Traffic signal control system based on traffic flow detection

Abstract

The invention discloses a traffic signal control system based on traffic flow detection, and relates to the technical field of intelligent traffic. The system comprises a traffic flow acquisition module, a two-dimensional traffic flow analysis module, a dynamic phase splicing module and a signal control execution module. The traffic flow acquisition module acquires original data of each lane; the two-dimensional analysis module generates a dynamic traffic flow characteristic value through spatial dimension correlation analysis and time dimension fluctuation period extraction; the dynamic phase splicing module splits and recombines the fixed phase, generates a dynamic phase sequence in combination with the conflict matrix and the priority, and determines the duration according to the traffic flow ratio and the fluctuation period; and the signal control module executes control. The system solves the problems that a traditional system is insufficient in detection precision and rigid in timing, improves the intersection passing efficiency, and has both practicability and economical efficiency.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to a traffic signal control system based on vehicle flow detection. Background Art

[0002] With the continuous development of urban transportation, traffic congestion is becoming increasingly serious. Traffic signal control systems, as an important means of alleviating traffic congestion, have a direct impact on the efficiency of intersections. Currently, traditional traffic signal control systems are mostly based on fixed timing schemes or simply adjust signal durations based on simple traffic flow statistics, which has many shortcomings. Traditional systems typically only count traffic flow in a single lane independently, ignoring the correlation between adjacent lanes. For example, at an intersection with a left-turn lane and a through lane, a sudden increase in left-turn traffic can impact traffic flow in the through lane. However, traditional detection methods are unable to capture this correlation, resulting in low traffic flow detection accuracy. Traditional signal timing systems use a fixed phase division, such as east-west through-traffic, east-west left-turn, north-south through-traffic, and north-south left-turn. However, in real-world traffic, traffic flow in each lane fluctuates dynamically, and fixed phases cannot flexibly adapt to such changes. For example, during a certain period of time, traffic flow in the east-west through-traffic lane may be very low, while traffic flow in the east-west left-turn lane may be very high. However, since they are in the same fixed phase, the through-traffic lane will occupy part of the green light time, resulting in a waste of green light resources and affecting traffic efficiency at the intersection. Therefore, there is an urgent need for a traffic signal control system that can more accurately detect vehicle flow and flexibly adjust signal timing. Summary of the Invention

[0003] The object of the present invention is to provide a traffic signal control system based on vehicle flow detection to solve the problems raised in the above background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solutions: a traffic signal control system based on vehicle flow detection, comprising a vehicle flow collection module, a two-dimensional vehicle flow analysis module, a dynamic phase splicing module, and a signal control execution module, wherein the vehicle flow collection module collects raw vehicle flow data of each lane at an intersection through traffic detection equipment, wherein the traffic detection equipment includes but is not limited to cameras and radars; The dual-dimensional traffic flow analysis module is used to process the original traffic flow data obtained by the traffic flow collection module to obtain the dynamic traffic flow characteristic value of each lane. The specific working steps are as follows: Step 1: Obtain the original traffic flow data of each lane in a continuous time period collected by the traffic flow collection module, and the data is expressed as the number of vehicles per unit time; Step 2: Perform lane correlation analysis in the spatial dimension: Calculate the synchronization coefficient of traffic flow changes in adjacent lanes per unit time. The calculation formula for the synchronization coefficient is: ,in Indicates lane and lanes The synchronization coefficient, Indicates lane In the Traffic flow per unit time, Indicates lane In the Traffic flow per unit time, Indicates the number of units of time for statistics; the closer the synchronization coefficient is to 1, the higher the synchronization of traffic flow changes in the two lanes and the greater the correlation; Step 3: Extract traffic flow fluctuation cycles in the time dimension: Use a sliding window combined with Fourier transform, setting the sliding window duration to 5 minutes and the sliding step to 1 minute. Perform a Fourier transform on the traffic flow data within each sliding window to obtain the traffic flow distribution in the frequency domain. From this, extract the main fluctuation cycles, that is, the time intervals during which traffic flow exhibits periodic changes within this time period. Step 4: Generate a dynamic traffic flow characteristic value for each lane by combining the lane correlation coefficient in the spatial dimension and the traffic flow fluctuation period in the temporal dimension. The dynamic traffic flow characteristic value is a comprehensive parameter that includes traffic flow size, lane correlation coefficient, and fluctuation period, and is used to reflect the dynamic characteristics of lane traffic flow. The dynamic phase splicing module is used to split and reorganize the traditional fixed phases based on the dynamic traffic flow characteristic values ​​output by the two-dimensional traffic flow analysis module, generate a dynamic phase sequence, and determine the duration of each dynamic phase. The specific working steps are as follows: Step S1: receiving the dynamic traffic flow characteristic value of each lane output by the two-dimensional traffic flow analysis module; Step S2: Setting a lane traffic flow difference threshold and a correlation threshold: When the difference in traffic flow between the dynamic traffic flow characteristic values ​​of two lanes exceeds the traffic flow difference threshold, and the correlation coefficient between the two lanes is lower than the correlation threshold, it is determined that the traffic flow characteristics of the two lanes are significantly different. Step S3: split the traditional fixed phase: traverse the lanes in each fixed phase and split the lanes with large differences in traffic flow characteristics from the original phase; Step S4: Recombining the split lanes: Recombining the split lanes with lanes that have similar traffic flow characteristics in other fixed phases, that is, lanes whose traffic flow difference is less than the traffic flow difference threshold and whose correlation coefficient is higher than the correlation threshold, to form a new dynamic phase; Step S5, determining the duration of each dynamic phase: Based on the dynamic traffic flow characteristic value of each lane in the dynamic phase, the green light duration required for the dynamic phase is calculated using a weighted average method, where the weight is the proportion of the traffic flow of each lane to the total traffic flow of the dynamic phase. At the same time, based on the traffic flow fluctuation cycle, the green light duration of the corresponding dynamic phase is increased during the peak period of the fluctuation cycle and appropriately reduced during the valley period; Step S6: Generate a dynamic phase sequence: determine the order of the dynamic phases according to the conflict relationship between the dynamic phases; The signal control execution module is used to receive the dynamic phase sequence and the duration of each dynamic phase generated by the dynamic phase splicing module, control the on and off status of the traffic light, and realize signal control of the intersection traffic.

