Intelligent traffic signal lamp parameter configuration method and system

Abstract

The invention relates to the technical field of traffic control, in particular to an intelligent traffic signal lamp parameter configuration method and system, and the method comprises the following steps: obtaining a vehicle passing behavior, extracting a transverse offset and longitudinal delay time sequence before and after a steering action, recognizing a signal behavior delay section, and extracting the waiting time length and the release completion time point of a vehicle at the tail of a channel. And calculating a time difference, detecting a longitudinal distance of vehicles in a queue in a real-time period, updating a tail-section control period, and obtaining a tail-section green light continuous regulation result. According to the invention, through careful identification and analysis of vehicle behaviors and position features, optimization of signal scheduling, dynamic adjustment of a signal lamp period through real-time data, reduction of traffic congestion caused by untimely response, calculation of waiting duration and release time of tail vehicles, and allocation of a variable green light time period for each channel, the traffic congestion caused by untimely response can be reduced. The green light time of the tail section can be flexibly regulated and controlled according to actual requirements, traffic delay is further reduced, and the road utilization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and particularly to an intelligent traffic signal parameter configuration method and system. Background Art

[0002] The technical field of traffic control encompasses a series of methods and devices for solving traffic management and optimization problems. The core content is to improve road usage efficiency, reduce traffic congestion, and enhance traffic safety. Specifically, it involves the design, implementation, adjustment, and monitoring technologies of traffic signals, including signal control systems, vehicle detection systems, and traffic data analysis methods. This field also includes the study of traffic flow patterns and strategies for adjusting traffic flow at different times and locations through technical means.

[0003] Among them, the intelligent traffic signal parameter configuration method refers to the process of intelligently setting and adjusting the control parameters of traffic signals. This technical topic mainly involves how to adjust the green light duration and red light duration of traffic signals according to real-time traffic flow data to optimize traffic mobility and reduce traffic congestion. This method includes collecting traffic data, analyzing traffic data, and adjusting the parameters of signal operation according to the analysis results. This configuration method relies on traffic flow detection devices and data processing software to update the operation mode of traffic signals in real time to adapt to changing traffic conditions.

[0004] Existing traffic control technologies mainly rely on fixed signal cycles and static signal control modes, which cannot fully adapt to real-time road conditions changes in a dynamic traffic environment. Especially during traffic peaks or emergencies, fixed signal settings are difficult to cope with instantaneous traffic flow increases or decreases, easily causing traffic congestion and resource waste. Existing methods also show limitations in the identification and analysis of vehicle behaviors, lacking a detailed judgment of the impact of specific behaviors. For example, the identification of lateral offset and longitudinal delay is insufficient, resulting in the signal control being unable to accurately reflect the actual passing state of vehicles and affecting the smoothness of traffic flow. The problem not only reduces the efficiency of traffic signals but also increases the risks of road usage. Especially in the management of complex intersections, the lack of an effective adjustment mechanism can lead to traffic paralysis and safety accidents. Summary of the Invention

[0005] In order to solve the technical problems existing in the prior art that in a dynamically changing traffic environment, it is impossible to fully adapt to real-time road condition changes. Especially during traffic peaks or emergencies, fixed signal settings are difficult to cope with instantaneous increases and decreases in traffic flow, which easily cause traffic congestion and waste of resources. Existing methods also show limitations in the recognition and analysis of vehicle behaviors, lacking a detailed judgment of the impact of specific behaviors. For example, the recognition of lateral offset and longitudinal delay is insufficient, resulting in the signal control being unable to accurately reflect the actual passing state of vehicles and affecting the smoothness of traffic flow. The problem not only reduces the efficiency of traffic signals but also increases the risk of road use. Especially in the management of complex intersections, the lack of an effective adjustment mechanism leads to traffic paralysis and safety accidents. The present invention provides an intelligent traffic signal parameter configuration method and system. The technical solution is as follows: On the one hand, an intelligent traffic signal parameter configuration method is provided, and the method includes: S1: Obtain the vehicle passing data of the lanes at the signal-controlled intersection, extract the lateral offset positions and longitudinal delay time series before and after the steering action, and judge through the vehicle deceleration frequency corresponding to the behavior to generate an interference behavior response tag group; S2: Call the marked positions in the interference behavior response tag group, obtain the starting time of the green light, the starting time of the first vehicle, and the release time of the last vehicle within the section, identify the behaviors and passing deviation segments within the signal control period, and generate a signal behavior delay section; S3: According to the channels indicated in the signal behavior delay section, extract the waiting duration of the trailing vehicles at the end of the green light within two cycles of the channel and the release completion time point, calculate the time gap with the real-time green light end point, judge whether the channel release is incomplete, screen the waiting duration corresponding to the incomplete channel as the priority sorting benchmark, and generate a channel green light variable period configuration table; S4: Call the starting configuration value in the channel green light variable period configuration table, detect the longitudinal distance of the queuing vehicles in the real-time cycle, and identify the continuous acceleration time points of three adjacent vehicles. If the acceleration occurrence times are all lower than the signal preparation switching point, record the synchronous start state and generate a signal cycle start instruction.

[0006] As a further solution of the present invention, the interference behavior response tag group includes a passing interruption identification number, an interference behavior classification code, and a lane space mapping position. The signal behavior delay section includes a signal start response delay length, a vehicle release rhythm lag length, and a behavior trigger and passing rhythm offset length. The channel green light variable period configuration table includes a channel sequence sorting weight, a green light period adjustment amplitude, and a scheduling control effective time. The signal cycle start instruction includes a cycle switching trigger state, a synchronous start trigger moment, and a vehicle distance compression identification state.

[0007] As a further solution of the present invention, the step of obtaining the interference behavior response tag group is specifically as follows: S101: Obtain the vehicle passing data of the lanes at the signal control intersection, extract the lateral coordinate change amount and the longitudinal position change time sequence before and after each vehicle performs a turning action, calculate the lateral offset distance and the longitudinal action delay period based on the lateral coordinate difference and the time length of the longitudinal sequence each time, and generate an offset and delay time sequence parameter set; S102: According to the lateral offset distance and the longitudinal action delay period of the vehicle in the offset and delay time sequence parameter set, combined with the speed change information of the vehicle at consecutive moments before and after the turning action, count the deceleration times of each vehicle in the offset and delay stages, identify the synchronous distribution relationship between the deceleration times and the delay periods, and generate the deceleration frequency during the vehicle turning stage; S103: Based on the deceleration frequency during the vehicle turning stage, judge the vehicle behavior with the deceleration frequency exceeding the preset turning behavior critical frequency, bind the corresponding behavior actions to the road section positions where they are located, and generate an interference behavior response tag group.

