Traffic signal lamp control method and system based on edge calculation
Through edge computing devices, the red light duration of right-turn traffic lights is dynamically adjusted by using linear regression algorithm, which solves the problem of insufficient red light duration settings in the existing technology, and improves the setting effect of right-turn traffic lights and the optimization of traffic flow.
Patent Information
- Application Number
- CN202510374607.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-27
- Publication Date
- 2025-05-30
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The setting time of the existing right-turn traffic light is usually pre-set based on manual experience or expert knowledge. The actual application scenario cannot be considered, resulting in too long or too short red light time, resulting in wasted resources or poor setting effect.
The edge computing device analyzes the pedestrian data, non-motor vehicle data and motor vehicle data associated with the right turn of the vehicle in real time, and uses a linear regression analysis algorithm to output correction factors to dynamically adjust the red light duration of the right turn traffic light.
Through real-time data analysis and dynamic adjustment, the setting effect of right-turn traffic lights is improved, the safety hazards brought about by vehicles are reduced, traffic flow is optimized, and traffic efficiency is improved.
Smart Images
Figure CN120071646A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of signal lamp control, and particularly to a traffic signal lamp control method and system based on edge computing. Background Art
[0002] Traffic signal lamps are an important part of road traffic management, mainly used to direct and control the traffic order of motor vehicles, non-motor vehicles and pedestrians. They indicate passing, waiting or prohibited passage through light signals of different colors (usually red, yellow, and green), thereby improving road safety, reducing traffic accidents, optimizing traffic flow, and improving road traffic efficiency. In recent years, more and more cities have begun to set up right-turn traffic lights (i.e., dedicated right-turn signal lights), especially in intersections with heavy traffic flow, and they are used more frequently. The increase in right-turn traffic lights is mainly to improve the safety of pedestrians and non-motor vehicles, reduce traffic conflicts, optimize road order, and cooperate with the overall optimization of the intelligent transportation system.
[0003] However, the setting duration of existing right-turn traffic lights is usually preset based on manual experience or expert knowledge, that is, the right-turn traffic lights usually have a fixed red light duration. However, since the preset red light duration cannot consider the actual application scenario, when the red light duration is too long, it may cause waste of passing resources, resulting in congestion in the right-turn lane. When the red light duration is too short, it may lead to poor setting effect of the right-turn traffic lights.
[0004] Based on this, the present invention proposes a traffic signal lamp control method and system based on edge computing. After the edge computing device analyzes the pedestrian data, non-motor vehicle data, and motor vehicle data associated with vehicle right-turns in real time, it dynamically adjusts the red light duration of the right-turn traffic lights to further improve the setting effect of the right-turn traffic lights. Summary of the Invention
[0005] The purpose of the present invention is to provide a traffic signal lamp control method and system based on edge computing to solve the deficiencies in the background art.
[0006] To achieve the above purpose, the present invention provides the following technical solution: A traffic signal lamp control method based on edge computing, the control method includes the following steps: The control system regularly obtains the traffic flow information of the road. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, it judges whether it is necessary to control the right-turn red light to turn on. If it is necessary to turn on, it controls the right-turn red light to operate for a preset initial duration. During the operation of the right-turn red light, the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane are periodically obtained through a camera device. After substituting the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor; The red light duration is dynamically adjusted based on the comparison result between the correction factor and the correction threshold, and the operation of the right-turn red light is controlled by the dynamically adjusted red light duration.
[0007] In a preferred embodiment, during the operation of the right-turn red light, the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane are periodically obtained through a camera device. The sidewalk data includes the pedestrian flow index, the non-motor vehicle lane data includes the number of non-motor vehicles waiting, and the motor vehicle lane data includes the number of vehicles in the right-turn lane, the traffic flow of vehicles entering the lane, and the vehicle density at the lane entrance.
[0008] In a preferred embodiment, after substituting the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor, including the following steps: The edge computing device substitutes the pedestrian flow index, the number of non-motor vehicles waiting, the number of vehicles in the right-turn lane, the traffic flow of vehicles entering the lane, and the vehicle density at the lane entrance into the linear regression analysis algorithm. The algorithm expression is: , where is the correction factor, are respectively the pedestrian flow index, the number of non-motor vehicles waiting, the number of vehicles in the right-turn lane, the traffic flow of vehicles entering the lane, and the vehicle density at the lane entrance, is the regression coefficient, and the regression coefficient is greater than 0.
[0009] In a preferred embodiment, the red light duration is dynamically adjusted based on the comparison result between the correction factor and the correction threshold, and the operation of the right-turn red light is controlled by the dynamically adjusted red light duration, including the following steps: After obtaining the correction factor, the larger the correction factor, the more the right-turn red light duration needs to be increased. Compare the obtained correction factor with the preset correction threshold. The correction threshold is used to determine whether to dynamically adjust the initial red light duration. If the correction factor is less than or equal to the correction threshold, it is determined that there is no need to dynamically adjust the initial red light duration. If the correction factor is greater than the correction threshold, it is determined that the initial red light duration needs to be dynamically adjusted; When it is determined that the initial red light duration needs to be dynamically adjusted, the initial red light duration is dynamically adjusted by the correction factor.
