and millimeter-wave radar detection information based on the effectiveness assessment results and historical effectiveness assessment results, including: Current effectiveness indicators for acquiring video image information and millimeter-wave radar detection information; Historical validity trend indicators for acquiring video image information and millimeter-wave radar detection information; Based on current effectiveness indicators and historical effectiveness trend indicators, calculate the comprehensive reliability score of video image information and millimeter-wave radar detection information; The dynamic weights of video image information and millimeter-wave radar detection information are determined based on the comprehensive reliability score.

[0017] The above approach dynamically adjusts data weights based on current and historical validity indicators, making the fusion results more adaptable and robust.

[0018] To improve the solution, this application also proposes a step of performing time-series smoothing on the initial fused queue length to obtain the vehicle queue length of the U-turn lane, including: The exponential smoothing method is used to process the initial fusion queue length to obtain the vehicle queue length of the U-turn lane.

[0019] The above scheme employs exponential smoothing to effectively filter out data noise, making queue length estimation more stable and reliable.

[0020] To improve the solution, this application also proposes a step of performing time-series smoothing on the initial fused queue length to obtain the vehicle queue length of the U-turn lane, including: Obtain historical data on the initial fusion queue length; The historical data of the initial fusion queue length are processed by weighted average to obtain the vehicle queue length of the U-turn lane.

[0021] The above scheme uses weighted averaging to process historical data, further smoothing the queue length estimate and improving its stability.

[0022] To improve the solution, this application also proposes that, when the type of ground traffic congestion is a temporary stop interference state for shuttle vehicles, the steps for scheduling the reduction of the green light in the affected lane and the extension of the green light in the unaffected key lane include: When the ground traffic congestion status is a temporary stop interference status for shuttle vehicles, real-time traffic flow information in each direction within the intersection of the rail transit station is obtained. Based on real-time traffic flow information, the impact of scheduling the reduction of green lights for affected lanes and the extension of green lights for unaffected key lanes on the traffic efficiency of the intersection is assessed, and the impact assessment results are obtained. Based on the impact assessment results, the duration of the green light in the affected lanes will be shortened and the duration of the green light in the unaffected key lanes will be extended. The scheduling is executed according to the adjusted compression and extension durations.

[0023] The above-mentioned solutions address the disruption caused by temporary stops of shuttle vehicles by assessing the impact and dynamically adjusting green light durations, thereby effectively easing traffic congestion and reducing interference.

[0024] An intersection signal control system for performing intersection signal control includes: The abnormal information acquisition module is used to acquire abnormal information about rail transit operation; The congestion type determination module is used to sense and analyze abnormal information to determine the type of ground traffic congestion. The signal dispatch control module is used to execute corresponding signal dispatch strategies based on the type of ground traffic congestion. The signal dispatch strategies include: if the ground traffic congestion is a bus congestion, the downstream channel will be diverted and buses will be released in a concentrated manner; if the ground traffic congestion is a temporary stop interference of shuttle vehicles, the green light of the affected lane will be shortened and the green light of the unaffected key lane will be extended.

[0025] The above scheme provides a system implementation plan that can effectively execute the above signal control method and realize intelligent traffic management.

[0026] In summary, the intersection signal control method and system provided in this application can effectively alleviate traffic congestion by acquiring abnormal information about rail transit, intelligently identifying the type of ground traffic congestion, and performing targeted scheduling. Thus, it has the advantage of effectively dealing with ground traffic congestion caused by abnormal operation of rail transit and improving the traffic efficiency of intersections. Attached Figure Description

[0027] Figure 1 This is a flowchart of a method for intersection signal control in one embodiment of the present invention; Figure 2 This is one of the flowcharts of an intersection signal control method according to another embodiment of the present invention; Figure 3 This is a second flowchart of a method for intersection signal control according to another embodiment of the present invention; Figure 4 This is a third flowchart of a method for intersection signal control according to another embodiment of the present invention; Figure 5 This is the fourth flowchart of a method for intersection signal control in another embodiment of the present invention; Figure 6 This is the fifth flowchart of a method for intersection signal control in another embodiment of the present invention; Figure 7 This is a flowchart of a method for intersection signal control according to another embodiment of the present invention, number six. Figure 8This is the seventh flowchart of a method for intersection signal control in another embodiment of the present invention; Figure 9 This is a system block diagram of an intersection signal control system according to another embodiment of the present invention; Explanation of reference numerals in the attached figures: 1. Intersection signal control system; 11. Abnormal information acquisition module; 12. Delay type determination module; 13. Signal dispatch control module. Detailed Implementation

[0028] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0029] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0030] Traditional intersection signal control methods suffer from the inability to accurately determine the type of traffic congestion when sudden operational anomalies occur in the rail transit system. This limitation prevents the system from implementing effective signal scheduling strategies based on the instantaneous and unexpected superposition of changes in actual traffic flow, thereby exacerbating intersection traffic congestion and reducing the overall traffic efficiency of the intersection.

[0031] For example, suppose in a core urban area, at an intersection near a rail transit station, when a power outage causes trains to stop, forcing a large number of passengers to turn to surface buses and shuttle vehicles, the existing signal control system can only passively adjust based on local flow sensor data. The system cannot identify whether this congestion is caused by a specific type of rail transit anomaly, such as queue overflow caused by buses stopping for extended periods at platforms, or lane obstruction caused by ride-hailing vehicles and taxis temporarily stopping on the roadside. This lack of information prevents the system from distinguishing the root causes and scope of impact of different congestion patterns, thus hindering the implementation of targeted traffic management measures, resulting in the intersection facing disorderly traffic flow overlap and conflicts in a short period of time.

[0032] In response, this application proposes an intersection signal control method, combining... Figure 1 As shown, it includes: S1, obtain abnormal information about rail transit operation; S2, based on abnormal information, senses and analyzes to determine the type of ground traffic congestion; S3, based on the type of ground traffic congestion, execute the corresponding signal dispatching strategy; the signal dispatching strategy includes: if the type of ground traffic congestion is bus congestion, execute the downstream channel diversion and centralized release of buses; and if the type of ground traffic congestion is temporary stop interference of shuttle vehicles, execute the dispatching to shorten the green light of the affected lane and extend the green light of the unaffected key lane.