[0008] As a further solution of the present invention, the step of obtaining the signal behavior delay section is specifically as follows: S201: Call the road position section identified in the interference behavior response tag group, obtain the three time data of the green light start time, the leading vehicle start time, and the trailing vehicle release time in the corresponding section during the signal control cycle, calculate the leading vehicle response time and the trailing vehicle passing time interval respectively, and generate passing key time interval data; S202: Based on the leading vehicle response time and the trailing vehicle passing time interval values in the passing key time interval data, call the total number of vehicles and the vehicle position distribution in the green light cycle of the corresponding section, identify the total queue length and the queue sparsity level, and obtain the passing density distribution information; S203: According to the passing density distribution information, combined with the time interval distribution of the front, middle, and rear position points of the queue in the passing key time interval, calculate the passing density offset characteristic value, identify the position interval of the density abnormal section in the green light control cycle, and generate the passing density offset time period; The formula for calculating the passing density offset characteristic value is as follows: ; Wherein, AD represents the passing density offset characteristic value, n represents the number of density detection sections participating in the evaluation during the green light cycle, represents the vehicle queue length in the i-th detection section, represents the passing time length of the trailing vehicle in the i-th detection section, represents the average longitudinal response density value of the front, middle, and rear groups of vehicles in the i-th detection section, represents the delay time of the response of the trailing vehicle to the leading vehicle in the queuing section of the i-th detection section, and BD represents the set reference traffic density benchmark value; S204: Based on the corresponding section position in the traffic density offset time period, evaluate the correlation between the time period in the signal control cycle and the vehicle release behavior segment, screen the continuous segments with offset vehicle release behavior, mark them as the areas where traffic behavior delays occur, and generate the signal behavior delay section.

[0009] As a further solution of the present invention, the acquisition steps of the channel green light variable time period configuration table are specifically as follows: S301: According to the channel number indicated in the signal behavior delay section, extract the stationary duration and release completion time point of the trailing vehicle at the end of the green light in two consecutive signal cycles of the channel, and combine the time difference between the release completion time point and the end time point of the green light in the current cycle to obtain the release lag information of the channel tail section; S302: Based on the release lag information of the channel tail section, judge whether the release lag value exceeds the channel release completion benchmark threshold, screen the channel numbers with incomplete releases, extract the waiting time of the trailing vehicle in the corresponding channels, sort them according to the waiting duration of the incomplete channels, and generate the channel release waiting priority sequence; S303: Call the sorting information in the channel release waiting priority sequence, sequentially configure additional green light time periods for the channels, adjust the length of the channel green light time period within the total green light duration, record the channel number and the corresponding adjusted green light duration, and generate the channel green light variable time period configuration table.

[0010] As a further solution of the present invention, the acquisition steps of the signal cycle start instruction are specifically as follows: S401: Call the channel start configuration time value recorded in the channel green light variable time period configuration table, detect the longitudinal distance distribution of vehicles in the corresponding channel queuing area in the real-time signal cycle, identify the continuous acceleration start time points of three adjacent vehicles in the queuing section, and generate the queuing acceleration start time; S402: According to the time points of the acceleration actions of three adjacent vehicles in the queuing acceleration start time, judge whether the acceleration time is earlier than the signal preparation switching time point in the channel green light variable time period. If the judgment condition is met, mark it as the synchronous start state, bind the channel number and the status information, and generate the signal cycle start instruction.

[0011] As a further solution of the present invention, the formula for judging whether the acceleration time is earlier than the signal preparation switching time point in the channel green light variable time period is as follows: ; where, AT represents the average start time difference, , , respectively represent the acceleration start times of the first, second, and third vehicles in the group represents the speed of the a-th vehicle at the sampling moment before the acceleration action, OV represents the passing reference speed set for the road section, k is the weighting coefficient, and m is the number of vehicles

[0012] As a further solution of the present invention, the method further includes step S5: S5: Invoke the marked period of the signal cycle start instruction, screen the passing trajectories of the trailing vehicles in the green light rear section, compare the vehicle speed change trend with the rear section duration. If the speed remains rising and does not reach stability, calculate the time required to reach the predetermined release point and update the rear section control cycle to obtain the continuous regulation result of the rear section green light duration The continuous regulation result of the rear section green light duration includes the remaining release path continuation duration, the predicted period for the trailing vehicle to complete the release, and the recommended signal holding time window

[0013] As a further solution of the present invention, the steps for obtaining the continuous regulation result of the rear section green light duration are specifically as follows: S501: Invoke the signal cycle number marked in the signal cycle start instruction, screen the trajectory data of the trailing vehicles in the green light rear section under the corresponding cycle, extract the continuous speed sequence and timestamp data from the start moment of the rear section to the moment when the vehicle exits the release line, and generate the rear section passing trajectory sequence S502: Based on the speed sequence in the rear section passing trajectory sequence, analyze the speed change trend of the trailing vehicle during the rear section green light duration, extract the speed increase value at the end of the sequence, and compare it with the corresponding duration of the rear section. If the speed continues to rise and does not enter the stable interval, calculate the additional time required for the vehicle to reach the predetermined release point, and generate the remaining release time interval for the rear section S503: Invoke the additional time value required for the channel in the remaining release time interval of the rear section, update the real-time rear section green light control cycle, correct the original configured end time point of the rear section green light, and re-set the control interval of the rear section of the channel to obtain the continuous regulation result of the rear section green light duration

[0014] On the other hand, the intelligent traffic signal parameter configuration system is used to execute the above intelligent traffic signal parameter configuration method, and the system includes: The behavior data recognition module detects the vehicle passing behaviors of the lanes at the signal-controlled intersection, collects the lateral offset positions and longitudinal delay time series of the vehicles, counts the deceleration frequencies of the vehicles, and generates an interference behavior response tag group The interference behavior analysis module, based on the interference behavior response tag group, identifies the key behavior actions that cause traffic interruption, binds the key behavior actions to the locations, and generates signal behavior delay sections The signal delay calculation module calls the signal behavior delay section, obtains the starting time of the green light, the starting time of the first vehicle, and the release time of the last vehicle within the section, calculates the time interval, and generates a channel green light variable time period configuration table; Based on the channel green light variable time period configuration table, the green light time dynamic configuration module analyzes the completion situation of the channel release, calculates the waiting duration of the uncompleted channels, reallocates the green light time period as needed, and generates a signal cycle start instruction; The signal start module calls the signal cycle start instruction, detects and evaluates the longitudinal distance and continuous acceleration points of the queued vehicles, adjusts the signal start time according to the vehicle speed change trend and the tail section duration, and obtains the tail section green light continuous regulation result.