[0010] In a preferred embodiment, the initial red light duration is dynamically adjusted by a correction factor, and the expression is: , where is the red light duration after dynamic adjustment, is the initial red light duration, is the correction factor.
[0011] In a preferred embodiment, the calculation logic of the pedestrian flow index is as follows: obtain the number of pedestrians waiting at the sidewalk intersection at the 3rd second before the end of the red light for the sidewalk, obtain the pedestrian density on the sidewalk at the 10th second of the green light for the sidewalk, perform normalization processing on the number of pedestrians waiting and the pedestrian density, map the value ranges of the number of pedestrians waiting and the pedestrian density to between [0, 1], obtain the normalized value of the number of pedestrians waiting and the normalized value of the pedestrian density, and sum the normalized value of the number of pedestrians waiting and the normalized value of the pedestrian density to obtain the pedestrian flow index.
[0012] In a preferred embodiment, the traffic flow information of the road is obtained regularly, and the traffic flow information includes the conflict point location, the speed of right-turning vehicles, the location of right-turning vehicles, the location of pedestrians, the walking speed of pedestrians, the location of non-motor vehicles, and the speed of non-motor vehicles.
[0013] In a preferred embodiment, the control system obtains the traffic flow information of the road regularly. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, it judges whether it is necessary to control the activation of the right-turn red light, including the following steps: The gap acceptance model outputs the right-turn conflict frequency of the road, including the following steps: Calculate the distances between right-turning vehicles and conflict points, between pedestrians and conflict points, and between non-motor vehicles and conflict points; Obtain the time durations for right-turning vehicles, pedestrians, and non-motor vehicles to reach the conflict points; Then calculate the pedestrian gap and the non-motor vehicle gap; Based on the comparison result between the pedestrian gap and the pedestrian gap threshold, judge whether there is a conflict of pedestrians passing through during this vehicle right-turn. Based on the comparison result between the non-motor vehicle gap and the non-motor vehicle gap threshold, judge whether there is a conflict of non-motor vehicles passing through during this vehicle right-turn; Record the number of pedestrian passing conflicts and the number of non-motor vehicle passing conflicts during the monitoring period, add the number of pedestrian passing conflicts and the number of non-motor vehicle passing conflicts to obtain the total number of conflicts, and divide the total number of conflicts by the monitoring duration to obtain the right-turn conflict frequency; After receiving the right-turn conflict frequency of the road through the gap acceptance model, compare the right-turn conflict frequency with the frequency threshold. The frequency threshold is used to determine whether to control the activation of the right-turn red light. If the right-turn conflict frequency is greater than the frequency threshold, it is determined that the activation of the right-turn red light needs to be controlled. If the right-turn conflict frequency is less than or equal to the frequency threshold, it is determined that the activation of the right-turn red light does not need to be controlled.
[0014] In a preferred embodiment, calculate the distances between the right-turning vehicle and the conflict point, the pedestrian and the conflict point, and the non-motor vehicle and the conflict point. The processing logic is as follows: calculate the distances between the right-turning vehicle and the conflict point, the pedestrian and the conflict point, and the non-motor vehicle and the conflict point respectively based on the positions of the conflict point, the right-turning vehicle, the pedestrian, and the non-motor vehicle. Obtain the time taken for the right-turning vehicle to reach the conflict point, the time taken for the pedestrian to reach the conflict point, and the time taken for the non-motor vehicle to reach the conflict point. The processing logic is as follows: obtain the time taken for the right-turning vehicle to reach the conflict point by dividing the distance between the right-turning vehicle and the conflict point by the speed of the right-turning vehicle, obtain the time taken for the pedestrian to reach the conflict point by dividing the distance between the pedestrian and the conflict point by the walking speed of the pedestrian, and obtain the time taken for the non-motor vehicle to reach the conflict point by dividing the distance between the non-motor vehicle and the conflict point by the speed of the non-motor vehicle. Calculate the pedestrian gap and the non-motor vehicle gap. The processing logic is as follows: the pedestrian gap is obtained by subtracting the time taken for the right-turning vehicle to reach the conflict point from the time taken for the pedestrian to reach the conflict point, and the non-motor vehicle gap is obtained by subtracting the time taken for the right-turning vehicle to reach the conflict point from the time taken for the non-motor vehicle to reach the conflict point. Determine whether there is a pedestrian passing conflict or a non-motor vehicle passing conflict during the right-turn of the vehicle. The processing logic is as follows: if the pedestrian gap is less than the pedestrian gap threshold, it is determined that there is a pedestrian passing conflict during the right-turn of the vehicle. If the pedestrian gap is greater than or equal to the pedestrian gap threshold, it is determined that there is no pedestrian passing conflict during the right-turn of the vehicle. If the non-motor vehicle gap is less than the non-motor vehicle gap threshold, it is determined that there is a non-motor vehicle passing conflict during the right-turn of the vehicle. If the non-motor vehicle gap is greater than or equal to the non-motor vehicle gap threshold, it is determined that there is no non-motor vehicle passing conflict during the right-turn of the vehicle.