[0033] Abnormal information in rail transit operation refers to event data that occurs outside of normal rail transit system operation and affects passenger travel patterns and ground traffic conditions. This information can be obtained in various forms, such as train delay, cancellation, and malfunction reports obtained through the rail transit operation data center interface, or data such as platform passenger flow concentration and increased exit gate traffic monitored by sensor networks. The purpose is to provide data input for subsequent assessment of ground traffic conditions. Sensing and analyzing abnormal information to determine the type of ground traffic congestion involves using data processing and pattern recognition technologies based on the acquired rail transit abnormal information to identify the current congestion or congestion pattern faced by ground traffic in the intersection area. This can be achieved using machine learning algorithms, expert system rule bases, or threshold-based judgment logic. For example, by analyzing the scale of passenger outflow, the impact on bus routes, and changes in the demand for connecting vehicles in the abnormal information, it can be determined whether the congestion is in the bus station area or a local lane disturbance caused by temporary stops of connecting vehicles. The purpose is to transform rail transit abnormalities into ground traffic problem types to facilitate scheduling measures. The signal scheduling strategy refers to the traffic signal timing adjustment schemes that are preset or dynamically generated for different types of ground traffic congestion. It can be implemented using a preset scheduling scheme library, real-time optimization algorithms, or rule-based decision-making systems. Its purpose is to optimize ground traffic flow by adjusting the green light duration and phase sequence of intersection traffic lights in the event of rail transit anomalies. In the case of a ground traffic congestion type of bus congestion, the downstream channel diversion and centralized release of buses scheduling strategy refers to a combined signal control strategy adopted by the system when it is determined that the main problem of ground traffic is the long-term stopping or queuing of buses at platforms or exit lanes. This can be achieved by activating a dedicated diversion phase for downstream exit lanes, combining periodic green light pulses with red light buffers, and centrally releasing bus convoys under specific conditions. Its purpose is to resolve the bottleneck in bus traffic, ensuring that buses can leave the platform area, thereby alleviating intersection congestion. Among them, the type of ground traffic congestion is the temporary parking interference state of shuttle vehicles. The scheduling of compressing the green light of the affected lane and extending the green light of the unaffected key lane refers to an asymmetric signal control strategy adopted by the system when it is determined that the main problem of ground traffic is the temporary parking of shuttle vehicles such as taxis or ride-hailing vehicles on the roadside, which leads to a decrease in the traffic capacity of some lanes. It can be achieved by real-time monitoring of traffic flow and queue length of the affected lane, assessing the impact on overall traffic efficiency, and dynamically adjusting the green light duration of the relevant lanes. Its purpose is to optimize the traffic capacity of the unaffected lanes by sacrificing the travel time of some affected lanes while ensuring overall traffic efficiency, thereby mitigating the impact of local interference on the intersection.

[0034] To further illustrate the implementation of this application, an example is provided below. In some embodiments, this application is specifically implemented as follows. First, obtaining abnormal information about rail transit operation can be achieved through real-time data exchange with the urban rail transit operation management center via a data interface. For example, data streams such as train timetables, real-time locations, fault alarms, and passenger congestion can be received via an API interface. These data streams can be received and pre-processed by a data acquisition and preprocessing module. Subsequently, the abnormal information is perceived and analyzed to determine the type of ground traffic congestion, which can be accomplished by an analysis module. This module can have a pre-set rule-based system or a trained model. For example, when receiving abnormal information such as "a train on a certain line is suspended, and passengers are expected to transfer to buses," the analysis module will combine historical data and a real-time passenger flow prediction model to determine that the current ground traffic congestion type is "bus congestion." If abnormal information such as "passengers are congested at a certain platform, and shuttle vehicles are flooding into the surrounding area" is received, it may be determined as "shuttle vehicle temporary stop interference." Once the type of ground traffic congestion is determined, the signal dispatch control module will execute the corresponding signal dispatch strategy according to the pre-set dispatch logic. Specifically, when a bus is determined to be stuck, the signal dispatch control module sends instructions to the intersection signal controller to activate the diversion phase for the downstream exit lanes and coordinate with a green light timing scheme for centralized bus passage. For example, the green light time for dedicated bus lanes can be temporarily extended, and signals at adjacent intersections can be coordinated to ensure buses can pass. When a temporary stop by shuttle vehicles is determined to be disruptive, the signal dispatch control module calculates real-time traffic flow data for the affected lanes and unaffected key lanes, and dynamically adjusts the green light duration according to preset optimization goals. For example, the green light time for the affected lanes can be shortened, and this time can be allocated to straight or left-turn lanes to maintain overall traffic flow. These dispatch instructions are sent to the on-site signal controller via wired or wireless communication networks, and the signal controller executes the specific light color changes.

[0035] Optional, combined Figure 2 As shown, the type of ground traffic congestion in step S3 is bus congestion. The steps for implementing downstream channel diversion and centralized release of buses include: A1, when the type of ground traffic congestion is bus congestion, the downstream exit lane diversion phase is activated; A2, in the downstream exit lane guidance phase, periodically executes green light pulses and red light buffers; A3, during the red light buffer period, monitor the length of the vehicle queue in the U-turn lane area; A4. Adjust the green light pulse according to the length of the vehicle queue to adjust the release rhythm of downstream vehicles; A5. When the downstream exit main lane is cleared and the vehicle queue length has not reached the preset overflow warning threshold, the downstream exit lane diversion phase ends. A6, after the downstream exit lane traffic control phase ends, the centralized release phase for buses begins.

[0036] The vehicle queue length refers to the length or number of vehicles waiting to pass in a specific lane or area, which can be achieved using technologies such as visual recognition, radar detection, or geomagnetic induction. The overflow warning threshold is a pre-set critical value used to determine whether the vehicle queue is about to reach saturation or overflow. Specifically, it can be a number of vehicles, a distance, or a time period, with the aim of preventing downstream traffic congestion from spreading to upstream areas.