[0015] The beneficial effects brought by the technical solutions provided in the embodiments of the present invention at least include: By carefully identifying and analyzing the vehicle behavior and position characteristics, the adaptability and accuracy of signal control are effectively improved. The lateral offset and longitudinal delay behaviors of the vehicles are extracted and combined with the actual traffic interruption behaviors, so that the traffic signal can more accurately reflect the real-time road conditions and reduce traffic jams caused by signal inadaptability. The analysis of the vehicle deceleration frequency during the identification process further optimizes the signal scheduling, dynamically adjusts the signal light cycle through real-time data, and reduces traffic congestion caused by untimely response. The calculation of the waiting duration and release time of the tail vehicles, and the allocation of variable green light time periods for each channel provide a more reasonable prediction of the passing time, ensuring the continuity and high efficiency of the traffic flow. The combination of the vehicle speed change trend and the adjustment of the control cycle enables the tail section green light time to be flexibly regulated according to actual needs, further reducing traffic delays and improving the road utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the working process of the present invention; Figure 2 It is a system flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The technical solutions in the present invention will be described below with reference to the accompanying drawings.

[0018] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "example" in the present invention should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0019] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0020] Please refer to Figure 1 , an embodiment of the present invention provides an intelligent traffic signal parameter configuration method, and the processing flow of this method may include the following steps: S1: Obtain the vehicle passing data of the lanes at the signal-controlled intersection, extract the lateral offset position and longitudinal delay time sequence before and after the turning action, judge through the vehicle deceleration frequency corresponding to the behavior, identify the behavior actions causing traffic interruption and bind them to the corresponding positions, and generate an interference behavior response tag group; S2: Call the marked positions in the interference behavior response tag group, obtain the starting time of the green light, the starting time of the first vehicle and the release time of the last vehicle within the section, calculate the time intervals among the three and compare the queue lengths, identify the behaviors and traffic deviation segments within the signal control period, and generate a signal behavior delay section; S3: According to the channels indicated in the signal behavior delay section, extract the waiting duration of the last vehicle at the end of the green light and the release completion time point within two cycles of the channel, calculate the time difference from the real-time green light end point, judge whether the channel release is not completed, screen the waiting durations corresponding to the uncompleted channels as the priority sorting basis, and allocate variable green light time periods to each channel in sequence to generate a channel green light variable time period configuration table; S4: Call the starting configuration value in the channel green light variable time period configuration table, detect the longitudinal distance of the queuing vehicles in the real-time period, and identify the continuous acceleration time points of three adjacent vehicles. If the acceleration occurrence times are all lower than the signal preparation switching point, record the synchronous start state and generate a signal cycle start instruction; S5: Call the marked cycle of the signal cycle start instruction, screen the passing trajectories of the last vehicle in the latter part of the green light, compare the vehicle speed change trend with the duration of the latter part. If the speed remains rising and does not reach stability, calculate the time required to reach the predetermined release point and update the latter part control cycle to obtain the latter part green light continuous regulation result; The interference behavior response tag group includes a traffic interruption identification number, an interference behavior classification code, and a lane space mapping position. The signal behavior delay section includes a signal start response delay length, a vehicle release rhythm lag length, and a behavior trigger and traffic rhythm offset length. The channel green light variable time period configuration table includes a channel sequence sorting weight, a green light time period adjustment amplitude, and a scheduling control effective time. The signal cycle start instruction includes a cycle switching trigger state, a synchronous start trigger moment, and a vehicle distance compression identification state. The latter part green light continuous regulation result includes a remaining release path continuation duration, a last vehicle release completion prediction period, and a signal hold recommended time window.

[0021] The specific steps for obtaining the interference behavior response tag group are as follows: S101: Obtain the vehicle passing data of the lanes at the signal-controlled intersection, extract the lateral coordinate change amount and the longitudinal position change time sequence before and after each vehicle executes a turning action, calculate the lateral offset distance and the longitudinal action delay period based on the lateral coordinate difference and the time length of the longitudinal sequence each time, and generate an offset and delay time sequence parameter set; It is necessary to rely on the camera recognition device and the geomagnetic induction device deployed at the signal-controlled intersection to record the vehicles passing through the lane all-weather by frames, and based on the positions of each vehicle at different time points in the frame image, identify its trajectory direction. Combining the image frame sequences of each vehicle before and after the turning action occurs, extract the lateral coordinate values of each vehicle within a continuous period, identify the coordinate difference before and after its turning, and at the same time obtain the corresponding time range of this process. The offset direction and amplitude of the vehicle trajectory in the lane can be marked through the lateral coordinate point sequence. Set that in an actual intersection, after the signal is released, a car deviates from the straight lane to the left-turn direction. The front of the car shows an obvious leftward movement trend at the 15th frame and completely enters the left-turn lane at the 22nd frame. This time period is marked as the turning section. At the same time, extract the time sequence of the vehicle's longitudinal displacement, combine the geomagnetic data to determine its relative coordinate sequence, generate the lateral offset data and the longitudinal action duration sequence of each vehicle. After completing the annotation of this process for the vehicle, combine the lane number, the passing time period and the vehicle ID to establish a data set with a multi-field structure, form the statistical parameters of the offset and the delay, and form an offset and delay time sequence parameter set.