[0005] Preferably, the specific implementation logic of generating the dynamic traffic flow characteristic value of each lane in step 4 is as follows: Step 4.1. Establish a three-dimensional feature matrix: Use lanes as row indexes and traffic volume, lane correlation coefficient, and fluctuation period as column indexes to construct the initial feature matrix. ,in Indicates lane The traffic volume, that is, the number of vehicles per unit time The average value of Indicates lane The average synchronization coefficient with the adjacent lanes is The average value of For lanes adjacent lanes, Indicates lane The main fluctuation cycle of Step 4.2, feature standardization: standardize the three-dimensional feature matrix to eliminate the dimension effect. The standardization formula is: ,in , Representation matrix Middle All elements of the column, is the standardized eigenvalue; Step 4.3, dynamic weight allocation: Use the improved entropy method to calculate the dynamic weight of each feature dimension. First calculate Information entropy of features , ,in is the total number of lanes, is the minimum value to avoid zero value in logarithmic operation; then calculate the weight : , the smaller the information entropy, the higher the discreteness of the feature and the greater the weight; Step 4.4, weighted fusion to generate feature value: weighted sum of the normalized feature value and the corresponding weight to obtain the dynamic traffic flow feature value of each lane : .

[0006] Preferably, the specific implementation logic of determining the length of each dynamic phase in step S5 is as follows: Step S5.1, calculate the basic green light duration: set the dynamic phase contains a set of lanes , the traffic flow size of lane is , wherein , the total traffic flow of dynamic phase is , the weight of lane is , and the calculation formula of the basic green light duration is: , wherein is the dynamic traffic flow feature value of lane , and is the basic conversion coefficient; Step S5.2, determine the fluctuation period adjustment coefficient: set the average fluctuation period of each lane in dynamic phase is , that is , take the average value of the fluctuation periods of each lane, and the relative position of the current time in the fluctuation period is , wherein ; the adjustment coefficient is calculated using the improved sine function , wherein is the fluctuation influence factor, wherein is set according to the fluctuation intensity of the intersection traffic flow; when is in the first half of the period, the value is positive, ; when in the second half, the value is negative, ; Step S5.3, calculate the final green light duration: , and limit it to the minimum green light duration and the maximum green light duration , that is .

[0007] Preferably, the specific implementation steps of generating the dynamic phase sequence in step S6 are as follows: Step S6.1, constructing a conflict matrix: perform pairwise combination judgment on all dynamic phases. If there is an intersection conflict between the driving directions of the lanes corresponding to the two dynamic phases (e.g., one phase contains an east-west through lane, and the other phase contains a south-to-north left-turn lane, and the two traffic paths intersect), it is marked as "conflict" in the conflict matrix, represented by the number 1; if there is no intersection conflict between the driving directions of the lanes corresponding to the two dynamic phases (e.g., one phase contains an east-to-west through lane, and the other phase contains a west-to-east through lane, and the two traffic paths are parallel and do not intersect), it is marked as "compatible", represented by the number 0; the conflict matrix is ​​a The square array, is the total number of dynamic phases, where the matrix elements Indicates dynamic phase and dynamic phase conflicting relationships, Indicates conflict, Indicates compatibility; Step S6.2, calculate the dynamic phase priority: calculate the priority based on the final green light duration of the dynamic phase and the dynamic traffic flow characteristic value. The priority calculation formula is: ,in Dynamic phase The final green light duration, Dynamic phase The collection of lanes contained, is the number of lanes in the set, is the weight coefficient, and , which is used to balance the impact of green light duration and traffic flow characteristic value on priority; the larger the priority value, the more urgent the traffic demand in the dynamic phase; Step S6.3, generating an initial phase sequence: sorting the dynamic phases in descending order of priority to form an initial dynamic phase sequence; Step S6.4, adjust the phase sequence to eliminate conflicts: traverse the initial phase sequence, for two adjacent dynamic phases, if the conflict matrix is ​​marked as "conflict", keep the order unchanged; if it is marked as "compatible", determine whether it can be merged into a combined phase. If the merger can reduce the total signal cycle length, that is, the combined phase length takes the maximum value of the two dynamic phase lengths, then merge, otherwise keep the order unchanged; after the above adjustment, a conflict-free and efficient dynamic phase sequence is finally formed.