[0037] In some preferred embodiments, this application is implemented as follows. When the intersection signal control system identifies a bus congestion status through its congestion type determination module, the signal dispatch control module immediately sends a command to the traffic light controller to initiate a downstream exit lane diversion phase. During this diversion phase, the traffic light controller alternates between green light pulses and red light buffers according to a preset cycle, such as every 30 seconds. The duration of the green light pulse can be set to 5 to 10 seconds, allowing some vehicles to pass; the duration of the red light buffer can be set to 20 to 25 seconds, providing time for downstream vehicles to disperse. During the red light buffer, the system uses video detectors and geomagnetic sensors installed in the U-turn lane area to acquire real-time vehicle information and calculate the vehicle queue length. For example, the number of vehicles can be identified through image processing algorithms, or the presence of vehicles can be detected and accumulated using geomagnetic sensors. The signal dispatch control module dynamically adjusts the duration of subsequent green light pulses based on the real-time monitored vehicle queue length. For example, if the vehicle queue length exceeds a certain intermediate threshold, the green light pulse duration can be shortened by 2 seconds; if the queue length continues to increase and approaches the overflow warning threshold, the green light pulse can be further shortened or paused. When the system confirms through the vehicle detectors in the downstream exit main lane that there are no vehicles in that lane, and the vehicle queue length in the U-turn lane area is lower than the preset overflow warning threshold (e.g., queue length is less than 5 vehicles or the queue area corresponding to the queue length does not exceed the U-turn lane range), the signal dispatch control module will immediately end the downstream exit lane guidance phase. Immediately afterwards, the system will initiate the centralized bus release phase. At this time, the traffic lights related to the bus travel direction will remain green for an extended period, such as 30 to 60 seconds, to ensure that the stranded buses can pass through the intersection quickly and in a concentrated manner.

[0038] Optional, combined Figure 3 As shown, the steps for monitoring the vehicle queue length in the U-turn lane area during the red light buffer period include: A31, acquire video image information of the U-turn lane area; A32, acquiring millimeter-wave radar detection information in the U-turn lane area; A33, based on video image information, identifies vehicle targets and calculates the length of the first vehicle queue; A34, based on millimeter-wave radar detection information, identifies vehicle targets and calculates the length of the second vehicle platoon; A35, based on video image information and millimeter-wave radar detection information, evaluates the effectiveness of the video image information and millimeter-wave radar detection information, and obtains the effectiveness evaluation results; A36. Based on the effectiveness assessment results, the vehicle queue length of the U-turn lane is obtained by merging the first vehicle queue length and the second vehicle queue length.

[0039] The process involves several key steps: First, acquiring video image information of the U-turn lane area. This involves using video surveillance equipment installed above the U-turn lane area to collect real-time visual data streams of the area. The purpose is to provide visual characteristic information about vehicles, such as their shape, color, and position. Second, acquiring millimeter-wave radar detection information of the U-turn lane area involves using millimeter-wave radar sensors deployed in the U-turn lane area to acquire real-time radar detection data of the area. The purpose is to provide physical attribute information about vehicles, such as their distance, speed, and presence. Third, identifying vehicle targets and calculating the initial vehicle queue length based on the video image information involves using image processing and computer vision algorithms to analyze the video image information, automatically detect and locate individual vehicles in the image, and determine the vehicle queue length from the video perspective based on these identified vehicle targets through counting or spatial arrangement analysis. The purpose is to obtain a preliminary estimate of the vehicle queue from a visual perspective. Based on millimeter-wave radar detection information, identifying vehicle targets and calculating the second vehicle queue length involves using radar signal processing algorithms to analyze the millimeter-wave radar detection information, identify vehicle targets within the radar beam coverage area, and calculate the radar-detected vehicle queue length based on the distance and number of these targets. The purpose is to obtain a preliminary estimate of the vehicle queue from the radar detection perspective. Evaluating the effectiveness of video image information and millimeter-wave radar detection information, and obtaining an effectiveness evaluation result, involves analyzing the quality and reliability of sensor data from two different sources to determine their usability and accuracy under current environmental conditions. Specifically, this can involve a comprehensive judgment of image clarity, lighting conditions, radar signal strength, noise level, and consistency between the two information sources. The purpose is to provide a reliable basis for subsequent information fusion and ensure the accuracy of the fusion result. Fusion of the first and second vehicle queue lengths to obtain the vehicle queue length of the U-turn lane involves using appropriate algorithms or strategies, based on the effectiveness evaluation results, to comprehensively process the vehicle queue lengths calculated separately from the video images and millimeter-wave radar to obtain a more accurate and robust final vehicle queue length. Its purpose is to overcome the limitations of a single sensor and improve measurement accuracy and reliability by complementing information from multiple sources.

[0040] In some preferred embodiments, this application is implemented as follows. To accurately monitor the vehicle queue length in the U-turn lane during the red light buffer period, a high-resolution network camera can be deployed to acquire video image information of the U-turn lane area, while a 77GHz vehicle-mounted millimeter-wave radar sensor is installed to acquire millimeter-wave radar detection information of the area. Specifically, after acquiring the video image information, a real-time target detection model based on deep learning, such as YOLOv5, can be used to identify vehicle targets in the video frames and calculate the first vehicle queue length. This may include performing perspective transformation correction on the identified vehicle bounding boxes, mapping them to the actual ground coordinate system, and then determining the queue length through counting and arrangement analysis. Simultaneously, for the millimeter-wave radar detection information, a clustering algorithm, such as DBSCAN, can be used to process the radar point cloud data, identify vehicle targets, and calculate the second vehicle queue length based on the distance and number of these targets. Further, to evaluate the effectiveness of the video image information and the millimeter-wave radar detection information, image quality analysis can be performed on the video images, such as calculating image sharpness, brightness, and contrast, and setting corresponding quality thresholds. Simultaneously, signal feature analysis can be performed on millimeter-wave radar detection information, such as evaluating the signal-to-noise ratio of radar echoes and the density of target point clouds. Furthermore, the consistency between vehicle targets identified in video images and those detected by millimeter-wave radar can be compared, for example, by matching vehicle positions and numbers. Based on these analysis results, a comprehensive effectiveness evaluation result can be obtained, for example, by using a multi-factor weighted model or fuzzy logic system to determine the reliability of the two information sources. Finally, based on the effectiveness evaluation result, the first vehicle queue length and the second vehicle queue length can be dynamically fused. This can include determining the dynamic weights of video image information and millimeter-wave radar detection information based on the current evaluation result and historical effectiveness evaluation results within a preset time period. For example, when the quality of the video image deteriorates, its weight can be reduced while the weight of the millimeter-wave radar is increased. Then, based on these dynamic weights, the first vehicle queue length and the second vehicle queue length are weighted and fused to obtain a preliminary fused queue length. To further improve the stability of the queue length, the preliminary fused queue length can be subjected to time-series smoothing, for example, using exponential smoothing, to obtain the final vehicle queue length of the U-turn lane. This processing effectively filters out transient noise and measurement fluctuations, providing smoother and more reliable queue length data.