[0022] S102: According to the lateral offset distance and the longitudinal action delay period of the vehicle in the offset and delay time sequence parameter set, and combining the speed change information of the vehicle at consecutive moments before and after the turning action, count the deceleration times of each vehicle in the offset and delay stages, identify the synchronous distribution relationship between the deceleration times and the delay periods, and generate the deceleration frequency of the vehicle turning stage; It is necessary to further call the speed change sequence of the vehicle per second before and after turning, extract the speed values of each time period in combination with the frame rate synchronization, and judge whether there is a continuous deceleration process. Taking the left-turn lane in the intersection as an example, during the turning process of a bus, its speed drops from the original 9 meters per second to 7 meters, and then to 6 meters, indicating that this process is affected by the turning path or the front block. It is necessary to mark it as an effective deceleration action. At the same time, detect whether there are two or more speed drops in the time periods before and after the vehicle deviates, record the deceleration action and its occurrence time period, and count the occurrence times. After making this judgment on the vehicle, construct a result list containing the deceleration action times of each vehicle, pair the deceleration times with their corresponding longitudinal delay times, and judge whether they show a synchronous distribution structure, that is, whether most deceleration actions occur in the lateral deviation section or its time period. If it is satisfied, it is classified into the synchronous group. In the cross-intersection test, vehicles in multiple lanes show an obvious synchronous deceleration trend, especially when vehicles are close to the signal light area during peak hours. This kind of data can be directly recorded and summarized as the behavioral performance characteristics of the corresponding lane in the turning stage, generating the deceleration frequency of the vehicle turning stage.

[0023] S103: Based on the deceleration frequency in the vehicle turning stage, judge the vehicle behavior with a deceleration frequency exceeding the preset critical frequency of turning behavior, bind the corresponding behavioral actions to the position of the road section where they are located, and generate a group of interference behavior response tags; It is necessary to set the judgment conditions for turning behavior and screen and process the deceleration frequency data to clarify that some vehicle behaviors are abnormal. Taking Area A of the crossroads as an example, summarize and count the vehicle samples throughout the day. Among them, divide the frequency distribution of vehicles with turning behavior and observe the range of deceleration times under normal conditions. After analysis, it is found that normal turning vehicles decelerate once within one second. If there are two or more times, it can be classified as a behavior-intensive section. Set the critical frequency standard of behavior as once per second. If the frequency of a certain vehicle reaches or exceeds this value, it is determined that the behavior is overly complex. Further, classify and organize the vehicle behavior information exceeding this frequency standard, and bind its occurrence position to the lane number it belongs to. Suppose a vehicle performs a left-turn action in the second lane, and its deceleration behavior exceeds the normal range and is concentrated at the end of the lane, indicating that there is interference or obstacle influence at this position. Archive and structure this kind of abnormal behavior information to form the corresponding behavior type identifier, clarify its road space position and time window, and constitute the associated tag data between vehicle behavior and road section, generating a group of interference behavior response tags.

[0024] The specific steps for obtaining the signal behavior delay section are as follows: S201: Call the road position section identified in the interference behavior response tag group to obtain three time data of the green light start time, the starting time of the first vehicle, and the release time of the last vehicle in the corresponding section during the signal control cycle. Calculate the time interval between the first vehicle's response time and the last vehicle's passing time respectively, and generate the passing critical time interval data; Based on the section number, interference behavior timestamp, and position coordinates stored in the tag group, determine the specific road segment and direction attribute where the interference behavior occurs. Retrieve the periodic release data of this section in the signal control. Each control cycle's record should extract the time point when the green light starts, the starting time point when the first vehicle starts to move, and the ending time point when the last vehicle in the queue completely exits the detection area. For example, in the second lane of a certain intersection, the green light starts at 8:01:10, the first vehicle starts to move at 8:01:14, and the last vehicle completely leaves the induction coil area at 8:01:28. Calculate that the first vehicle's response time is 4 seconds and the release duration of the last vehicle is 14 seconds respectively. Similar data can be obtained through automatic detection equipment or by backtracking and annotating image frames with a camera. The timestamps are recorded in milliseconds to eliminate manual errors. By batch processing the time points of each channel and each cycle, record the time interval values of the key events of the first vehicle and the last vehicle item by item, and associate each group of data with the corresponding road section and signal cycle number. Finally, summarize and output as structured data to obtain the passing critical time interval data.

[0025] S202: Based on the first vehicle's response time and the last vehicle's passing time interval values in the passing critical time interval data, call the total number of vehicles and the vehicle position distribution within the green light cycle of the corresponding section to identify the total queue length and the queue sparsity level, and obtain the passing density distribution information; Extract the number of vehicles entering during the green light control period and their initial arrangement position information under the corresponding signal cycle from the same data source. Restore the vehicle queue structure through geomagnetic coils or camera induction. For the stationary vehicles in the same lane before the green light of the current cycle, record their coordinate ranges. Mark the total queue length by calculating the maximum longitudinal distance from the head to the tail of the vehicle, and at the same time count the number of vehicles within this range to calculate the number of vehicles per unit distance. Judge its queue sparsity degree. If there are 12 queuing vehicles and the total queue length is 96 meters, then there is 1 vehicle per 8 meters on average. According to the set standard, judge that this value is greater than the dense standard of 1 vehicle per 6 meters, so it is a sparse queue. By horizontally comparing the vehicle number and position data sequences at different times, it is possible to further identify whether there are breaks or vacancies in the queue. Then output the distribution labels of the density and sparsity degrees of each section in matrix form, and combine and match the distribution information with the aforementioned time interval data to form the fusion analysis data of passing behavior in the two dimensions of time and space, and obtain the passing density distribution information.