[0015] The traffic signal control system based on edge computing includes an activation judgment module, an edge computing module, and a dynamic adjustment module. Activation judgment module: Regularly obtain the traffic flow information of the road. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, determine whether to control the activation of the right-turn red light. If activation is required, control the right-turn red light to operate for a preset initial duration. Edge computing module: During the operation of the right-turn red light, regularly obtain the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane through the camera device. After substituting the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor. Dynamic adjustment module: Dynamically adjust the red light duration according to the comparison result between the calibration factor and the calibration threshold, and control the operation of the right-turn red light through the dynamically adjusted red light duration.
[0016] In the above technical solution, the technical effects and advantages provided by the present invention are as follows: 1. During the operation of the right-turn red light, the present invention regularly obtains the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane through the camera device. After substituting the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a calibration factor, and dynamically adjusts the red light duration according to the comparison result between the calibration factor and the calibration threshold, and controls the operation of the right-turn red light through the dynamically adjusted red light duration. By analyzing the pedestrian data, non-motor vehicle data, and motor vehicle data associated with vehicle right-turn in real time through the edge computing device, the red light duration of the right-turn traffic lights is dynamically adjusted, further improving the setting effect of the right-turn traffic lights; 2. The present invention regularly obtains the traffic flow information of the road. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, it judges whether it is necessary to control the activation of the right-turn red light. If it is necessary to activate, it controls the right-turn red light to operate with a preset initial duration, which not only effectively reduces the safety hazards caused by vehicles rushing through, but also reduces the additional constraints on right-turn vehicles and improves the traffic efficiency. Description of the Drawings
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.
[0018] Figure 1 It is the method flow chart of the present invention. Detailed Embodiments
[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
[0020] Embodiment 1: Please refer to Figure 1 As shown, the traffic signal control method based on edge computing described in this embodiment includes the following steps: The control system regularly obtains the traffic flow information of the road. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, it determines whether to control the activation of the right-turn red light. If activation is required, it controls the right-turn red light to operate for a preset initial duration. During the operation of the right-turn red light, the camera device regularly obtains the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane. After the edge computing device substitutes the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor, and dynamically adjusts the red light duration based on the comparison result between the correction factor and the correction threshold, and controls the operation of the right-turn red light with the dynamically adjusted red light duration.
[0021] In this application, during the operation of the right-turn red light, the camera device regularly obtains the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane. After the edge computing device substitutes the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor, and dynamically adjusts the red light duration based on the comparison result between the correction factor and the correction threshold, and controls the operation of the right-turn red light with the dynamically adjusted red light duration. By the edge computing device analyzing the pedestrian data, non-motor vehicle data, and motor vehicle data associated with vehicle right-turns in real time, the red light duration of the right-turn traffic lights is dynamically adjusted, further improving the setting effect of the right-turn traffic lights.
[0022] In this application, by regularly obtaining the traffic flow information of the road, after the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, it determines whether to control the activation of the right-turn red light. If activation is required, it controls the right-turn red light to operate for a preset initial duration, which not only effectively reduces the safety hazards caused by vehicles cutting in line, but also reduces the additional constraints on right-turning vehicles and improves the traffic efficiency.
[0023] Embodiment 2: The control system regularly obtains the traffic flow information of the road. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, it determines whether to control the activation of the right-turn red light. If activation is required, it controls the right-turn red light to operate for a preset initial duration, including the following steps: In this application, the initial duration of the right-turn red light is set to a relatively small value, such as 5s - 8s.
[0024] Regularly obtain the traffic flow information of the road. The traffic flow information includes the conflict point location, right-turn vehicle speed, right-turn vehicle location, pedestrian location, pedestrian walking speed, non-motor vehicle location, and non-motor vehicle speed; The steps for the gap acceptance model to output the right-turn conflict frequency of the road are: Calculate the distances between the right-turning vehicle and the conflict point, between the pedestrian and the conflict point, and between the non-motor vehicle and the conflict point. Specifically: Calculate the distances between the right-turning vehicle and the conflict point, between the pedestrian and the conflict point, and between the non-motor vehicle and the conflict point respectively based on the positions of the conflict point, the right-turning vehicle, the pedestrian, and the non-motor vehicle; In this application, the conflict points on the road are pre-input into the control system manually. Usually, when the right-turn signal is green and the pedestrian signal is also green at the same time, the non-motor vehicles and pedestrians will move together. The intersection point between the right-turning vehicle and the non-motor vehicles and pedestrians before the right-turning vehicle enters the right-turning road is the conflict point.
