A tidal lane control method, device, equipment and medium

Abstract

The application discloses a tidal lane control method and device, equipment and medium, wherein the method comprises: acquiring vehicle queuing data of a target area in real time, the vehicle queuing data at least comprising queuing time data and queuing quantity data; predicting vehicle flow data of each lane in the next period based on historical vehicle flow data and vehicle queuing data of the current period; in the case that the tidal lane meets a preset opening condition, determining an opening time and an opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane, and opening the tidal lane based on the opening time and the opening scale. The application solves the problem that the existing tidal lane control method cannot effectively alleviate traffic congestion due to response lag.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of control, in particular to a tidal lane control method, device, equipment and medium. BACKGROUND

[0002] In the field of tidal lane control, the existing technology mainly adopts pre-programmed control based on fixed time table or real-time intervention control based on manual observation. The fixed time table control mode enables the tidal lane according to the periodicity of historical traffic flow within the preset time period; and the manual observation mode manually switches the lane guidance when obvious tidal flow characteristics are observed by traffic management personnel monitoring real-time road conditions.

[0003] However, the above existing methods have the following problems: the fixed time table control lacks flexibility and cannot adapt to sudden traffic mode changes such as morning and evening peak flow fluctuations and large event dispersal; the manual observation mode is delayed in response and limited by subjective judgment, and it is difficult to timely find the load imbalance state of multiple routes or local congestion of a specific turning lane, resulting in a lag in the activation time of the tidal lane and the inability to effectively alleviate traffic pressure. SUMMARY

[0004] The main purpose of the present application is to provide a tidal lane control method, device, equipment and medium, which aims to solve the technical problem that the existing tidal lane control method cannot effectively alleviate traffic congestion due to delayed response.

[0005] To achieve the above purpose, the present application provides a tidal lane control method applied to a tidal lane control system, wherein the system stores historical traffic flow data and a tidal lane resource library, and the historical traffic flow data at least includes historical queuing time data and historical queuing quantity data, and the method comprises: acquiring vehicle queuing data of a target area in real time, wherein the vehicle queuing data at least includes queuing time data and queuing quantity data; predicting vehicle flow data of each lane in the next period based on the historical traffic flow data and the vehicle queuing data of the current period; determining the opening time and opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane in the case that the tidal lane meets the preset opening condition, and opening the tidal lane based on the opening time and the opening scale.

[0006] Optionally, the prediction of the vehicle flow data of each lane in the next period based on the historical traffic flow data and the vehicle queuing data of the current period comprises: determining at least one feature vector based on the historical traffic flow data, training a preset neural network based on the feature vector, wherein the feature vector is used to represent the traffic condition of the target area; inputting the vehicle queuing data of the current period into the trained preset neural network to obtain the vehicle flow data of each lane in the next period.

[0007] Optionally, the preset opening condition is determined based on the historical traffic flow data; the method further comprises: constructing a conditional decision function based on the historical traffic flow data and setting a decision score threshold, the conditional decision function being used to determine whether the tidal lane needs to be opened; inputting the predicted traffic flow data of each lane in the next period into the conditional decision function to obtain a decision score in the next period; and the preset opening condition is that the decision score is greater than the decision score threshold.

[0008] Optionally, the system further stores weather data, the preset neural network is a GBDT model, and the feature vector includes a weather influence factor; before the at least one feature vector is determined based on the historical traffic flow data, the method further comprises: classifying the weather data and determining weather influence factors of each type of weather data, wherein the weather influence factor represents the influence degree of weather on traffic flow.

[0009] Optionally, before the vehicle queuing data in the current period is input into the trained preset neural network to obtain the traffic flow data of each lane in the next period, the method further comprises: determining a weather influence factor in the current period; and the inputting of the vehicle queuing data in the current period into the trained preset neural network to obtain the traffic flow data of each lane in the next period comprises: inputting the vehicle queuing data in the current period and the weather influence factor in the current period into the trained GBDT model to obtain the traffic flow data of each lane in the next period.

[0010] Optionally, the tidal lane resource library includes a lane position and a lane variable direction, and the opening scale includes an opening number and a vehicle traffic guidance.

[0011] Optionally, the method further comprises: in the case where the opening state of the tidal lane changes, sending prompt information to a management terminal.

[0012] In addition, to achieve the above-mentioned purposes, the application further provides a tidal lane control device, comprising: a data acquisition module, configured to acquire vehicle queuing data of a target area in real time, the vehicle queuing data at least including queuing time data and queuing quantity data; a traffic flow prediction module, configured to predict traffic flow data of each lane in a next period based on historical traffic flow data and vehicle queuing data in a current period; and a tidal lane control module, configured to, in the case where a tidal lane meets a preset opening condition, determine an opening time and an opening scale of the tidal lane based on vehicle queuing data in the current period and the predicted traffic flow data of each lane, and open the tidal lane based on the opening time and the opening scale.

[0013] The application further provides a tidal lane control device, comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method in any possible implementation manner described above.

[0014] The application further provides a computer-readable storage medium, comprising: a computer program stored therein, and the computer program is executed by a processor to implement the method in any possible implementation manner described above.