[0026] S203: Calculate the traffic density offset eigenvalue according to the traffic density distribution information, and in combination with the time interval distribution of the front, middle, and rear position points in the queue during the critical traffic time interval, identify the position interval of the density anomaly section in the green light control cycle, and generate the traffic density offset time period. The formula for calculating the traffic density offset eigenvalue is as follows: ; where AD represents the traffic density offset eigenvalue, n represents the number of density detection sections participating in the evaluation within the green light cycle, represents the vehicle queue length within the i-th detection section, represents the passing time length of the trailing vehicle within the i-th detection section, represents the average longitudinal response density value of the front, middle, and rear groups of vehicles in the i-th detection section, represents the delay time of the trailing vehicle in the queue section responding to the leading vehicle in the i-th detection section, and BD represents the set reference traffic density baseline value; Meaning of parameters and derivation process of formula calculation: During one signal control cycle, five density detection sections are obtained through road induction equipment and cameras, that is, n = 5; In each data collection, the vehicle queue length to is obtained by multiplying the number of vehicles counted from the lane tail to the head by the headway. After actual measurement: = 48m, = 60m, = 52m, = 44m, = 50m; The passing time length of the trailing vehicle to is monitored by the time taken from when the green light comes on to when the trailing vehicle drives out of the release line. The result is: = 10s, = 12s, = 9s, = 11s, = 10s; The longitudinal response density values of the front, middle, and rear vehicles in each detection section to are obtained by tracking the three-vehicle spacing at 1-second intervals through the camera and converting to the average longitudinal response density value, with the unit of vehicles per second. The specific values are: = 4.5, = 5.0, = 4.3, = 4.8, = 4.7; Response delay time of the last vehicle to the first vehicle within the queuing section to By detecting the time difference between the start of the first vehicle and the start of the last vehicle, the collected results are as follows: = 1.2s, = 0.8s, = 1.4s, = 0.9s, = 1.1s; The reference passing density benchmark value BD is determined based on the sample statistics of the green light efficient release period in the past 30 days as: BD = 4.6; Substitute the above values into each item of the formula and calculate them in turn: The first part is the average absolute difference: ; ; ; ; ; Average after summing the first part: ; The second part is the root mean square of the residuals: ; ; ; ; ; Calculation of the root mean square of the residuals: ; Substitute into the formula for calculation: AD = 0.576 + 0.4655 = 1.0415; This result indicates that the passing density deviation characteristic value is 1.0415. This value is greater than the deviation critical judgment standard of 0.85, indicating that there is a significant deviation in density during the current signal cycle, and the corresponding time segment needs to be marked as a passing density deviation time period.

[0027] S204: Based on the corresponding section position in the passing density deviation time period, evaluate the correlation between the time section and the vehicle release behavior segment within the signal control cycle, screen out the continuous segments with deviation in vehicle release behavior, mark them as the areas where passing behavior delays occur, and generate signal behavior delay sections; It is necessary to parse each vehicle release trajectory within the signal cycle, map and match the time points when each vehicle's release behavior occurs during the passage process with this time period, and compare whether there are trajectory segments where vehicles have significant delays in abnormal sections. If a vehicle in a certain lane starts before the start of the density offset period, but its adjacent following vehicle is delayed for up to 6 seconds without starting, this behavior is reflected as a release chain break in the trajectory. After summarizing such behaviors, they are classified and screened to identify vehicle behavior segments that meet the characteristics of "separation of start and release", and mark their position ranges within the signal control cycle. By analyzing the segments, a set of passage behaviors with tail response times far exceeding the average response level and significantly lagging release trajectories can be summarized, which are bound to the signal cycle number and road position and output, and integrated to form structured identification result data, generating signal behavior delay sections.

[0028] The steps for obtaining the configuration table of the variable green light period of the lane are specifically as follows: S301: According to the lane number indicated in the signal behavior delay section, extract the stationary duration and release completion time point of the tail vehicle of the lane at the end of the green light in two consecutive signal cycles, and combine the time difference between the release completion time point and the end time point of the green light in the current cycle to obtain the release lag information of the tail section of the lane. It is necessary to retrieve the road data calibrated by the delay label to further obtain the status change records of the tail vehicle of this lane during the green light period of the current cycle and the previous cycle. In intelligent transportation, this type of data is jointly recorded by the video acquisition device and induction coil at the end of the lane. The time point when the vehicle head leaves the stop line is used as the release completion mark. In actual operation, the last 5 seconds of the green light at the tail section of each lane cycle are set as the end time zone. Statistics are made on whether the stationary state of the tail vehicle continues within this interval, and the time point from the start of movement to completely leaving the stop line. For example, in the first cycle of a certain lane, the tail vehicle remains stationary until the end of the green light and does not complete the release, while in the second cycle, the release time of this vehicle is 2 seconds before the end of the green light, indicating that it is in a delayed release state. Compare the release completion time point with the official end time point of the green light and record the time difference. If the difference is more than 0.5 seconds, it is classified into the lag record. The data of the tail vehicle in each cycle are summarized on this basis, automatically recording the stationary duration and release completion point, and forming a data structure composed of the lane number and release difference, sorting out the release lag information of the tail section of the lane.

[0029] S302: Based on the release lag information of the tail section of the lane, judge whether the release lag value exceeds the release completion benchmark threshold of the lane, screen the lane numbers with incomplete releases, extract the waiting times of the tail vehicles in the corresponding lanes, sort them according to the waiting durations of the incomplete lanes, and generate a lane release waiting priority sequence. During the evaluation process, the hysteresis values of each channel need to be classified to determine whether they exceed the preset benchmark threshold for channel release completion. The setting of this benchmark value refers to the regional traffic flow characteristics and data samples. For the morning peak commuting period, the hysteresis threshold is set to 1.5 seconds. If the tail release hysteresis value of a certain channel reaches 2.1 seconds, it is marked as an uncompleted release channel. Based on this, the channel numbers exceeding the benchmark are screened out, and the cumulative waiting time of the corresponding tail vehicle in the green light phase in this channel is called to obtain the complete waiting duration of the tail vehicle from the start of the green light to the completion of release. Taking a certain channel as an example, the total waiting duration of its tail vehicle from the start of the green light to the completion of release is 17 seconds, which exceeds the upper limit of the average waiting time in the channel and needs to be given priority for regulation. The channel numbers of uncompleted releases are sorted in descending order according to their corresponding waiting times, and data pairs corresponding to the channel numbers and waiting durations are established. The priority levels are marked according to the sorting numbers for subsequent use as the basis for the time period resource allocation order, and a channel release waiting priority sequence is generated.

[0030] S303: Call the sorting information in the channel release waiting priority sequence, and sequentially allocate additional green light time periods for the channels. Adjust the length of the channel green light time period within the total green light duration, record the channel number and the corresponding adjusted green light duration, and generate a channel green light variable time period configuration table; According to the sorting levels, additional green light time periods are allocated to the channels one by one, and they are re-divided and adjusted within the total configurable green light duration of the current signal control to determine the total green light variable time budget, which is set to 20 seconds. Channels with higher priorities are allocated larger proportions of time, and it decreases sequentially. If the channel ranked 1st is numbered A01, an additional 6 seconds of green light can be configured for it, and the 2nd and 3rd ranked channels are each allocated 5 seconds and 4 seconds, and the remaining channels are allocated according to the remaining budget. Each allocation records the channel number, the sum of the original green light duration and the newly configured time, and updates the green light time of this channel in the next cycle. In the actual configuration table, each record includes five data fields: channel number, current signal cycle number, original duration, newly added duration, and combined green light duration. Structured output is automatically generated, and interface parameters for the configuration items in the corresponding section are established to ensure matching with the traffic control terminal, and a channel green light variable time period configuration table is generated.