[0041] Optional, combined Figure 4 As shown, the A35 evaluates the effectiveness of video image information and millimeter-wave radar detection information based on the latter, and the steps to obtain the effectiveness evaluation results include: A351 performs image quality analysis on video image information and obtains the image quality analysis results. A352 performs signal feature analysis on millimeter-wave radar detection information and obtains signal feature analysis results; A353, compare the consistency between the vehicle target in the video image information and the vehicle target in the millimeter-wave radar detection information to obtain the consistency result; A354, based on the image quality analysis results, signal feature analysis results, and consistency results, judges the validity of video image information and millimeter-wave radar detection information, and obtains the validity assessment results.

[0042] Image quality analysis refers to the evaluation of the visual attributes of video image data. Specifically, it can be achieved by analyzing indicators such as image sharpness, brightness, contrast, noise level, and whether there is occlusion or blurring. Its purpose is to determine whether the video image information is affected by environmental factors such as lighting, rain, fog, or sensor failure, thereby evaluating its reliability for vehicle target recognition.

[0043] Signal feature analysis refers to the evaluation of the characteristics of the raw or processed signals detected by millimeter-wave radar. Specifically, it can be achieved by analyzing indicators such as signal strength, signal-to-noise ratio, Doppler frequency shift, target cross section (RCS), and whether there are false targets or clutter interference. Its purpose is to determine whether the millimeter-wave radar detection information is affected by environmental interference, multipath effect, or sensor performance degradation, thereby evaluating its reliability for vehicle target identification.

[0044] Consistency refers to the degree of matching between the vehicle targets identified in video image information and the vehicle targets identified in millimeter-wave radar detection information in terms of spatial location, motion state, or number. Specifically, it can be achieved by aligning the data from the two sensors in time and space and comparing whether the position coordinates, velocity vectors, size estimates, and number of targets identified by each sensor are mutually consistent. The purpose is to verify whether the information from two different sources points to the same physical entity, thereby determining whether there are obvious deviations or errors between them.

[0045] The effectiveness evaluation result refers to a comprehensive judgment on whether the video image information and millimeter-wave radar detection information are reliable and suitable for subsequent vehicle queue length estimation. Specifically, it can be achieved by setting a threshold or adopting logical rules, combining image quality analysis results, signal feature analysis results, and consistency results, and outputting a quantitative score or classification label representing the effectiveness of the information. Its purpose is to provide a reliable basis for subsequent fusion of data from different sensors and avoid introducing low-quality or inconsistent data into the final queue length calculation.

[0046] In some preferred embodiments, this application is implemented as follows: When performing image quality analysis on video image information, image processing algorithms can be used, such as calculating the image's sharpness score (e.g., gradient magnitude based on the Laplacian operator), average brightness, contrast (e.g., grayscale range based on histogram equalization), and noise level (e.g., residual analysis after Gaussian filtering). These indicators can be quantified and compared with preset thresholds. For example, if the sharpness score is below a certain threshold, or the brightness is too high or too low, the image quality analysis result can indicate poor video image information quality.

[0047] When performing signal feature analysis on millimeter-wave radar detection information, the signal strength, signal-to-noise ratio (SNR), and stability of Doppler frequency shift for each detected target can be extracted. For example, if the signal strength of a target is lower than the background noise level, or its SNR is lower than a preset threshold, or its Doppler frequency shift changes in a short period of time, the signal feature analysis results can indicate that the millimeter-wave radar detection information is subject to interference or instability.

[0048] When comparing the consistency between vehicle targets identified in video images and those detected by millimeter-wave radar, spatiotemporal registration can be performed first. For example, the pixel coordinates of the vehicle targets in the video image can be converted to actual geographic coordinates and matched with the geographic coordinates of the targets detected by the millimeter-wave radar. After successful matching, the consistency of the matched targets' size, velocity vector, and target ID can be compared. For instance, if a large vehicle identified in the video corresponds to a small target in the radar data, or if their velocities are opposite, or if multiple vehicles are identified in the video but only one is detected by the radar in a certain area, the consistency result can indicate a discrepancy between the two.

[0049] Finally, when judging the validity of video image information and millimeter-wave radar detection information based on image quality analysis results, signal feature analysis results, and consistency results, a multi-factor weighted decision model or a rule-based expert system can be used. For example, different weights can be assigned to image quality, signal features, and consistency, and then a comprehensive validity score can be calculated. If the comprehensive score is higher than a certain validity threshold, the information is judged to be valid; otherwise, it is judged to be invalid or low-validity. For example, when the image quality analysis results show that the image is clear, the signal feature analysis results show that the signal is stable, and the consistency results show that the matching degree is high, both types of information can be judged to be highly valid; conversely, if any one or more indicators are poor, the validity evaluation result should be reduced accordingly.

[0050] Optional, combined Figure 5As shown, the steps by which A36, based on the effectiveness assessment results, integrates the first vehicle queue length and the second vehicle queue length to obtain the vehicle queue length for the U-turn lane include: A361, based on the effectiveness assessment results, obtains the historical effectiveness assessment results of video image information and millimeter-wave radar detection information within a preset time period; A362 determines the dynamic weights of video image information and millimeter-wave radar detection information based on the effectiveness assessment results and historical effectiveness assessment results; A363, based on dynamic weights, performs a weighted fusion of the lengths of the first and second vehicle queues to obtain a preliminary fused queue length; A364, initially merge queue lengths and perform time series smoothing to obtain the vehicle queue length of the U-turn lane.

[0051] Time series smoothing refers to processing a series of data arranged in chronological order to eliminate short-term fluctuations or noise, reveal potential trends or more stable values. It can be achieved using various algorithms such as moving average, exponential smoothing, and Kalman filtering. Its purpose is to improve the stability and reliability of the data series and reduce the impact of instantaneous errors on subsequent decisions.