[0008] Preferably, the specific implementation steps of adjusting the phase sequence to eliminate the conflict in step S6.4 are as follows: Step S6.4.1, initialize the adjustment sequence: take the initial phase sequence as the sequence to be adjusted, denoted as ,in to is the dynamic phase after initial sorting; Step S6.4.2, set the traversal pointer: let the pointer , start traversing from the first phase of the sequence; Step S6.4.3, determine the adjacent phase relationship: when When the current phase and the next phase , query conflict matrix If the value of , then the pointer Increment by 1 and continue to traverse the next set of adjacent phases; if , then execute step S6.4.4; Step S6.4.4, assess the feasibility of the merger: calculate the total duration of the two phases before the merger , the duration of the combined phase after merging ;like , then the merger is deemed feasible; otherwise, the merger is deemed infeasible; Step S6.4.5, perform the merge or keep the order: If the merge is feasible, and Merge into combined phase , which lasts for , and use it in the sequence to be adjusted replace and , sequence length minus 1, pointer Remain unchanged and continue to check the relationship between the new combined phase and the next phase; if the merger is not feasible, the pointer Increment by 1, keeping the order of the two phases unchanged; Step S6.4.6, repeat the adjustment: when When , a round of traversal is completed; if there is a phase merging operation in this round of traversal, the pointer is reset to 1, and steps S6.4.3 to S6.4.5 are performed again on the updated sequence; if there is no merging operation in this round of traversal, the adjustment is stopped; Step S6.4.7, output the final sequence: After the above iterative adjustment, the sequence to be adjusted is a dynamic phase sequence with no conflict and the optimal total cycle length.

[0009] Compared with existing technologies, this invention achieves the following benefits: It improves traffic flow detection accuracy: The dual-dimensional traffic flow analysis module breaks through the traditional single-lane independent statistical model and calculates the synchronization coefficient of traffic flow changes in adjacent lanes in the spatial dimension, capturing the traffic flow correlation between lanes. Simultaneously, in the temporal dimension, a sliding window combined with Fourier transform is used to extract traffic flow fluctuation cycles, comprehensively reflecting the dynamic changes in traffic flow. By constructing a three-dimensional feature matrix, performing standardization processing, and assigning dynamic weights using an improved entropy method, the generated dynamic traffic flow characteristic values ​​accurately depict the size, correlation, and periodicity of traffic flow in each lane, providing reliable data support for subsequent signal timing.

[0010] Dynamically optimize signal timing: The dynamic phase splicing module splits and reorganizes traditional fixed phases based on dynamic traffic flow characteristics. By setting traffic flow difference thresholds and correlation thresholds, lanes with significantly different characteristics are split and reassembled into dynamic phases, avoiding the waste of green light resources under fixed phases. When determining the phase duration, the basic duration is calculated based on the weighted average of the lane traffic flow ratio. The fluctuation cycle adjustment coefficient is used to achieve dynamic adaptation, increasing the green light duration during peak hours and reducing it during off-peak hours, making signal timing more aligned with actual traffic flow needs.

[0011] Improve intersection efficiency: By constructing a conflict matrix to identify phase conflicts, and combining priority sorting and iterative merging and adjustment, a conflict-free dynamic phase sequence is generated to ensure orderly traffic flow. Merging and optimizing compatible phases further shortens the total signal cycle time, reducing vehicle wait times, effectively alleviating intersection congestion and improving overall traffic efficiency.

[0012] Taking into account both practicality and economy: The system's vehicle flow collection module and signal control execution module use mature cameras, radars and other detection equipment and conventional control logic to reduce development difficulty and cost; the core innovation module focuses on vehicle flow analysis and dynamic phase adjustment, ensuring technological advancement while facilitating practical promotion and application, and is suitable for traffic signal control scenarios at various urban intersections. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 Schematic diagram of the system structure of the present invention; Figure 2 This is a schematic diagram of the workflow of the dual-dimensional traffic flow analysis module of the present invention; Figure 3 Schematic diagram of the workflow of the dynamic phase splicing module of the present invention. DETAILED DESCRIPTION