[0031] The specific steps for obtaining the signal cycle start instruction are as follows: S401: Call the channel start configuration time value recorded in the channel green light variable time period configuration table, detect the longitudinal distance distribution of vehicles in the corresponding channel queuing area within the real-time signal cycle, identify the continuous acceleration start time points of three adjacent vehicles in the queuing section, and generate the queuing acceleration start time; Extract the channel number and the starting time of the green light for each record in the configuration table, and locate the specific starting point of the green light segment where the channel is located in the current cycle. Obtain the position coordinates and identification numbers of the stationary or moving vehicles in the queuing area of the channel through continuous frame cameras or geomagnetic coils deployed at the lane entrance and the central area, and statistically calculate the position of the center point of the front of each vehicle with the longitudinal coordinate as the main axis to form a longitudinal distance distribution map. Sample the longitudinal coordinates of the vehicles at 1-second intervals in the distribution map, and perform a sequence comparison on the longitudinal coordinates of the vehicles to identify whether the longitudinal displacement of the vehicle is greater than the set minimum acceleration threshold in two consecutive frames. If the displacements of vehicles B, C, and D are continuously greater than 1 meter, mark them as in a continuous acceleration state, and record the time point when each of them first meets this condition, which is the starting time point of acceleration. Set vehicle B to start accelerating 2 seconds after the start of the green light, vehicle C at 2.2 seconds, and vehicle D at 2.4 seconds, and all three show a sequential acceleration state. Record the time stamp as an acceleration group event, and construct a data set structure together with its channel number, vehicle number, and acceleration starting time, and summarize to generate the starting time of queuing acceleration.

[0032] S402: According to the time points of the acceleration actions of three adjacent vehicles in the starting time of queuing acceleration, judge whether the acceleration times are all earlier than the signal preparation switching time point within the variable period of the channel green light. If the judgment condition is met, mark it as a synchronous start state, bind the channel number and status information, and generate a signal cycle start instruction; The formula for judging whether the acceleration time is earlier than the signal preparation switching time point within the variable period of the channel green light is as follows: ; Among them, AT represents the average starting time difference, 、 、 respectively represent the acceleration starting times of the first, second, and third vehicles in this group, represents the speed of the a-th vehicle at the sampling moment before the acceleration action, OV represents the passing reference speed set for the road section, k is the weighting coefficient, and m is the number of vehicles; Parameter meaning and formula calculation derivation process: 、 、 : Through a non-contact speedometer or a drum test bench with high-frequency sampling (≥10Hz), record the speed-time curve of the vehicle during the starting acceleration process. When the vehicle speed first rises from 0 km / h to 1 km / h, record this moment as the starting time of acceleration. Set the starting times of acceleration for the three vehicles as follows: The first vehicle (vehicle 1): = 0.8s; The second vehicle (vehicle 2): = 1.1s; The third vehicle (Vehicle 3): = 1.4 s; , , : The speed values obtained at the last sampling moment before vehicle acceleration through in-vehicle wheel speed sensors or GPS. Set the speeds of the three vehicles as follows: Vehicle 1: = 0.5 m / s; Vehicle 2: = 0.6 m / s; Vehicle 3: = 0.4 m / s; OV: The design passing reference speed of the urban arterial road is 60 km / h, i.e., OV = 60 km / h = 16.67 m / s; k: The weighting coefficient used to correct the influence of speed difference on the starting time difference. It is set based on the sensitivity analysis of the speed difference to the time difference and is set as: k = 0.05; Formula calculation process: Calculate the time difference: ; ; Calculate the speed difference term: ; ; Calculate the square root term: ; ; Substitute into the formula for calculation: ; This result indicates that the average starting time sequence difference of the three vehicles is 1.348 seconds. If this value is less than the set synchronous start threshold (e.g., 2 seconds), it can be determined that the current queue is in a synchronous start state, and a signal cycle start instruction is generated accordingly.

[0033] The specific steps for obtaining the regulation result of the green light duration in the tail section are as follows: S501: Call the signal cycle number marked in the signal cycle start instruction, filter the trajectory data of the tail vehicle in the second half of the green light under the corresponding cycle, extract the continuous speed sequence and timestamp data from the start moment of the tail section to the moment when the vehicle exits the release line, and generate the passing trajectory sequence of the tail section; First, match the control cycle information corresponding to the current analysis object from the signal regulation data table, and then screen the passing trajectory records of the trailing vehicle in this cycle from the vehicle trajectory database. In actual operation, through the geomagnetic sensors or camera recognition deployed in the detection area at the tail section, continuously record the speed changes and time information of the last three vehicles within the end section of the green light. Align the sampling frames based on the starting time of the green light, and define the starting time of the tail section as the stage after the fifth second before the end of the green light. During this time period, the motion state of each vehicle is sampled at intervals of 0.5 seconds, and the speed change values and corresponding timestamps are extracted to form a complete time series. For example, a trailing vehicle samples its speed every 0.5 seconds starting from the 7th second of the green light, and a total of ten sets of speed data are obtained to construct a complete passing trajectory time series within the tail section. Archive this time series with the vehicle number as the primary key, mark the channel number, signal cycle number, and start and end times of the frame, and output the passing trajectory sequence of the tail section.

[0034] S502: Based on the speed sequence in the passing trajectory sequence of the tail section, analyze the speed change trend of the trailing vehicle within the green light duration of the tail section, extract the speed increase value at the end of the sequence, and compare it with the corresponding duration of the tail section. If the speed continues to rise and has not entered the stable range, calculate the additional time required for the vehicle to reach the predetermined release point, and generate the remaining release time of the tail section; It is necessary to conduct a trend analysis on the motion characteristics of the trailing vehicle within the green light duration of the tail section to determine whether the vehicle is in a continuous acceleration state and has not entered the stable operation range. The specific analysis steps are as follows: Take three consecutive sets of data from the end of the speed sequence and calculate their increase values for comparison with the previous speed increase amplitude. If the difference maintains a positive growth trend and the change does not tend to converge, it is determined that the vehicle has not passed stably. If the speed of the vehicle increases from 5.6 m / s to 6.1 m / s and then to 6.7 m / s in the last three frames, and the continuous increase amplitude exceeds 0.5 m / s each time, it is judged that the vehicle is still in the acceleration state. Then, compare it with the remaining time of the tail section. If the remaining time of the current green light tail section is only 1.5 seconds, and the vehicle still needs 2.3 seconds to reach the release point, then 0.8 seconds of additional time needs to be added. Record this time value and archive it together with the channel number and vehicle number. Each set of data generated includes the acceleration trend state, the speed increase at the end of the section, the additional time, and the signal cycle number, forming the remaining release time of the tail section.