[0052] In some preferred embodiments, this application is implemented as follows: After obtaining the validity assessment results, the system maintains a historical data buffer to store the validity assessment results of video image information and millimeter-wave radar detection information per second over the past five minutes, for example. When a new validity assessment result is generated, the system extracts the historical validity assessment results within a preset time period from this buffer. Next, to determine the dynamic weights of the video image information and millimeter-wave radar detection information, the system can calculate a weighted average of the current validity assessment result and the historical validity assessment results. For example, assigning a weight of 0.6 to the current validity assessment result and a weight of 0.4 to the historical validity assessment result, thereby obtaining their respective comprehensive reliability scores. Then, based on these comprehensive reliability scores, the dynamic weights of the video image information and millimeter-wave radar detection information are calculated through normalization processing or the Softmax function, ensuring that the sum of their weights is 1. For example, if the comprehensive reliability score of the video image information is high, its dynamic weight can be set to 0.7, and the dynamic weight of the millimeter-wave radar detection information is 0.3. Subsequently, based on these dynamic weights, the first vehicle queue length identified through video image information and the second vehicle queue length identified through millimeter-wave radar detection information are weighted and fused. For example, the initial fused queue length = dynamic weight - video * first vehicle queue length + dynamic weight - radar * second vehicle queue length. Finally, to perform time-series smoothing on the initial fused queue length, the system can use exponential smoothing. Specifically, the new vehicle queue length of the U-turn lane can be obtained by weighted averaging the current initial fused queue length and the smoothed queue length from the previous moment. For example, the new queue length = α * current initial fused queue length + (1-α) * smoothed queue length from the previous moment, where α is a smoothing coefficient, for example, it can be set to 0.3, which can effectively eliminate instantaneous fluctuations and obtain a more stable vehicle queue length for the U-turn lane.

[0053] Optional, combined Figure 6 The steps by which A362 determines the dynamic weights of video image information and millimeter-wave radar detection information based on effectiveness assessment results and historical effectiveness assessment results include: A3621, a current validity indicator for acquiring video image information and millimeter-wave radar detection information; A3622, a historical effectiveness trend indicator for acquiring video image information and millimeter-wave radar detection information; A3623 calculates the comprehensive reliability score of video image information and millimeter-wave radar detection information based on current effectiveness indicators and historical effectiveness trend indicators. A3624 determines the dynamic weights of video image information and millimeter-wave radar detection information based on the comprehensive reliability score.

[0054] The current validity index refers to a quantitative value used to measure the quality and reliability of video image information and millimeter-wave radar detection information at the current moment. It can be characterized using parameters such as image sharpness, illumination intensity, signal-to-noise ratio, and target detection confidence, aiming to reflect the availability of sensor data in the immediate environment. The historical validity trend index refers to a statistical measure reflecting the changing patterns of the validity of video image information and millimeter-wave radar detection information over a past period. It can be calculated using the average, variance, rate of change, or time series model predictions of the historical validity index, aiming to assess the stability and reliability of sensor data in long-term operation. The comprehensive reliability score is a quantitative result that comprehensively evaluates the reliability of video image information and millimeter-wave radar detection information based on the current validity index and the historical validity trend index. It can be calculated using methods such as weighted averaging, fuzzy logic inference, and machine learning models, aiming to provide a single evaluation value that comprehensively reflects the quality of sensor data. Dynamic weights refer to the contribution ratios that are adjusted in real time based on the comprehensive reliability scores when fusing video image information and millimeter-wave radar detection information. They can be determined by normalization, Sigmoid function mapping, or rule-based lookup tables. The purpose is to ensure that sensor data with higher reliability occupies a larger proportion in the fusion result under different environmental conditions.

[0055] In some preferred embodiments, the step of determining the dynamic weights of video image information and millimeter-wave radar detection information based on the effectiveness evaluation results and historical effectiveness evaluation results can be specifically implemented as follows: First, obtain the current effectiveness indicators of the video image information and millimeter-wave radar detection information. For example, for video image information, its image clarity, brightness, contrast, and the vehicle target confidence output by the target detection algorithm can be analyzed as current effectiveness indicators. For millimeter-wave radar detection information, its signal-to-noise ratio, target distance accuracy, target speed accuracy, and target tracking stability can be obtained as current effectiveness indicators. These indicators can be calculated in real time or read directly from the sensor interface. Second, obtain the historical effectiveness trend indicators of the video image information and millimeter-wave radar detection information. This may involve maintaining a database of historical effectiveness indicators within a sliding time window. For example, the average and standard deviation of each current effectiveness indicator of the video image information and millimeter-wave radar detection information over the past five minutes can be calculated as its historical effectiveness trend indicators. In addition, the historical performance of each sensor effectiveness indicator under different weather or lighting conditions can be recorded and used as a trend reference. Then, based on the current effectiveness indicators and historical effectiveness trend indicators, calculate the comprehensive reliability score of the video image information and millimeter-wave radar detection information. Specifically, a comprehensive reliability function can be defined for each sensor, which weights and sums the current effectiveness index (e.g., weighted at 0.6) and the historical effectiveness trend index (e.g., weighted at 0.4). For example, a threshold can be set: if the current sharpness is below the threshold, the current effectiveness score decreases; if the historical average signal-to-noise ratio is consistently low, the historical trend score also decreases. Finally, these weighted scores are normalized to the range of 0 to 1 to obtain their respective comprehensive reliability scores. Finally, based on the comprehensive reliability scores, the dynamic weights of video image information and millimeter-wave radar detection information are determined. For example, a Softmax function or a simple proportional allocation method can be used. If the comprehensive reliability score of the video image information is 0.8 and the comprehensive reliability score of the millimeter-wave radar detection information is 0.6, then a total score of 1.4 can be calculated, and the weights of the video image information are set to 0.8 / 1.4, and the weights of the millimeter-wave radar detection information are set to 0.6 / 1.4. In this way, sensors with higher comprehensive reliability scores are assigned greater weights in subsequent fusion processes, thus ensuring the accuracy of the fusion results.

[0056] Optionally, A364 performs time-series smoothing on the initial fused queue length to obtain the vehicle queue length for the U-turn lane, including the following steps: The exponential smoothing method is used to process the initial fusion queue length to obtain the vehicle queue length of the U-turn lane.

[0057] Exponential smoothing is a time series forecasting method that assigns different weights to historical data, with more recent data receiving greater weights, in order to predict future values ​​or smooth current data. It can be implemented in various forms, such as single exponential smoothing, double exponential smoothing, or triple exponential smoothing. Its purpose is to effectively filter out random fluctuations and noise in the data while maintaining sensitivity to changes in data trends.

[0058] Optional, combined Figure 7 As shown, the steps of A364 to perform time-series smoothing on the initial fused queue length to obtain the vehicle queue length for the U-turn lane include: A3641, Obtain historical data on the initial fusion queue length; A3642, the historical data of the preliminary fusion queue length is processed by weighted average to obtain the vehicle queue length of the U-turn lane.