[0014] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0015] See also Figure 1-3 The present invention provides a technical solution: a traffic signal control system based on vehicle flow detection, comprising a vehicle flow collection module, a two-dimensional vehicle flow analysis module, a dynamic phase splicing module, and a signal control execution module. The vehicle flow collection module collects raw vehicle flow data of each lane at an intersection through traffic detection equipment, wherein the traffic detection equipment includes but is not limited to cameras and radars. The specific vehicle flow collection module is the data input core of the traffic signal control system. Through traffic detection equipment such as cameras and radar, it collects raw traffic flow data for each lane of the intersection in real time over a continuous period of time. The data is expressed as the number of vehicles per unit time (such as 10 seconds). Its function is to provide basic data support for the subsequent two-dimensional traffic flow analysis module, ensuring that the system can accurately analyze and adjust signal timing based on actual traffic conditions. This module uses mature traffic detection technology to achieve real-time and accurate perception of traffic flow in each lane, which is the prerequisite for the intelligent control of the entire system. The two-dimensional traffic flow analysis module is used to process the original traffic flow data obtained by the traffic flow collection module to obtain the dynamic traffic flow characteristic value of each lane. The specific working steps are as follows: Step 1: Obtain the original traffic flow data of each lane in a continuous time period collected by the traffic flow collection module, and the data is expressed as the number of vehicles per unit time; Step 2: Perform lane correlation analysis in the spatial dimension: Calculate the synchronization coefficient of traffic flow changes in adjacent lanes per unit time. The calculation formula for the synchronization coefficient is: ,in Indicates lane and lanes The synchronization coefficient, Indicates lane In the Traffic flow per unit time, Indicates lane In the Traffic flow per unit time, Indicates the number of units of time for statistics; the closer the synchronization coefficient is to 1, the higher the synchronization of traffic flow changes in the two lanes and the greater the correlation; Step 3: Extract traffic flow fluctuation cycles in the time dimension: Use a sliding window combined with Fourier transform, setting the sliding window duration to 5 minutes and the sliding step to 1 minute. Perform a Fourier transform on the traffic flow data within each sliding window to obtain the traffic flow distribution in the frequency domain. From this, extract the main fluctuation cycles, that is, the time intervals during which traffic flow exhibits periodic changes within this time period. Step 4: Combine the lane correlation in the spatial dimension and the traffic flow fluctuation period in the temporal dimension to generate a dynamic traffic flow characteristic value for each lane. The dynamic traffic flow characteristic value is a comprehensive parameter that includes traffic flow size, lane correlation coefficient, and fluctuation period, and is used to reflect the dynamic characteristics of lane traffic flow. The specific implementation logic is as follows: Step 4.1. Establish a three-dimensional feature matrix: Use lanes as row indexes and traffic volume, lane correlation coefficient, and fluctuation period as column indexes to construct the initial feature matrix. ,in Indicates lane The traffic volume, that is, the number of vehicles per unit time The average value of Indicates lane The average synchronization coefficient with adjacent lanes is The average value of For lanes adjacent lanes, Indicates lane The main fluctuation cycle of Step 4.2, feature standardization: standardize the three-dimensional feature matrix to eliminate the dimension effect. The standardization formula is: ,in , Representation matrix Middle All elements of the column, is the standardized eigenvalue; Step 4.3, dynamic weight allocation: Use the improved entropy method to calculate the dynamic weight of each feature dimension. First calculate Information entropy of features , ,in is the total number of lanes, is the minimum value to avoid zero value in logarithmic operation; then calculate the weight : , the smaller the information entropy, the higher the discreteness of the feature and the greater the weight; Step 4.4: Weighted fusion to generate eigenvalues: Perform weighted summation of the standardized eigenvalues ​​and the corresponding weights to obtain the dynamic traffic flow eigenvalues ​​of each lane. : The dynamic phase splicing module is used to split and reorganize the traditional fixed phases based on the dynamic traffic flow characteristic values ​​output by the two-dimensional traffic flow analysis module, generate a dynamic phase sequence, and determine the duration of each dynamic phase. The specific working steps are as follows: Step S1: receiving the dynamic traffic flow characteristic value of each lane output by the two-dimensional traffic flow analysis module; Step S2: Setting a lane traffic flow difference threshold and a correlation threshold: When the difference in traffic flow between the dynamic traffic flow characteristic values ​​of two lanes exceeds the traffic flow difference threshold, and the correlation coefficient between the two lanes is lower than the correlation threshold, it is determined that the traffic flow characteristics of the two lanes are significantly different. Step S3: Split the traditional fixed phase: traverse the lanes in each fixed phase and split the lanes with large differences in traffic flow characteristics from the original phase; Step S4: Recombining the split lanes: Recombining the split lanes with lanes that have similar traffic flow characteristics in other fixed phases, that is, lanes whose traffic flow difference is less than the traffic flow difference threshold and whose correlation coefficient is higher than the correlation threshold, to form a new dynamic phase; Step S5: Determine the duration of each dynamic phase: Based on the dynamic traffic flow characteristic values ​​of each lane in the dynamic phase, a weighted average is used to calculate the green light duration required for the dynamic phase. The weight is the proportion of the traffic flow of each lane to the total traffic flow of the dynamic phase. At the same time, based on the traffic flow fluctuation cycle, the green light duration of the corresponding dynamic phase is increased during the peak period of the fluctuation cycle and appropriately reduced during the valley period. The specific implementation logic is as follows: Step S5.1, calculate the basic green light duration: set the dynamic phase Contains lane collection ,Lane The traffic volume is ,in , dynamic phase The total traffic volume is , then the lane The weight of , basic green light duration The calculation formula is: ,in For lanes The dynamic traffic flow characteristic value, is the basic conversion factor; Step S5.2, determine the fluctuation period adjustment coefficient: set the dynamic phase The average fluctuation period of each lane is ,Right now Take the fluctuation period of each lane The average value of the current moment is the relative position in the fluctuation cycle. ,in ; Use improved sine function to calculate the adjustment coefficient ,in is the volatility factor, where , set according to the traffic flow fluctuation intensity at the intersection; when In the first half of the cycle, The value is positive, ; In the second half, The value is negative, ; Step S5.3, calculate the final green light duration: , and limited to the minimum green light duration and maximum green light duration ,Right now .