[0025] Calculate the distances between the right-turning vehicle and the conflict point, between the pedestrian and the conflict point, and between the non-motor vehicle and the conflict point respectively based on the positions of the conflict point, the right-turning vehicle, the pedestrian, and the non-motor vehicle. The distance between the right-turning vehicle and the conflict point is obtained through the existing distance calculation formula, and the expression is: , where is the distance between the right-turning vehicle and the conflict point, is the position of the right-turning vehicle, is the position of the conflict point. The calculation methods for the distances between the pedestrian and the conflict point and between the non-motor vehicle and the conflict point are the same as above, and will not be elaborated in this application.
[0026] Obtain the time for the right-turning vehicle to reach the conflict point, the time for the pedestrian to reach the conflict point, and the time for the non-motor vehicle to reach the conflict point. Specifically: Obtain the time for the right-turning vehicle to reach the conflict point by dividing the distance between the right-turning vehicle and the conflict point by the speed of the right-turning vehicle, obtain the time for the pedestrian to reach the conflict point by dividing the distance between the pedestrian and the conflict point by the walking speed of the pedestrian, and obtain the time for the non-motor vehicle to reach the conflict point by dividing the distance between the non-motor vehicle and the conflict point by the speed of the non-motor vehicle; Then calculate the pedestrian gap and the non-motor vehicle gap. Specifically: The pedestrian gap is obtained by subtracting the time for the right-turning vehicle to reach the conflict point from the time for the pedestrian to reach the conflict point, and the non-motor vehicle gap is obtained by subtracting the time for the right-turning vehicle to reach the conflict point from the time for the non-motor vehicle to reach the conflict point; Judge whether there is a pedestrian passing through the conflict during this vehicle right-turn based on the comparison result between the pedestrian gap and the pedestrian gap threshold, and judge whether there is a non-motor vehicle passing through the conflict during this vehicle right-turn based on the comparison result between the non-motor vehicle gap and the non-motor vehicle gap threshold. Specifically: If the pedestrian gap is less than the pedestrian gap threshold, judge that there is a pedestrian passing through the conflict during this vehicle right-turn; if the pedestrian gap is greater than or equal to the pedestrian gap threshold, judge that there is no pedestrian passing through the conflict during this vehicle right-turn; if the non-motor vehicle gap is less than the non-motor vehicle gap threshold, judge that there is a non-motor vehicle passing through the conflict during this vehicle right-turn; if the non-motor vehicle gap is greater than or equal to the non-motor vehicle gap threshold, judge that there is no non-motor vehicle passing through the conflict during this vehicle right-turn; Record the number of pedestrian passing conflicts and the number of non-motor vehicle passing conflicts during the monitoring period. Add the number of pedestrian passing conflicts to the number of non-motor vehicle passing conflicts to obtain the total number of conflicts. Divide the total number of conflicts by the monitoring duration to get the right-turn conflict frequency.
[0027] The greater the right-turn conflict frequency, it indicates that during the current period of the road, due to the large volume of motor vehicles, non-motor vehicles, and pedestrians, the more times the right-turning vehicles affect the non-motor vehicles and pedestrians, and the more necessary it is to turn on the right-turn red light.
[0028] After the right-turn conflict frequency of the road is output through the gap acceptance model, compare the right-turn conflict frequency with the frequency threshold. The frequency threshold is used to determine whether to control the activation of the right-turn red light. If the right-turn conflict frequency is greater than the frequency threshold, it is determined that the right-turn red light needs to be controlled to be turned on. If the right-turn conflict frequency is less than or equal to the frequency threshold, it is determined that there is no need to control the activation of the right-turn red light.
[0029] There are the following parameters: Vehicle right-turn speed: Assume the average is 15 km / h (about 4.17 m / s).
[0030] Pedestrian walking speed: Assume the average is 1.4 m / s.
[0031] Non-motor vehicle speed: Assume the speed of electric vehicles, bicycles, etc. is about 4 m / s.
[0032] Intersection width (distance pedestrians need to pass through): 10 m.
[0033] Acceptable gap threshold (the minimum gap that drivers usually accept): For pedestrians > 3.5 seconds, for non-motor vehicles > 2.5 seconds.
[0034] Suppose at the current moment, a pedestrian is about to cross the sidewalk, and there is an electric vehicle approaching the intersection from behind. The driver needs to decide whether to turn right. The conflict point between the pedestrian and the right-turning vehicle is in the middle of the sidewalk, about 5 m away from the vehicle (assuming the vehicle has reached the intersection and is waiting to turn right). The time for the pedestrian to reach this point:
[0035] The non-motor vehicle is about 10 m away from the conflict point (from the side). The time for the non-motor vehicle to reach this point:
[0036] The right-turning vehicle needs to pass through 6 m (assuming the safe passing distance). The time to reach the conflict point is:
[0037] Pedestrian gap (Gap) = 3.57 s - 1.44 s = 2.13 s (less than 3.5 s, there is a conflict), non-motor vehicle gap (Gap) = 2.5 s - 1.44 s = 1.06 s (less than 2.5 s, there is a conflict), record all the conflict times within the monitoring time period.