[0059] Weighted average processing involves assigning a weight to each data point in a set of data, multiplying each data point by its corresponding weight, summing all the products and dividing by the sum of all weights to obtain a weighted average. This can be achieved using various methods such as linear weighting, exponential decay weighting, and sliding window weighting. Its purpose is to more precisely reflect the degree of influence of different historical data points on the current state, thereby improving the accuracy of the estimation.

[0060] In some preferred embodiments, the historical data of the preliminary fused queue length is processed by a weighted average, specifically using an exponentially decaying weighted average. For example, the system can maintain a historical sequence of preliminary fused queue lengths, such as queue length data for the most recent N time steps. When calculating the current queue length of the U-turn lane, these historical data points are assigned a weight that decays over time; that is, data points closer to the current time have a larger weight, while data points farther away have a smaller weight. For example, a decay factor α (0 < α < 1) can be set, so the queue length at the current moment can be calculated by the weighted sum of historical data points, where the weight of each historical data point is a power of α. In this way, the system can ensure that the latest data has a greater impact on the result, while still utilizing earlier historical data to smooth out short-term fluctuations, thus obtaining a U-turn lane queue length that is both responsive to current changes and has good stability. This processing method can effectively reduce queue length estimation bias caused by instantaneous sensor errors or short-term fluctuations in traffic flow, enabling the signal dispatching system to make decisions based on more reliable data.

[0061] Optional, combined Figure 8As shown, the type of ground traffic congestion state in step S3 is a temporary stop interference state for shuttle vehicles. The steps of scheduling to shorten the green light of the interfered lane and extend the green light of the unaffected key lane include: B1, when the ground traffic congestion status is a temporary stop interference status for shuttle vehicles, obtain real-time traffic flow information in each direction within the intersection of the rail transit station; B2. Based on real-time traffic flow information, assess the impact of scheduling the reduction of green lights for affected lanes and the extension of green lights for unaffected key lanes on the intersection's traffic efficiency, and obtain the impact assessment results. B3. Based on the impact assessment results, adjust the shortened duration of the green light for the affected lanes and the extended duration of the green light for the unaffected key lanes. B4, execute the scheduling according to the adjusted compression duration and extension duration.

[0062] Real-time traffic flow information refers to the data set of vehicles operating in all directions within an intersection, acquired at a specific point in time through various traffic sensors or data sources. This data can include traffic volume, vehicle speed, queue length, lane occupancy, etc., and aims to provide an accurate and dynamic data foundation for subsequent scheduling decisions. Furthermore, evaluating the impact of shortening the green light of affected lanes and extending the green light of undisturbed key lanes on intersection efficiency involves using traffic simulation models, prediction algorithms, or expert systems, based on the acquired real-time traffic flow information, to quantitatively analyze the potential impact of different green light duration adjustment schemes on the overall intersection capacity, average delay, queue length, and other indicators. This results in a determination of the potential impact of each scheme on traffic efficiency, aiming to quantify the advantages and disadvantages of different scheduling schemes and provide a basis for optimization decisions. Based on this, adjusting the shortened green light duration of the affected lane and the extended green light duration of the unaffected key lane refers to dynamically calculating and determining the specific amount of time the affected lane's green light should be shortened and the specific amount of time the unaffected key lane's green light should be extended, based on the impact assessment results and using optimization algorithms, rule engines, or machine learning models. The aim is to alleviate local interference while maximizing the balance of the overall traffic flow at the intersection and avoiding the creation of new congestion points.

[0063] This application's solution achieves refined management of signal dispatching strategies by introducing the acquisition and analysis of real-time traffic flow information. When the system determines that the ground traffic congestion is due to temporary stopping interference from connecting vehicles, it no longer simply executes preset green light compression and extension, but first acquires real-time traffic flow information in each direction within the rail transit station intersection. This information, such as traffic volume, queue length, and vehicle speed, comprehensively reflects the current traffic operation status of the intersection. Based on this real-time data, the system can assess the impact of different combinations of green light compression and extension durations on the overall traffic efficiency of the intersection, thereby obtaining a quantitative impact assessment result. This assessment process considers the traffic load in each direction of the intersection, avoiding the situation of focusing only on local problems while ignoring overall efficiency. It is precisely because of this comprehensive assessment that the system can dynamically adjust the compression duration of the green light for the affected lanes and the extension duration of the green light for the unaffected key lanes based on the assessment results. This adjustment is no longer a fixed value, but is optimized based on real-time traffic conditions and predictions of the impact on overall efficiency, ensuring that while alleviating the interference of temporary stopping interference from connecting vehicles, the negative impact on traffic flow in other directions is minimized. Ultimately, the system executes scheduling according to the adjusted compression and extension durations, enabling signal timing to respond more flexibly and intelligently to sudden traffic events. Compared to basic solutions that only execute preset scheduling strategies based on the type of congestion, the proposed solution, after identifying the interference of temporary stops by shuttle vehicles, further introduces the perception of real-time traffic flow information, the assessment of the impact on scheduling, and the dynamic adjustment of green light durations. This in-depth analysis and optimization mechanism transforms the original strategy of "compressing the green light of the affected lane and extending the green light of the unaffected key lane" from extensive execution to refined management. In this way, not only can local congestion caused by temporary stops of shuttle vehicles be effectively alleviated, but secondary congestion caused by over-adjustment can also be avoided, thereby significantly improving the overall traffic efficiency and traffic flow balance of the intersection while solving specific traffic problems.