[0016] Step S6: Generate a dynamic phase sequence: Determine the order of the dynamic phases according to the conflict relationship between the dynamic phases. The specific implementation steps are as follows: Step S6.1, constructing a conflict matrix: perform pairwise combination judgment on all dynamic phases. If there is an intersection conflict between the driving directions of the lanes corresponding to the two dynamic phases (e.g., one phase contains an east-west through lane, and the other phase contains a south-to-north left-turn lane, and the two traffic paths intersect), it is marked as "conflict" in the conflict matrix, represented by the number 1; if there is no intersection conflict between the driving directions of the lanes corresponding to the two dynamic phases (e.g., one phase contains an east-to-west through lane, and the other phase contains a west-to-east through lane, and the two traffic paths are parallel and do not intersect), it is marked as "compatible", represented by the number 0; the conflict matrix is ​​a The square array, is the total number of dynamic phases, where the matrix elements Indicates dynamic phase and dynamic phase conflicting relationships, Indicates conflict, Indicates compatibility; Step S6.2, calculate the dynamic phase priority: calculate the priority based on the final green light duration of the dynamic phase and the dynamic traffic flow characteristic value. The priority calculation formula is: ,in Dynamic phase The final green light duration, Dynamic phase The collection of lanes contained, is the number of lanes in the set, is the weight coefficient, and , which is used to balance the impact of green light duration and traffic flow characteristic value on priority; the larger the priority value, the more urgent the traffic demand in the dynamic phase; Step S6.3, generating an initial phase sequence: sorting the dynamic phases in descending order of priority to form an initial dynamic phase sequence; Step S6.4, adjust the phase sequence to eliminate conflicts: traverse the initial phase sequence. For two adjacent dynamic phases, if the conflict matrix is ​​marked as "conflicting", the order is maintained unchanged. If it is marked as "compatible", determine whether they can be merged into a combined phase. If the merger can reduce the total signal cycle length, that is, the combined phase length takes the maximum of the two dynamic phase lengths, then merge, otherwise keep the order unchanged. After the above adjustments, a conflict-free and efficient dynamic phase sequence is finally formed. The specific implementation steps are as follows: Step S6.4.1, initialize the adjustment sequence: take the initial phase sequence as the sequence to be adjusted, denoted as ,in to is the dynamic phase after initial sorting; Step S6.4.2, set the traversal pointer: let the pointer , start traversing from the first phase of the sequence; Step S6.4.3, determine the adjacent phase relationship: when When the current phase and the next phase , query conflict matrix If the value of , then the pointer Increment by 1 and continue to traverse the next set of adjacent phases; if , then execute step S6.4.4; Step S6.4.4, assess the feasibility of the merger: calculate the total duration of the two phases before the merger , the duration of the combined phase after merging ;like , then the merger is deemed feasible; otherwise, the merger is deemed infeasible; Step S6.4.5, perform the merge or keep the order: If the merge is feasible, and Merge into combined phase , which lasts for , and use it in the sequence to be adjusted replace and , sequence length minus 1, pointer Remain unchanged and continue to check the relationship between the new combined phase and the next phase; if the merger is not feasible, the pointer Increment by 1, keeping the order of the two phases unchanged; Step S6.4.6, repeat the adjustment: when When , a round of traversal is completed; if there is a phase merging operation in this round of traversal, the pointer is reset to 1, and steps S6.4.3 to S6.4.5 are performed again on the updated sequence; if there is no merging operation in this round of traversal, the adjustment is stopped; Step S6.4.7, output the final sequence: After the above iterative adjustment, the sequence to be adjusted is a dynamic phase sequence with no conflict and the optimal total cycle length.

[0017] The signal control execution module is used to receive the dynamic phase sequence and the duration of each dynamic phase generated by the dynamic phase splicing module, control the on and off status of the traffic light, and realize signal control of the intersection traffic.

[0018] The following is explained with reference to specific embodiments: Implementation scenario: Take a typical intersection as an example. This intersection includes four directions: east, west, south, and north. Each direction has a through lane, a left-turn lane, and a right-turn lane, for a total of 12 lanes. The traffic flow collection module uses a combination of high-definition cameras and radar to collect the number of vehicles in each lane per unit time (10 seconds).

[0019] Module Implementation: Traffic Flow Collection Module: Collects raw traffic flow data for 12 lanes during a continuous time period, such as the morning rush hour from 7:30 AM to 8:30 AM. For example, the traffic flow data collected for the eastbound through lane over multiple 10-second time periods might be 15, 18, 20, 17, and so on; the data for the eastbound left-turn lane might be 8, 10, 12, 9, and so on.

[0020] Two-dimensional traffic flow analysis module Step 1: Receive the original traffic flow data of the 12 lanes in the above continuous time period collected by the traffic flow collection module.

[0021] Step 2: Perform lane correlation analysis in the spatial dimension. ) and the eastbound left turn lane (lane ) as an example, calculate its synchronization coefficient. Assume that the number of statistical units of time , through the formula Calculate and obtain the synchronization coefficient of the two lanes , indicating that the synchronization of traffic flow changes in the two lanes is low and the correlation is small. Similarly, calculate the synchronization coefficient between other adjacent lanes; Step 3: Extract traffic flow fluctuation cycles in the time dimension. Set the sliding window duration to 5 minutes and the sliding step size to 1 minute. Perform a Fourier transform on the traffic flow data within each sliding window. For example, in the window from 7:30 to 7:35, traffic flow in the eastbound through lane exhibits a clear periodic distribution in the frequency domain. The main fluctuation period is extracted to be 3 minutes, indicating that traffic flow in this lane exhibits periodic changes every 3 minutes.