[0038] Specifically: The right-turn red light is synchronized with the pedestrian green light: In many cases, when the pedestrian green light comes on, the right-turn red light also comes on. This helps to ensure that right-turning vehicles stop when pedestrians cross the road, avoiding conflicts with pedestrians. When the pedestrian green light is on, right-turning vehicles must wait to ensure that pedestrians can pass safely without being interfered with by right-turning vehicles.
[0039] The right-turn red light is synchronized with the non-motor vehicle green light: If there is a dedicated non-motor vehicle lane at the intersection, when the non-motor vehicle green light is on, the right-turn red light usually also comes on. The purpose of this is to ensure that right-turning vehicles do not conflict with non-motor vehicles such as electric vehicles and bicycles. This synchronous control can effectively avoid the intersection of right-turning vehicles and the non-motor vehicle lane, especially in the case of a large non-motor vehicle flow in the city.
[0040] This application only gives the above two examples, and there are other situations where the right-turn red light comes on in actual applications, which will not be elaborated one by one.
[0041] During the operation of the right-turn red light, sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane are obtained regularly through camera devices. The sidewalk data includes the pedestrian flow index, the non-motor vehicle lane data includes the number of non-motor vehicles waiting, and the motor vehicle lane data includes the number of vehicles in the right-turn lane, the vehicle flow of vehicles entering the lane, and the vehicle density at the lane entrance. The calculation logic of the pedestrian flow index is as follows: Obtain the number of pedestrians waiting at the sidewalk entrance at the 3rd second before the end of the sidewalk red light, obtain the pedestrian density on the sidewalk at the 10th second of the sidewalk green light, perform normalization processing on the number of pedestrians waiting and the pedestrian density so that the value ranges of the number of pedestrians waiting and the pedestrian density are mapped to between [0, 1], obtain the normalized value of the number of pedestrians waiting and the normalized value of the pedestrian density, and sum the normalized value of the number of pedestrians waiting and the normalized value of the pedestrian density to obtain the pedestrian flow index. The larger the pedestrian flow index, the greater the pedestrian flow on the current road, and the more necessary it is to increase the duration of the right-turn red light to ensure the safe passage of pedestrians.
[0042] The acquisition logic of the number of non-motor vehicles waiting is as follows: Obtain the number of non-motor vehicles waiting at the non-motor vehicle lane entrance at the 3rd second before the end of the non-motor vehicle lane red light. The larger the number of non-motor vehicles waiting, the more non-motor vehicles there are on the current road, and the more necessary it is to increase the duration of the right-turn red light to avoid traffic accidents.
[0043] The acquisition logic for the number of vehicles in the right-turn lane is as follows: Record the number of vehicles in the right-turn lane starting from the time point when the right-turn lane changes from green to red, and stop recording when the right-turn lane changes from red to green to obtain the number of vehicles in the right-turn lane. The more vehicles there are in the right-turn lane, the more right-turning vehicles there are on this road, and the more necessary it is to reduce the red-light duration of the right turn to avoid congestion of right-turning vehicles.
[0044] The acquisition logic for the traffic flow of vehicles entering the lane is as follows: For the roads at intersections, the motor vehicles entering the road generally include those going straight, turning left, and turning right. Therefore, the lane entered by right-turning vehicles is the influencing lane, which generally affects the vehicles going straight and turning left. When the right-turn lane is green and the straight-through lane is also green, record the traffic flow of vehicles going straight. When the right-turn lane is green and the left-turn lane is also green, record the traffic flow of vehicles turning left. Compare the traffic flow of vehicles going straight with the traffic flow of vehicles turning left, and select the larger value as the traffic flow of vehicles entering the influencing lane. Since in traffic rules, the right-of-way for going straight is greater than that for turning left and greater than that for turning right, the greater the traffic flow of vehicles entering the influencing lane, the greater the traffic flow on the current road, and the more necessary it is to increase the red-light duration of the right turn to ensure the safe driving of vehicles going straight and turning left.
[0045] The acquisition logic for the vehicle density at the lane entrance is as follows: Within a specified distance from the lane entrance (for example, 150 meters along the driving direction of the lane entrance), obtain the number of vehicles entering the lane, and divide the number of vehicles by the specified distance to obtain the vehicle density at the lane entrance. The greater the vehicle density at the lane entrance, the more likely it is that there will be vehicle congestion problems in the lane being entered. To avoid the further deterioration of congestion, the more necessary it is to increase the red-light duration of the right turn.