[0035] S503: Call the additional time value required for the channel in the remaining release time of the tail section, update the real-time green light control cycle of the tail section, correct the originally configured end time point of the green light in the tail section, and reset the control interval of the tail section of the channel to obtain the continuous regulation result of the green light in the tail section; The dynamic adjustment of the original end-segment green light control configuration will be carried out in the current signal cycle. Read the end time point of each channel in the current cycle, and then update the end time one by one according to the make-up time recorded in the remaining release time of the end segment. Set the current end time of channel B03 to the 14th second of the green light. After judging that 1.2 seconds need to be made up, extend its end point to 15.2 seconds. This operation needs to consider whether the total cycle time is sufficient and complete the duration allocation on the basis of not affecting the channel control rhythm. Record the timestamps before and after each adjustment, and synchronously update the controller output data. Structurally organize the configuration results according to the four items of channel number, cycle number, original end time, and new end time, and output a control parameter configuration file in a standardized format for real-time update and scheduling of the signal control terminal, and output the continuous regulation result of the end-segment green light.

[0036] Please refer to Figure 2 , an intelligent traffic signal parameter configuration system, which includes: The behavior data recognition module detects the vehicle passing behavior of the lanes at the signal-controlled intersection, collects the lateral offset position and longitudinal delay time series of the vehicles, counts the deceleration frequency of the vehicles, and generates an interference behavior response tag group; The interference behavior analysis module, based on the interference behavior response tag group, identifies the key behavior actions that cause traffic interruption, binds the key behavior actions to the location, and generates a signal behavior delay section; The signal delay calculation module calls the signal behavior delay section, obtains the green light start time, the starting time of the first vehicle, and the release time of the last vehicle within the section, calculates the time interval, and generates a channel green light variable time period configuration table; The green light time dynamic configuration module, based on the channel green light variable time period configuration table, analyzes the completion situation of the channel release, calculates the waiting time of the uncompleted channels, reallocates the green light time period as needed, and generates a signal cycle start instruction; The signal start module calls the signal cycle start instruction, detects and evaluates the longitudinal distance and continuous acceleration points of the queued vehicles, adjusts the signal start time according to the vehicle speed change trend and the end-segment duration, and obtains the continuous regulation result of the end-segment green light.

[0037] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the said claims.

Claims

1. A method for configuring intelligent traffic signal light parameters, characterized in that: The following steps are involved: S1: Obtain vehicle traffic data of lanes at signal-controlled intersections, extract the lateral offset position and longitudinal delay sequence before and after the turning action, make a judgment based on the vehicle deceleration frequency corresponding to the behavior, and generate an interference behavior response label group; S2: calling the marked position in the interference behavior response tag group, obtaining the green light start time, the first vehicle start time and the last vehicle release time in the section, identifying the behavior and traffic deviation fragments in the signal control cycle, and generating a signal behavior delay section; S3: According to the channel indicated in the signal behavior delay section, the waiting time of the rear vehicle at the end of the green light and the release completion time point within two cycles of the channel are extracted, the time difference between the end point of the real-time green light is calculated, and it is determined whether the channel release is not completed. The waiting time corresponding to the unfinished channel is selected as the priority sorting basis, and the channel green light variable time period configuration table is generated; S4: Call the starting configuration value in the channel green light variable period configuration table, detect the longitudinal distance of the vehicles in the queue in the real-time cycle, and mark the continuous acceleration time points of the three adjacent vehicles. If the acceleration time is lower than the signal preparation switching point, record the synchronous start status and generate a signal cycle start instruction.

2. The intelligent traffic signal light parameter configuration method according to claim 1, characterized in that: The interference behavior response label group includes the traffic interruption identification number, the interference behavior classification code, and the lane space mapping position; the signal behavior delay section includes the signal start response delay length, the vehicle release rhythm lag length, and the behavior trigger and traffic rhythm offset length; the channel green light variable time period configuration table includes the channel sequence sorting weight, the green light time period adjustment amplitude, and the scheduling control effective time; the signal cycle start instruction includes the cycle switching trigger state, the synchronous start trigger time, and the vehicle distance compression recognition state.

3. The intelligent traffic signal light parameter configuration method according to claim 1, characterized in that: The steps for obtaining the interference behavior response label group are specifically as follows: S101: Acquire vehicle traffic data of lanes at signal-controlled intersections, extract the lateral coordinate change and longitudinal position change time sequence of each vehicle before and after turning, calculate the lateral offset distance and longitudinal action delay period based on the time length of each lateral coordinate difference and longitudinal sequence, and generate an offset and delay time sequence parameter set; S102: according to the lateral offset distance and longitudinal action delay period of the vehicle in the offset and delay timing parameter set, combined with the speed change information of the vehicle at consecutive moments before and after the turning action, counting the number of decelerations of each vehicle in the offset and delay stage, identifying the synchronous distribution relationship between the number of decelerations and the delay period, and generating the deceleration frequency of the vehicle in the turning stage; S103: Based on the deceleration frequency of the vehicle in the turning phase, determine the vehicle behavior whose deceleration frequency exceeds the preset critical frequency of the turning behavior, bind the corresponding behavior action with the road section position, and generate an interference behavior response label group.