[0064] In some preferred embodiments, when the system identifies ground traffic congestion as a temporary stop interference from shuttle vehicles, a refined scheduling process can be initiated. Specifically, obtaining real-time traffic flow information in each direction within the intersection of the rail transit station can be achieved through various sensor devices deployed at the intersection's entrance and exit lanes, such as loop detectors, video detectors, and millimeter-wave radar, to collect data on traffic flow, average vehicle speed, queue length, and lane occupancy rates in real time. This data is then aggregated and processed by the central control unit. Based on this real-time traffic flow information, the impact of shortening the green light for the affected lanes and extending the green light for the unaffected key lanes on the intersection's traffic efficiency can be evaluated. The impact assessment can be obtained by establishing a microscopic traffic simulation model, such as VISSIM or SUMO, using real-time traffic flow data as input to simulate the intersection's operation over a future period under different green light shortening and extension schemes, including indicators such as average delay, saturation, and queue overflow risk in each direction, thereby quantifying the impact of each scheme on overall traffic efficiency. For example, the system can simulate compressing the green light of the affected lane by 5, 10, or 15 seconds, while simultaneously extending the green light of the unaffected critical lane by the corresponding duration. The impact of these schemes on the total intersection delay can then be compared. Based on this, and according to the impact assessment results, adjusting the compressed green light duration of the affected lane and the extended green light duration of the unaffected critical lane can be achieved using an optimization algorithm, such as a genetic algorithm or particle swarm optimization algorithm, with the objective function of minimizing the total intersection delay or maximizing traffic capacity. The optimal green light compression and extension durations can be dynamically calculated based on the assessment results. For instance, if the assessment results show that compressing the green light of the affected lane by 10 seconds and extending the green light of the critical lane by 10 seconds achieves the best overall traffic efficiency, the system will determine to adopt this scheme. Finally, scheduling can be performed according to the adjusted compression and extension durations. This can be done through the traffic signal control interface, sending the calculated new green light duration parameters to the intersection traffic signal controller. The traffic signal controller then switches the traffic lights according to the new timing scheme, thereby achieving dynamic management of traffic flow.

[0065] An intersection signal control system for performing intersection signal control, combined with Figure 9 As shown, the intersection signal control system 1 includes: The abnormal information acquisition module 11 is used to acquire abnormal information of rail transit operation; The congestion type determination module 12 is used to perceive and analyze abnormal information to determine the type of ground traffic congestion status; The signal dispatch control module 13 is used to execute corresponding signal dispatch strategies according to the type of ground traffic congestion. The signal dispatch strategies include: if the type of ground traffic congestion is bus congestion, the downstream channel will be diverted and buses will be released in a concentrated manner; if the type of ground traffic congestion is temporary parking interference of shuttle vehicles, the green light of the affected lane will be shortened and the green light of the unaffected key lane will be extended.

[0066] The abnormal information acquisition module is a unit used to collect and process rail transit operation status data. Specifically, it can be a data interface or a sensor network, and its purpose is to provide real-time and accurate raw data for subsequent traffic status analysis. The congestion type determination module is a unit used to perform in-depth analysis and pattern recognition on the acquired abnormal information to determine the specific nature of the current ground traffic congestion or congestion. Specifically, it can be a processor with built-in machine learning algorithms or expert rule bases, and its purpose is to transform the raw abnormal information into specific traffic status types that can be used for signal dispatching decisions. The signal dispatching control module is a unit that generates and executes corresponding traffic signal timing schemes based on the determined ground traffic congestion status type. Specifically, it can be a programmable logic controller or a traffic signal controller, and its purpose is to optimize the traffic efficiency of intersections and alleviate traffic pressure by dynamically adjusting the green light duration and phase sequence of traffic lights.

[0067] This application's solution, through a modular design, achieves intelligent control of intersection signals, thereby addressing potential traffic problems arising from rail transit anomalies. Specifically, the anomaly information acquisition module, as the system's input, is responsible for acquiring real-time anomaly information regarding rail transit operations, such as train delays or cancellations, providing foundational data for subsequent decision-making. This anomaly information is then transmitted to the congestion type determination module, which performs in-depth perception and analysis of the received anomaly information to identify the specific congestion state of the current ground traffic, such as bus congestion or temporary stop interference from shuttle vehicles. This determination result is crucial for the system's precise scheduling. Finally, the signal scheduling control module selects and executes the corresponding signal scheduling strategy based on the output of the congestion type determination module. For example, when a bus congestion state is determined, the system will implement downstream channel diversion and centralized release of buses; when a temporary stop interference from shuttle vehicles is determined, the system will implement shortening of the green light for the affected lanes and extending the green light for the unaffected key lanes. This collaborative working mechanism enables the system to dynamically adjust signal timing according to actual conditions, optimizing intersection traffic efficiency. This system serves as a vehicle for implementing the aforementioned intersection signal control method, enabling the method to be executed efficiently and automatically. By concretizing the method steps into operable system modules, the system of this application can respond to rail traffic anomalies in real time, transforming abstract scheduling logic into actual signal control commands. This overcomes the limitations of relying solely on the method in terms of execution efficiency and real-time performance, ensuring the smooth operation of intersection traffic flow during sudden traffic events and preventing a decline in overall intersection traffic efficiency or even paralysis caused by abnormal surges or congestion of single-mode traffic flow.

[0068] In some preferred embodiments, the intersection signal control system can be specifically implemented as follows: The anomaly information acquisition module can be a server connected to the rail transit operation data center via an API interface. This server is configured with a data acquisition program capable of receiving and parsing anomaly information such as train operation status and fault reports from the data center in real time. The congestion type determination module can be a high-performance edge computing unit pre-installed with a deep learning-based traffic flow pattern recognition model. This model, by analyzing anomaly information and real-time traffic data from roadside sensors (such as video detectors and geomagnetic coils), can accurately identify various ground traffic congestion states, such as bus congestion and temporary stop interference from shuttle vehicles. The signal dispatch control module can be an intelligent traffic signal controller with a built-in library of various preset signal dispatch strategies, and is connected to the congestion type determination module via Ethernet. When the congestion type determination module outputs a specific congestion status type, the signal dispatch control module can immediately retrieve and execute the corresponding signal dispatch strategy from the strategy library. For example, in the case of a bus congestion, the signal controller can automatically adjust the green light phase of the downstream exit lane and coordinate with the dedicated phase for centralized release of buses. In the case of temporary stop interference from connecting vehicles, the signal controller can dynamically compress the green light duration of the affected lanes while extending the green light duration of the unaffected key lanes to ensure the optimization of the overall traffic efficiency of the intersection. The entire system is managed and monitored through an integrated software platform, enabling automated and intelligent control of intersection signals.

[0069] Through the above technical solution, this application provides an intersection signal control system. This system can effectively acquire abnormal information about rail transit operations and accurately determine the type of ground traffic congestion based on this information. By implementing customized signal scheduling strategies according to different congestion types, such as scheduling downstream lane diversion and centralized release of buses for bus congestion, and scheduling shortening the green light of affected lanes and extending the green light of unaffected key lanes for temporary stop interference from connecting vehicles, the system can achieve intelligent and dynamic control of intersection signals. This enables the system to respond promptly and accurately to changes in ground traffic flow caused by rail transit anomalies, effectively alleviating traffic congestion and avoiding a decline in overall intersection efficiency caused by abnormal surges or stagnation of a single traffic mode, thereby improving intersection capacity and traffic management efficiency.