[0046] After the edge computing device substitutes the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor, including the following steps: The edge computing device substitutes the pedestrian flow index, the number of non-motor vehicles waiting, the number of vehicles in the right-turn lane, the traffic flow of vehicles entering the influencing lane, and the vehicle density at the lane entrance into the linear regression analysis algorithm. The algorithm expression is: , where is the correction factor, are respectively the pedestrian flow index, the number of non-motor vehicles waiting, the number of vehicles in the right-turn lane, the traffic flow of vehicles entering the influencing lane, and the vehicle density at the lane entrance, are the regression coefficients, and the regression coefficients are greater than 0; Logical factors composing the correction factor in the use of the present invention: Taking the influence of various data on the stable operation of road traffic as an example, one is the index, that is, the factor causing the change in the stable operation of road traffic (in the present invention, it refers to the influence of various data on the stable operation of road traffic); the second is the weight of these indexes, that is, the proportion occupied when each main influencing data is generated; the third is the operation equation, that is, through what kind of mathematical operation process to obtain the result, and the correction factor obtained by operating the indexes with their respective weights through the operation equation.
[0047] Perform data conversion and processing on the main influencing data obtained from the sample, and convert it into the data language recognized by computer software; secondly, use SPSS software to perform Logistic regression analysis on these evaluation factors, and screen out the factors and their weights that are significantly correlated with the result; thirdly, substitute the evaluation factors and weights into the Logistic regression equation for calculation, so as to obtain the result, specifically: First, ensure the integrity of the main influencing data, process missing values and outliers, convert the data into a format that SPSS software can recognize, usually store the data in formats such as.csv and.xlsx, and then import it into SPSS. Open the SPSS software, import the processed data file, and convert variables as needed. For example, for continuous variables, perform standardization or normalization. Select the "Analysis" menu, and then select the "Binary Logistic" option under "Regression". In the dialog box, add the dependent variable (result) and independent variable (main influencing data) to the corresponding boxes. SPSS will fit the Logistic regression model according to the selected variables. In the output result, information such as the coefficients, standard errors, and p-values of the model will be seen. Check the coefficients and p-values in the output result to judge which variables are significantly correlated with the result. Usually, a p-value less than 0.05 is considered significant. While the model is being fitted, use variable selection methods such as stepwise regression to help screen out the most relevant factors. According to the coefficients of the Logistic regression model, the magnitude of the coefficient reflects the degree of influence of each factor on the result, and the positive or negative sign of the coefficient indicates the direction of the influence. After obtaining the significant factors and their coefficients, obtain the Logistic regression equation, which is used to calculate the probability of each sample and then predict the result.
[0048] Dynamically adjust the red light duration according to the comparison result of the correction factor and the correction threshold, and control the operation of the right-turn red light through the dynamically adjusted red light duration, including the following steps: After obtaining the correction factor, the larger the correction factor, the more necessary it is to increase the red light duration for right turns. Compare the obtained correction factor with a preset correction threshold. The correction threshold is used to determine whether it is necessary to dynamically adjust the initial red light duration. If the correction factor is less than or equal to the correction threshold, it is determined that there is no need to dynamically adjust the initial red light duration. If the correction factor is greater than the correction threshold, it is determined that it is necessary to dynamically adjust the initial red light duration; When it is determined that it is necessary to dynamically adjust the initial red light duration, the initial red light duration is dynamically adjusted through the correction factor. The expression is: , where is the red light duration after dynamic adjustment, is the initial red light duration, is the correction factor.
[0049] Specifically: When dynamically adjusting the red light duration for right turns, the control system regularly obtains the traffic flow information of the road. The traffic flow information includes the conflict point location, the speed of right-turning vehicles, the location of right-turning vehicles, the location of pedestrians, the walking speed of pedestrians, the location of non-motor vehicles, and the speed of non-motor vehicles; And the right-turn conflict frequency of the road is output through the gap acceptance model. The greater the right-turn conflict frequency, it indicates that during the current period of the road, due to the large flow of motor vehicles, non-motor vehicles, and pedestrians, the more times the right-turning vehicles affect the non-motor vehicles and pedestrians themselves, and the more necessary it is to turn on the right-turn red light.
[0050] After the right-turn conflict frequency of the road is output through the gap acceptance model, compare the right-turn conflict frequency with the frequency threshold. If the right-turn conflict frequency is greater than the frequency threshold, the right-turn red light continues to be turned on. If the right-turn conflict frequency is less than or equal to the frequency threshold, control the right-turn red light to be turned off.
[0051] Embodiment 3: The traffic signal control system based on edge computing described in this embodiment includes an opening judgment module, an edge computing module, and a dynamic adjustment module; Opening judgment module: Regularly obtain the traffic flow information of the road. After the edge computing device outputs the right-turn conflict frequency of the road through the gap acceptance model, determine whether it is necessary to control the right-turn red light to be turned on. If it is necessary to turn on, control the right-turn red light to operate with a preset initial duration. The initial duration is sent to the dynamic adjustment module, and the operation information of the right-turn red light is sent to the edge computing module; Edge computing module: During the operation of the right-turn red light, regularly obtain the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data associated with the right-turn lane through the camera device. After the edge computing device substitutes the sidewalk data, non-motor vehicle lane data, and motor vehicle lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs the correction factor, and the correction factor is sent to the dynamic adjustment module; Dynamic adjustment module: Dynamically adjust the red light duration according to the comparison result between the correction factor and the correction threshold, and control the operation of the right-turn red light through the dynamically adjusted red light duration.