[0022] Step 4: Generate dynamic traffic flow characteristic values ​​for each lane; Step 4.1: Create a three-dimensional feature matrix M. For the eastbound through lane (lane ), is the average value of traffic volume per unit time, which is calculated to be 18; is the average synchronization coefficient with the adjacent lanes (eastbound left-turn lane and eastbound right-turn lane). Assuming that the synchronization coefficient between the eastbound through lane and the eastbound right-turn lane is 0.6, then ; Its main fluctuation period is 3 minutes. Similarly, the three-dimensional feature matrix of the other 11 lanes is constructed; Step 4.2: Feature standardization. Take the column of traffic volume ( ) as an example, assuming the matrix The minimum value of the first column is 5 and the maximum value is 25. , then after standardization Similarly, the lane correlation coefficient and fluctuation period are standardized. Step 4.3: Dynamic weight assignment. Assume that the total number of lanes , calculate the information entropy of the first feature (traffic volume) , through the formula Calculate and obtain ; Similarly, we can calculate , . Then calculate the weight ; ; ; Step 4.4: Weighted fusion generates characteristic values. Dynamic traffic flow characteristic values ​​of the eastbound through lane , assuming , ,but Similarly, calculate the dynamic traffic flow characteristic values ​​of the other 11 lanes; Dynamic phase splicing module: Step S1: Receive the dynamic traffic flow characteristic values ​​of 12 lanes output by the two-dimensional traffic flow analysis module.

[0023] Step S2: Set the lane traffic flow difference threshold to 0.2 and the correlation threshold to 0.5.

[0024] Step S3: Split traditional fixed phases. For traditional fixed phases such as the eastbound through and left-turn phases, the eastbound left-turn lane is separated from the fixed phase because the calculated difference in traffic volume between the eastbound through lane and the left-turn lane exceeds 0.2, and the correlation coefficient (0.3) is less than 0.5. Therefore, the traffic flow characteristics of the two lanes are significantly different. Similarly, the splitting process is repeated for other fixed phases.

[0025] Step S4: Reorganize the split lanes. Compare the split eastbound left-turn lane with the southbound left-turn lane. If the difference in traffic volume between the two is less than 0.2 and the correlation coefficient is higher than 0.5, reorganize them to form a new dynamic phase. Similarly, complete the reorganization of other split lanes to form a dynamic phase 、 wait.

[0026] Step S5: Determine the duration of each dynamic phase: Step S5.1: Calculate the basic green light duration. (including the eastbound left-turn lane and the southbound left-turn lane) as an example, the lane Traffic volume (eastbound left-turn lane) ,Lane Traffic volume in the southbound left-turn lane , dynamic phase Total traffic volume .Lane Weight ,Lane Weight Known , , basic conversion factor seconds, the basic green light duration Second; Step S5.2: Determine the fluctuation period adjustment coefficient. Dynamic phase The average fluctuation period of each lane , the fluctuation period of the eastbound left-turn lane is 4 minutes, and the fluctuation period of the southbound left-turn lane is 5 minutes, then Minutes. The current moment is in the relative position of the fluctuation cycle Minutes (in the first half of the cycle), volatility impact factor Adjustment factor .

[0027] Step S5.3: Calculate the final green light duration. Seconds, set the minimum green light duration Seconds, maximum green light duration Seconds, finally Similarly, calculate the final green light duration of other dynamic phases.

[0028] Step S6: Generate dynamic phase sequence: Step S6.1: Construct conflict matrix. 、 、 etc. to make pairwise combination judgments, such as (turn left when heading east and south) and There is a cross conflict (eastbound and westbound straight) in the conflict matrix ; and (Northbound and westbound right turn) No intersection conflict, .

[0029] Step S6.2: Calculate the dynamic phase priority. For example, the final green light duration seconds, assuming that the total green light duration of all dynamic phases is 100 seconds, The sum of the dynamic traffic flow characteristic values ​​in the included lane set is , the number of lanes is 2. Weight coefficient , then the priority Similarly, the priorities of other dynamic phases are calculated.

[0030] Step S6.3: Generate the initial phase sequence. Sort by priority from high to low, and the initial phase sequence is: ; Step S6.4: Adjust the phase sequence to eliminate the conflict. Traverse the initial sequence, and The conflict matrix is ​​marked as 1, and the order remains unchanged; and The conflict matrix is ​​marked as 0 to evaluate the feasibility of the merger. Total time before the merger , combined phase duration after merging ,like , then merge and The phases are combined, otherwise the order is maintained. After multiple rounds of iterative adjustments, a conflict-free and efficient dynamic phase sequence is finally formed.

[0031] The Signal Control Execution Module receives the dynamic phase sequence and durations generated by the Dynamic Phase Splicing Module and controls the traffic lights to flash according to this sequence and duration, achieving precise signal control at the intersection. In the aforementioned morning rush hour scenario, this system's control ensures more orderly traffic flow across lanes at the intersection, effectively alleviating traffic congestion.