[0070] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for controlling intersection signals, characterized in that, include: Obtain abnormal information about rail transit operations; Based on the abnormal information, the type of ground traffic congestion is determined; The corresponding signal dispatching strategy is executed according to the type of ground traffic congestion. The signal dispatching strategy includes: when the ground traffic congestion status is a bus congestion status, the downstream channel is diverted and the buses are released in a centralized manner. The type of ground traffic congestion is temporary parking interference for shuttle vehicles, which involves shortening the green light of the affected lanes and extending the green light of the unaffected key lanes.

2. The intersection signal control method according to claim 1, characterized in that, The type of ground traffic congestion is a bus congestion. The steps for implementing downstream channel diversion and centralized release of buses include: When the type of ground traffic congestion is bus congestion, the downstream exit lane diversion phase is activated. During the downstream exit lane traffic management phase, green light pulses and red light buffers are periodically executed; During the red light buffer period, monitor the length of the vehicle queue in the U-turn lane area; The green light pulse is adjusted according to the length of the vehicle queue, thereby adjusting the release rhythm of downstream vehicles; When the downstream exit main lane is cleared and the length of the vehicle queue has not reached the preset overflow warning threshold, the downstream exit lane diversion phase ends. After the downstream exit lane traffic control phase ends, the centralized release phase for buses is initiated.

3. The intersection signal control method according to claim 2, characterized in that, The step of monitoring the vehicle queue length in the U-turn lane area during the red light buffer period includes: Acquire video image information of the U-turn lane area; Acquire millimeter-wave radar detection information in the U-turn lane area; Based on the video image information, identify vehicle targets and calculate the length of the first vehicle queue; Based on the millimeter-wave radar detection information, vehicle targets are identified and the length of the second vehicle platoon is calculated; Based on the video image information and the millimeter-wave radar detection information, the effectiveness of the video image information and the millimeter-wave radar detection information is evaluated, and an effectiveness evaluation result is obtained; Based on the effectiveness assessment results, the vehicle queue length of the U-turn lane is obtained by merging the first vehicle queue length and the second vehicle queue length.

4. The intersection signal control method according to claim 3, characterized in that, The step of evaluating the effectiveness of the video image information and the millimeter-wave radar detection information based on the video image information and the millimeter-wave radar detection information, and obtaining the effectiveness evaluation result, includes: The video image information is subjected to image quality analysis to obtain the image quality analysis results; Signal feature analysis is performed on the millimeter-wave radar detection information to obtain signal feature analysis results; The consistency between the vehicle target in the video image information and the vehicle target in the millimeter-wave radar detection information is compared to obtain a consistency result; Based on the image quality analysis results, the signal feature analysis results, and the consistency results, the validity of the video image information and the millimeter-wave radar detection information is determined, and a validity evaluation result is obtained.

5. The intersection signal control method according to claim 3, characterized in that, The step of obtaining the vehicle queue length of the U-turn lane by fusing the first vehicle queue length and the second vehicle queue length based on the effectiveness evaluation results includes: Based on the effectiveness assessment results, obtain the historical effectiveness assessment results of the video image information and the millimeter-wave radar detection information within a preset time period; Based on the effectiveness assessment results and historical effectiveness assessment results, the dynamic weights of the video image information and the millimeter-wave radar detection information are determined. Based on the dynamic weights, the lengths of the first vehicle queue and the second vehicle queue are weighted and fused to obtain a preliminary fused queue length. The vehicle queue length of the U-turn lane is obtained by performing time series smoothing on the preliminary fusion queue length.

6. The intersection signal control method according to claim 5, characterized in that, The step of determining the dynamic weights of the video image information and the millimeter-wave radar detection information based on the effectiveness evaluation results and historical effectiveness evaluation results includes: Obtain the current validity indicators of the video image information and the millimeter-wave radar detection information; Obtain historical validity trend indicators for the video image information and the millimeter-wave radar detection information; Based on current validity indicators and historical validity trend indicators, calculate the comprehensive reliability score of the video image information and the millimeter-wave radar detection information; The dynamic weights of the video image information and the millimeter-wave radar detection information are determined based on the comprehensive reliability score.

7. The intersection signal control method according to claim 5, characterized in that, The step of performing time-series smoothing on the preliminary fused queue length to obtain the vehicle queue length of the U-turn lane includes: The vehicle queue length of the U-turn lane is obtained by processing the preliminary fusion queue length using the exponential smoothing method.

8. The intersection signal control method according to claim 5, characterized in that, The step of performing time-series smoothing on the preliminary fused queue length to obtain the vehicle queue length of the U-turn lane includes: Obtain historical data on the length of the initial fusion queue; The historical data of the preliminary fusion queue length are processed by weighted average to obtain the vehicle queue length of the U-turn lane.

9. The intersection signal control method according to claim 1, characterized in that, The type of ground traffic congestion is a temporary stop interference state for shuttle vehicles. The steps for scheduling the reduction of the green light in the affected lane and the extension of the green light in the unaffected key lane include: When the ground traffic congestion status is a temporary stop interference status for shuttle vehicles, real-time traffic flow information in each direction within the intersection of the rail transit station is obtained. Based on the real-time traffic flow information, the impact of scheduling the reduction of green lights for affected lanes and the extension of green lights for unaffected key lanes on the intersection's traffic efficiency is assessed, and the impact assessment results are obtained. Based on the impact assessment results, the duration of the green light in the affected lanes will be shortened and the duration of the green light in the unaffected key lanes will be extended. The scheduling is executed according to the adjusted compression and extension durations.

10. An intersection signal control system for performing intersection signal control, characterized in that, include: The abnormal information acquisition module is used to acquire abnormal information about rail transit operation; The congestion type determination module is used to sense and analyze abnormal information to determine the type of ground traffic congestion. The signal dispatch control module is used to execute corresponding signal dispatch strategies based on the type of ground traffic congestion. The signal dispatching strategy includes: when the ground traffic congestion status is a bus congestion status, the downstream channel is diverted and the buses are released in a centralized manner. The type of ground traffic congestion is temporary parking interference for shuttle vehicles, which involves shortening the green light of the affected lanes and extending the green light of the unaffected key lanes.

Images