[0052] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0053] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.
[0054] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific implementation manners. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the relevant technical fields can understand and utilize the present invention well. The present invention is only limited by the claims and their full scope and equivalents.
Claims
1. A traffic light control method based on edge computing, characterized in that: The control method comprises the following steps: The control system periodically obtains traffic flow information on the road. After the edge computing device outputs the right-turn conflict frequency of the road through the intermittent receiving model, it determines whether it is necessary to control the right-turn red light to turn on. If it is necessary to turn it on, the right-turn red light is controlled to run for a preset initial duration. During the right turn red light operation, the sidewalk data, non-motorized lane data and motor lane data associated with the right turn lane are obtained by the camera device at a regular interval. The edge computing device substitutes the sidewalk data, non-motorized lane data and motor lane data into the linear regression analysis algorithm for analysis, and the linear regression analysis algorithm outputs the correction factor. The red light duration is dynamically adjusted based on the comparison result between the correction factor and the correction threshold, and the right turn red light operation is controlled by the dynamically adjusted red light duration.
2. The traffic signal light control method based on edge computing according to claim 1, characterized in that: During the operation of the right turn red light, the sidewalk data, non-motorized lane data and motor lane data associated with the right turn lane are obtained at a regular intervals through the camera equipment. The sidewalk data includes the pedestrian flow index, the non-motorized lane data includes the number of non-motorized vehicles waiting, and the motor lane data includes the number of vehicles in the right turn lane, the flow of vehicles entering the lane, and the density of vehicles entering the lane.
3. The traffic signal light control method based on edge computing according to claim 2 is characterized in that: After the edge computing device substitutes the pedestrian lane data, non-motorized lane data, and motorized lane data into the linear regression analysis algorithm for analysis, the linear regression analysis algorithm outputs a correction factor, including the following steps: The edge computing device substitutes the pedestrian flow index, the number of non-motor vehicles waiting, the number of vehicles in the right-turn lane, the flow of vehicles entering the lane, and the density of vehicles entering the lane into the linear regression analysis algorithm. The algorithm expression is: , where is the correction factor, They are pedestrian flow index, number of non-motor vehicles waiting, number of vehicles in the right-turn lane, vehicle flow in the lane, and vehicle density at the entrance of the lane. is the regression coefficient, and the regression coefficient is greater than 0.
4. The traffic signal light control method based on edge computing according to claim 3 is characterized in that: The red light duration is dynamically adjusted according to the comparison result between the correction factor and the correction threshold, and the right turn red light operation is controlled by the dynamically adjusted red light duration, including the following steps: After obtaining the correction factor, the larger the correction factor, the more it is necessary to increase the right turn red light duration. The obtained correction factor is compared with a preset correction threshold. The correction threshold is used to determine whether it is necessary to dynamically adjust the initial red light duration. If the correction factor is less than or equal to the correction threshold, it is determined that it is not necessary to dynamically adjust the initial red light duration. If the correction factor is greater than the correction threshold, it is determined that it is necessary to dynamically adjust the initial red light duration. When it is determined that the initial red light duration needs to be dynamically adjusted, the initial red light duration is dynamically adjusted using a correction factor.
5. The traffic signal light control method based on edge computing according to claim 4 is characterized in that: The initial red light duration is dynamically adjusted through the correction factor, and the expression is: , where is the length of the red light after dynamic adjustment. is the initial red light duration, is the correction factor.
6. The traffic signal light control method based on edge computing according to claim 5 is characterized in that: The calculation logic of the pedestrian flow index is as follows: obtain the number of pedestrians waiting at the crosswalk at the 3rd second before the red light on the crosswalk turns red, obtain the pedestrian density on the crosswalk at the 10th second before the green light on the crosswalk turns green, normalize the number of pedestrians waiting and the pedestrian density so that the value ranges of the number of pedestrians waiting and the pedestrian density are mapped to [0,1], obtain the normalized value of the number of pedestrians waiting and the normalized value of the pedestrian density, and sum the normalized value of the number of pedestrians waiting and the normalized value of the pedestrian density to obtain the pedestrian flow index.
7. The traffic signal light control method based on edge computing according to claim 6 is characterized in that: The traffic flow information of the road is obtained at regular intervals, and the traffic flow information includes the location of the conflict point, the speed of the right-turning vehicle, the location of the right-turning vehicle, the location of the pedestrian, the walking speed of the pedestrian, the location of the non-motor vehicle, and the speed of the non-motor vehicle.