[0032] This invention discloses a traffic signal control system based on traffic flow detection, relating to the field of intelligent transportation technology. The system includes a traffic flow acquisition module, a two-dimensional traffic flow analysis module, a dynamic phase splicing module, and a signal control execution module. The traffic flow acquisition module collects raw traffic flow data from each lane at an intersection using cameras, radar, and other equipment. The two-dimensional traffic flow analysis module generates dynamic traffic flow characteristic values, including traffic flow size, correlation coefficient, and fluctuation period, through lane correlation analysis in the spatial dimension (calculating synchronization coefficients) and traffic flow fluctuation period extraction in the temporal dimension (using a sliding window combined with Fourier transform). The dynamic phase splicing module splits and reorganizes traditional fixed phases based on these characteristic values. It identifies phase conflicts using a conflict matrix, prioritizes them, and iteratively merges and adjusts them to generate a dynamic phase sequence. The duration of each phase is dynamically determined based on traffic flow proportions and fluctuation periods. The signal control execution module controls the on / off of traffic lights based on the dynamic phase sequence and duration. This invention solves the problems of insufficient traffic flow detection accuracy and rigid signal timing in traditional systems, improving intersection traffic efficiency and achieving both practicality and cost-effectiveness.

[0033] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A traffic signal control system based on vehicle flow detection, characterized in that: It includes a traffic flow collection module, a two-dimensional traffic flow analysis module, a dynamic phase splicing module and a signal control execution module. The traffic flow collection module collects the original traffic flow data of each lane of the intersection through traffic detection equipment, where the traffic detection equipment includes but is not limited to cameras and radars; The dual-dimensional traffic flow analysis module is used to process the original traffic flow data obtained by the traffic flow collection module to obtain the dynamic traffic flow characteristic value of each lane. The specific working steps are as follows: Step 1: Obtain the original traffic flow data of each lane in a continuous time period collected by the traffic flow collection module, and the data is expressed as the number of vehicles per unit time; Step 2: Perform lane correlation analysis in the spatial dimension: Calculate the synchronization coefficient of traffic flow changes in adjacent lanes per unit time. The calculation formula for the synchronization coefficient is: ,in Indicates lane and lanes The synchronization coefficient, Indicates lane In the Traffic flow per unit time, Indicates lane In the Traffic flow per unit time, Indicates the number of units of time for statistics; The closer the synchronization coefficient is to 1, the higher the synchronization of traffic flow changes in the two lanes is and the greater the correlation is; Step 3: Extract traffic flow fluctuation cycles in the time dimension: Use a sliding window combined with Fourier transform, setting the sliding window duration to 5 minutes and the sliding step to 1 minute. Perform a Fourier transform on the traffic flow data within each sliding window to obtain the traffic flow distribution in the frequency domain. From this, extract the main fluctuation cycles, that is, the time intervals during which traffic flow exhibits periodic changes within this time period. Step 4: Generate a dynamic traffic flow characteristic value for each lane by combining the lane correlation coefficient in the spatial dimension and the traffic flow fluctuation period in the temporal dimension. The dynamic traffic flow characteristic value is a comprehensive parameter that includes traffic flow size, lane correlation coefficient, and fluctuation period, and is used to reflect the dynamic characteristics of lane traffic flow. The dynamic phase splicing module is used to split and reorganize the traditional fixed phases based on the dynamic traffic flow characteristic values ​​output by the two-dimensional traffic flow analysis module, generate a dynamic phase sequence, and determine the duration of each dynamic phase. The specific working steps are as follows: Step S1: receiving the dynamic traffic flow characteristic value of each lane output by the two-dimensional traffic flow analysis module; Step S2: Setting a lane traffic flow difference threshold and a correlation threshold: When the difference in traffic flow between the dynamic traffic flow characteristic values ​​of two lanes exceeds the traffic flow difference threshold, and the correlation coefficient between the two lanes is lower than the correlation threshold, it is determined that the traffic flow characteristics of the two lanes are significantly different. Step S3: split the traditional fixed phase: traverse the lanes in each fixed phase and split the lanes with large differences in traffic flow characteristics from the original phase; Step S4: Recombining the split lanes: Recombining the split lanes with lanes that have similar traffic flow characteristics in other fixed phases, that is, lanes whose traffic flow difference is less than the traffic flow difference threshold and whose correlation coefficient is higher than the correlation threshold, to form a new dynamic phase; Step S5, determining the duration of each dynamic phase: Based on the dynamic traffic flow characteristic value of each lane in the dynamic phase, the green light duration required for the dynamic phase is calculated using a weighted average method, where the weight is the proportion of the traffic flow of each lane to the total traffic flow of the dynamic phase. At the same time, based on the traffic flow fluctuation cycle, the green light duration of the corresponding dynamic phase is increased during the peak period of the fluctuation cycle and appropriately reduced during the valley period; Step S6: Generate a dynamic phase sequence: determine the order of the dynamic phases according to the conflict relationship between the dynamic phases; The signal control execution module is used to receive the dynamic phase sequence and the duration of each dynamic phase generated by the dynamic phase splicing module, control the on and off status of the traffic light, and realize signal control of the intersection traffic.

2. A traffic signal control system based on vehicle flow detection according to claim 1, characterized in that: The specific implementation logic of generating the dynamic traffic flow characteristic value of each lane in step 4 is as follows: Step 4.

1. Establish a three-dimensional feature matrix: Use lanes as row indexes and traffic volume, lane correlation coefficient, and fluctuation period as column indexes to construct the initial feature matrix. ,in Indicates lane The traffic volume, that is, the number of vehicles per unit time The average value of Indicates lane The average synchronization coefficient with adjacent lanes is The average value of For lanes adjacent lanes, Indicates lane The main fluctuation cycle of Step 4.2, feature standardization: standardize the three-dimensional feature matrix to eliminate the dimension effect. The standardization formula is: ,in , Representation matrix Middle All elements of the column, is the standardized eigenvalue; Step 4.3, dynamic weight allocation: Use the improved entropy method to calculate the dynamic weight of each feature dimension. First calculate Information entropy of features , ,in is the total number of lanes, is the minimum value to avoid zero value in logarithmic operation; then calculate the weight : , the smaller the information entropy, the higher the discreteness of the feature and the greater the weight; Step 4.4: Weighted fusion to generate eigenvalues: Perform weighted summation of the standardized eigenvalues ​​and the corresponding weights to obtain the dynamic traffic flow eigenvalues ​​of each lane. : .