4. The intelligent traffic signal light parameter configuration method according to claim 3, characterized in that: The steps for obtaining the signal behavior delay section are specifically as follows: S201: calling the road position section identified in the interference behavior response tag group, obtaining three time data of the green light start time, the first vehicle start time and the last vehicle release time of the corresponding section within the signal control cycle, respectively calculating the first vehicle response time and the last vehicle passing time interval, and generating the key time interval data for passing; S202: Based on the response time of the first vehicle and the time interval of the last vehicle in the critical time interval data, the total number of vehicles and the vehicle position distribution in the green light cycle of the corresponding section are called, the total length of the queue and the sparseness level of the queue are identified, and the traffic density distribution information is obtained; S203: Calculate the traffic density offset characteristic value based on the traffic density distribution information and the time interval distribution of the front, middle and rear position points of the queue in the critical traffic time interval, identify the position interval of the density abnormal section in the green light control cycle, and generate the traffic density offset time period; The formula for calculating the traffic density offset characteristic value is as follows: ； Among them, AD represents the characteristic value of traffic density deviation, n represents the number of density detection segments involved in the evaluation within the green light cycle, represents the length of the vehicle queue in the i-th detection segment, represents the length of time the tail vehicle passes through in the i-th detection segment, Represents the average longitudinal response density value of the front, middle and rear three groups of vehicles in the i-th detection segment, represents the delay time of the tail vehicle in the queue section in the i-th detection section responding to the head vehicle, and BD represents the set reference traffic density benchmark value; S204: Based on the corresponding segment position in the traffic density offset time period, the correlation between the time segment and the vehicle release behavior segment within the signal control cycle is evaluated, and the continuous segments with offset in the vehicle release behavior are screened and marked as the traffic behavior delay occurrence area, and the signal behavior delay segment is generated.

5. The intelligent traffic signal light parameter configuration method according to claim 4, characterized in that: The steps for obtaining the channel green light variable time period configuration table are specifically as follows: S301: According to the channel number indicated in the signal behavior delay section, extract the stationary duration and release completion time point of the rear vehicle at the end of the green light in two consecutive signal cycles of the channel, and combine the time difference between the release completion time point and the end time point of the green light in the current cycle to obtain the release hysteresis information of the end of the channel; S302: Based on the release hysteresis information of the tail section of the channel, determine whether the release hysteresis value exceeds the channel release completion reference threshold, filter the channel numbers of the uncompleted releases, extract the waiting time of the tail vehicles in the corresponding channels, sort the uncompleted channels according to the waiting time, and generate a channel release waiting priority sequence; S303: Call the sorting information in the channel release waiting priority sequence, configure additional green light time periods for the channels in sequence, adjust the channel green light time period length within the total green light duration, record the channel number and the corresponding adjusted green light duration, and generate a channel green light variable time period configuration table.

6. The intelligent traffic signal light parameter configuration method according to claim 5, characterized in that: The steps for obtaining the signal cycle start instruction are specifically as follows: S401: calling the channel start configuration time value recorded in the channel green light variable period configuration table, detecting the longitudinal distance distribution of vehicles in the corresponding channel queue area within the real-time signal cycle, identifying the continuous acceleration start time points of three adjacent vehicles in the queue section, and generating the queue acceleration start time; S402: According to the time points of the acceleration actions of the three adjacent vehicles in the queue acceleration start time, determine whether the acceleration time is earlier than the signal preparation switching time point within the variable period of the channel green light. If the judgment condition is met, it is marked as a synchronous start state, and the channel number and status information are bound to generate a signal cycle start instruction.

7. The intelligent traffic light parameter configuration method according to claim 6, characterized in that: The formula for judging whether the acceleration time is earlier than the signal preparation switching time point within the variable period of the channel green light is as follows: ； Among them, AT represents the average starting timing difference, , , Respectively represent the acceleration start time of the first, second, and third vehicles in the group. It represents the speed of the a-th vehicle at the sampling time before the acceleration action, OV represents the reference speed set for the road section, k is the weighting coefficient, and m is the number of vehicles.

8. The intelligent traffic signal light parameter configuration method according to claim 1, characterized in that: The method further comprises step S5: S5: calling the marking cycle of the signal cycle start instruction, screening the passage trajectory of the rear green light of the rear vehicle, comparing the vehicle speed change trend with the tail segment duration, and if the speed keeps rising but does not reach stability, calculating the time required to reach the predetermined release point and updating the tail segment control cycle to obtain the tail segment green light continuous control result; The result of the continuous control of the tail green light includes the duration of the remaining release path, the predicted time period for the completion of the tail vehicle release, and the recommended time window for signal maintenance.

9. The intelligent traffic signal light parameter configuration method according to claim 8, characterized in that: The steps for obtaining the tail green light continuous control result are as follows: S501: calling the signal cycle number marked in the signal cycle start instruction, filtering the trajectory data of the rear vehicle in the rear section of the green light under the corresponding cycle, extracting the continuous speed sequence and timestamp data from the start time of the rear section to the vehicle driving out of the release line, and generating the rear section passing trajectory sequence; S502: Based on the speed sequence in the tail passage trajectory sequence, the speed change trend of the tail vehicle within the green light duration of the tail segment is analyzed, the speed increase value of the tail segment of the sequence is extracted, and compared with the duration corresponding to the tail segment. If the speed continues to increase and does not enter the stable interval, the time required for the vehicle to reach the predetermined release point is calculated to generate the remaining release time of the tail segment; S503: Call the required supplementary time value of the channel in the remaining release time interval of the tail segment, update the real-time tail segment green light control cycle, correct the originally configured tail segment green light end time point, reset the control interval of the channel tail segment, and obtain the tail segment green light continuous control result.

10. An intelligent traffic light parameter configuration system, characterized in that: The system is used to implement the intelligent traffic signal light parameter configuration method according to any one of claims 1 to 9, and the system includes: The behavior data recognition module detects the vehicle traffic behavior in the lane of the signal-controlled intersection, collects the lateral offset position and longitudinal delay time sequence of the vehicle, counts the vehicle's deceleration frequency, and generates an interference behavior response label group; The interference behavior analysis module identifies the key behavior actions that cause traffic interruption based on the interference behavior response tag group, binds the key behavior actions to locations, and generates a signal behavior delay section; The signal delay calculation module calls the signal behavior delay section, obtains the green light start time, the first vehicle start time and the last vehicle release time in the section, calculates the time interval, and generates a channel green light variable time period configuration table; The green light time dynamic configuration module analyzes the completion of channel release based on the channel green light variable time period configuration table, calculates the waiting time of the unfinished channel, reallocates the green light time period as needed, and generates a signal cycle start instruction; The signal start module calls the signal cycle start instruction, detects and evaluates the longitudinal distance and continuous acceleration points of the vehicles in the queue, adjusts the signal start time according to the vehicle speed change trend and the tail segment duration, and obtains the tail segment green light continuous control result.

Images