8. The traffic signal light control method based on edge computing according to claim 7 is characterized in that: The control system periodically obtains the traffic flow information of the road. After the edge computing device outputs the right turn conflict frequency of the road through the intermittent receiving model, it determines whether it is necessary to control the right turn red light to turn on, including the following steps: The right turn conflict frequency of the road output by the gap acceptance model includes the following steps: Calculate the distance between right-turning vehicles and conflict points, the distance between pedestrians and conflict points, and the distance between non-motor vehicles and conflict points; Obtain the time it takes for right-turning vehicles, pedestrians, and non-motor vehicles to arrive at the conflict point; Then the pedestrian clearance and non-motor vehicle clearance are calculated; Based on the comparison result between the pedestrian clearance and the pedestrian clearance threshold, it is determined whether there is a conflict with pedestrians when the vehicle turns right. Based on the comparison result between the non-motor vehicle clearance and the non-motor vehicle clearance threshold, it is determined whether there is a conflict with non-motor vehicles when the vehicle turns right. During the monitoring period, the number of pedestrian and non-motor vehicle conflicts was recorded, and the total number of conflicts was obtained by adding the number of pedestrian conflicts to the number of non-motor vehicle conflicts. The right-turn conflict frequency was obtained by dividing the total number of conflicts by the monitoring time. After the right-turn conflict frequency of the road is output by the gap acceptance model, the right-turn conflict frequency is compared with the frequency threshold. The frequency threshold is used to determine whether it is necessary to control the right-turn red light to be turned on. If the right-turn conflict frequency is greater than the frequency threshold, it is determined that the right-turn red light needs to be turned on. If the right-turn conflict frequency is less than or equal to the frequency threshold, it is determined that it is not necessary to control the right-turn red light to be turned on.
9. The traffic signal light control method based on edge computing according to claim 8, characterized in that: Calculate the distance between the right-turning vehicle and the conflict point, the distance between the pedestrian and the conflict point, and the distance between the non-motor vehicle and the conflict point. The processing logic is: calculate the distance between the right-turning vehicle and the conflict point, the distance between the pedestrian and the conflict point, and the distance between the non-motor vehicle and the conflict point respectively according to the position of the conflict point, the position of the right-turning vehicle, the position of the pedestrian, and the position of the non-motor vehicle; Obtain the time it takes for right-turning vehicles, pedestrians and non-motor vehicles to reach the conflict point. The processing logic is as follows: obtain the time it takes for right-turning vehicles to reach the conflict point by dividing the distance between the right-turning vehicle and the conflict point by the speed of the right-turning vehicle; obtain the time it takes for pedestrians to reach the conflict point by dividing the distance between the pedestrian and the conflict point by the walking speed of the pedestrian; and obtain the time it takes for non-motor vehicles to reach the conflict point by dividing the distance between the non-motor vehicle and the conflict point by the speed of the non-motor vehicle. Pedestrian clearance and non-motor vehicle clearance are calculated. The processing logic is as follows: pedestrian clearance is obtained by subtracting the time it takes for pedestrians to reach the conflict point from the time it takes for right-turning vehicles to reach the conflict point; non-motor vehicle clearance is obtained by subtracting the time it takes for non-motor vehicles to reach the conflict point from the time it takes for right-turning vehicles to reach the conflict point; To determine whether there is a conflict with pedestrians or non-motor vehicles when the vehicle turns right, the processing logic is: if the pedestrian gap is less than the pedestrian gap threshold, it is determined that there is a pedestrian passing conflict when the vehicle turns right; if the pedestrian gap is greater than or equal to the pedestrian gap threshold, it is determined that there is no pedestrian passing conflict when the vehicle turns right; if the non-motor vehicle gap is less than the non-motor vehicle gap threshold, it is determined that there is a non-motor vehicle passing conflict when the vehicle turns right; if the non-motor vehicle gap is greater than or equal to the non-motor vehicle gap threshold, it is determined that there is no non-motor vehicle passing conflict when the vehicle turns right.
10. A traffic light control system based on edge computing, used to implement the control method according to any one of claims 1 to 9, characterized in that: Including start judgment module, edge computing module, and dynamic adjustment module; Turn-on judgment module: The traffic flow information of the road is obtained regularly. After the edge computing device outputs the right-turn conflict frequency of the road through the intermittent receiving model, it determines whether it is necessary to control the right-turn red light to turn on. If it is necessary to turn it on, the right-turn red light is controlled to run for a preset initial duration. Edge computing module: During the right-turn red light operation, the sidewalk data, non-motorized lane data, and motor lane data associated with the right-turn lane are acquired by the camera device at regular intervals. The edge computing device substitutes the sidewalk data, non-motorized lane data, and motor lane data into the linear regression analysis algorithm for analysis, and the linear regression analysis algorithm outputs the correction factor; Dynamic adjustment module: dynamically adjusts the red light duration according to the comparison result between the correction factor and the correction threshold, and controls the right turn red light operation through the dynamically adjusted red light duration.