3. The traffic signal control system based on vehicle flow detection according to claim 1, characterized in that: The specific implementation logic of determining the duration of each dynamic phase in step S5 is as follows: Step S5.1, calculate the basic green light duration: set the dynamic phase Contains lane collection ,Lane The traffic volume is ,in , dynamic phase The total traffic volume is , then the lane The weight is , basic green light duration The calculation formula is: ,in For lanes The dynamic traffic flow characteristic value, is the basic conversion factor; Step S5.2, determine the fluctuation period adjustment coefficient: set the dynamic phase The average fluctuation period of each lane is ,Right now Take the fluctuation period of each lane The average value of the current moment is the relative position in the fluctuation cycle. ,in ; Use the improved sine function to calculate the adjustment coefficient ,in is the volatility factor, where , set according to the traffic flow fluctuation intensity at the intersection; when In the first half of the cycle, The value is positive, ; In the second half, The value is negative, ; Step S5.3, calculate the final green light duration: , and limited to the minimum green light duration and maximum green light duration ,Right now .

4. The traffic signal control system based on vehicle flow detection according to claim 1, characterized in that: The specific implementation steps of generating the dynamic phase sequence in step S6 are as follows: Step S6.1, constructing a conflict matrix: perform pairwise combination judgment on all dynamic phases. If there is an intersection conflict between the driving directions of the lanes corresponding to the two dynamic phases (e.g., one phase contains an east-west through lane, and the other phase contains a south-to-north left-turn lane, and the two traffic paths intersect), it is marked as "conflict" in the conflict matrix, represented by the number 1; if there is no intersection conflict between the driving directions of the lanes corresponding to the two dynamic phases (e.g., one phase contains an east-to-west through lane, and the other phase contains a west-to-east through lane, and the two traffic paths are parallel and do not intersect), it is marked as "compatible" and represented by the number 0; the conflict matrix is ​​a The phalanx, is the total number of dynamic phases, where the matrix elements Indicates dynamic phase and dynamic phase conflicting relationships, Indicates conflict, Indicates compatibility; Step S6.2, calculate the dynamic phase priority: calculate the priority based on the final green light duration of the dynamic phase and the dynamic traffic flow characteristic value. The priority calculation formula is: ,in Dynamic phase The final green light duration, Dynamic phase The collection of lanes contained, is the number of lanes in the set, is the weight coefficient, and , which is used to balance the impact of green light duration and traffic flow characteristic value on priority; the larger the priority value, the more urgent the traffic demand in the dynamic phase; Step S6.3, generating an initial phase sequence: sorting the dynamic phases in descending order of priority to form an initial dynamic phase sequence; Step S6.4, adjust the phase sequence to eliminate conflicts: traverse the initial phase sequence. For two adjacent dynamic phases, if the conflict matrix is ​​marked as "conflict", the order is kept unchanged; if it is marked as "compatible", determine whether they can be merged into a combined phase. If the merger can reduce the total signal cycle length, that is, the combined phase length takes the maximum of the two dynamic phase lengths, then merge, otherwise keep the order unchanged; after the above adjustment, a conflict-free and efficient dynamic phase sequence is finally formed.

5. The traffic signal control system based on vehicle flow detection according to claim 4, characterized in that: The specific implementation steps of step S6.4 of adjusting the phase sequence to eliminate the conflict are as follows: Step S6.4.1, initialize the adjustment sequence: take the initial phase sequence as the sequence to be adjusted, denoted as ,in to is the dynamic phase after initial sorting; Step S6.4.2, set the traversal pointer: let the pointer , start traversing from the first phase of the sequence; Step S6.4.3, determine the adjacent phase relationship: when When the current phase and the next phase , query conflict matrix If , then the pointer Increment by 1 and continue to traverse the next set of adjacent phases; if , then execute step S6.4.4; Step S6.4.4, assess the feasibility of the merger: calculate the total duration of the two phases before the merger , the duration of the combined phase after merging ;like , then the merger is deemed feasible; Otherwise, the merger is deemed infeasible; Step S6.4.5, perform the merge or keep the order: If the merge is feasible, and Merge into combined phase , which lasts for , and use it in the sequence to be adjusted replace and , sequence length minus 1, pointer Remain unchanged and continue to check the relationship between the new combined phase and the next phase; if the merger is not feasible, the pointer Increase by 1, keeping the order of the two phases unchanged; Step S6.4.6, repeat the adjustment: when When , a round of traversal is completed; if there is a phase merging operation in this round of traversal, the pointer is reset to 1, and steps S6.4.3 to S6.4.5 are performed again on the updated sequence; if there is no merging operation in this round of traversal, the adjustment is stopped; Step S6.4.7, output the final sequence: After the above iterative adjustment, the sequence to be adjusted is a dynamic phase sequence with no conflict and the optimal total cycle length.

Images