[0015] The tidal lane control method, device, equipment and medium provided by the application can first accurately determine the congestion degree and passing efficiency of the current traffic state by acquiring the vehicle queuing data of a target area in real time; secondly, the next-period lane-level traffic flow prediction can be performed based on historical traffic flow data and current-period data, so that the hysteresis of the existing fixed schedule and manual observation method can be limitedly improved; and then, the opening time and scale of the tidal lane can be determined based on the real-time queuing data and the prediction result, so that the control strategy can be ensured to be based on the actual traffic demand and congestion aggravation caused by response delay can be avoided. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of the tidal lane control method provided for Embodiment One of the application;

[0017] Figure 2 A flowchart of the tidal lane control method provided for Embodiment Two of the application;

[0018] Figure 3 A structural block diagram of the tidal lane control device provided for an embodiment of the application;

[0019] Figure 4 A structural schematic diagram of the tidal lane control device provided for an embodiment of the application.

[0020] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are merely intended to explain the application and not to limit the application.

[0022] In the field of tidal lane control, the existing technology mainly adopts pre-programmed control based on fixed schedule or real-time intervention control based on manual observation. The fixed schedule control mode enables the tidal lane according to the periodicity of historical traffic flow in a preset time period; and the manual observation mode manually switches the lane guidance when obvious tidal flow characteristics are observed by traffic management personnel.

[0023] However, the above-mentioned existing method has the following problems: the fixed schedule control lacks flexibility and cannot adapt to sudden traffic mode changes such as morning and evening peak flow fluctuations, large event dispersal, etc.; the manual observation method is delayed in response and limited by subjective judgment, and it is difficult to find the load imbalance state of multiple routes or local congestion of a specific turning lane in time, resulting in a lag in the activation time of the tidal lane and the inability to effectively alleviate traffic pressure.

[0024] To solve the above problems, the present application provides a tidal lane control method, device, equipment and medium, and the following will introduce the application scheme in detail.

[0025] Figure 1 The flow chart of the tidal lane control method provided by the first embodiment of the present application, the method can be applied to a tidal lane control system, the system stores historical traffic data and a tidal lane resource library, the historical traffic data at least includes historical queuing time data and historical queuing quantity data, and the tidal lane resource library includes lane position and lane variable direction. The method can be executed by a control device in communication connection with the tidal lane control system, and the control device can be a tidal lane control device. Referring to Figure 1 The tidal lane control method can include the following steps:

[0026] S11, real-time acquisition of vehicle queuing data of a target area, the vehicle queuing data at least including queuing time data and queuing quantity data.

[0027] Among them, the historical traffic data also includes date data, and the vehicle queuing data also includes vehicle average speed, vehicle occupancy rate and road condition data.

[0028] In the specific implementation process, first, the road condition image of the target area can be collected in real time by using the collection device, and then the vehicle average speed and the vehicle occupancy rate of each lane can be obtained by using the computer vision model such as YOLO model to detect the vehicle. At the same time, the queuing time data and the queuing quantity data of the vehicle in the target area are obtained based on the road condition image and through the trajectory tracking algorithm.

[0029] In addition, the GPS data of the taxi or online car navigation can also be accessed to correct the above-mentioned vehicle queuing data.

[0030] S12, predicting the vehicle flow data of each lane in the next period based on the historical traffic data and the vehicle queuing data of the current period.

[0031] Among them, the period length can be set by the implementer, for example, 5 minutes can be set as a period, and the period length is not limited in the embodiment.

[0032] It should be noted that the neural network algorithm is used to predict the traffic flow data of each lane in the next period in this embodiment, and other prediction algorithms suitable for multi-dimensional data can also be used in other embodiments, and the LSTM model (Long Short Term Memory Network) is taken as an example for description in this embodiment.

[0033] In one embodiment, in step S12, the traffic flow data of each lane in the next period is predicted based on the historical traffic flow data and the vehicle queuing data in the current period, and specifically can include:

[0034] S121, determining at least one feature vector based on the historical traffic flow data, and training a preset neural network based on the feature vector, wherein the feature vector is used to represent the traffic condition of the target area;

[0035] S122, inputting the vehicle queuing data of each lane in the current period into the trained preset neural network to obtain the traffic flow data of each lane in the next period.

[0036] In the specific implementation process, first, at least one feature vector is determined based on the historical traffic flow data, and the feature vector can include: queuing quantity, queuing time, historical same period flow, date type. It should be noted that the date type refers to whether the current date is a holiday.

[0037] Further, the LSTM model is trained based on these feature vectors and historical traffic flow data, the model input is the vehicle queuing data of each lane in the current period, and the model output is the predicted traffic flow data of each lane in the next period.

[0038] Further, the vehicle queuing data of the current period is input into the trained LSTM model to obtain the traffic flow data of each lane in the next period.

[0039] In another embodiment, an ARIMA model (time series model) is used to predict the traffic flow data of each lane in the next period.

[0040] Specifically, first, the data of each historical traffic flow lane is respectively constituted into a time series; after missing value filling and abnormal value processing are performed on each sequence, a time series graph is drawn and statistical tests such as ADF and KPSS are used to determine whether the sequence is stationary; if it is not stationary, it is made stationary through difference (once or twice) and the difference order d is recorded; second, the autocorrelation function (ACF) and the partial autocorrelation function (PACF) are calculated based on the time series graph, and the values of the AR term p and the MA term q are determined by combining the information criterion (AIC, BIC), thereby constructing an ARIMA (p, d, q) model; finally, the ARIMA model constructed is used to perform rolling prediction on the next period to obtain the predicted traffic flow of each lane.

[0041] It should be noted that if the difference processing is performed, the predicted value needs to be recovered to the original scale by inverse difference.

[0042] S13, in the case that the tidal lane meets the preset opening condition, determining the opening time and opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane, and opening the tidal lane based on the opening time and opening scale.

[0043] The opening scale includes an opening number and a vehicle traffic guidance.

[0044] In an embodiment, the preset opening condition can be determined based on historical traffic data, and before step S13, the method further comprises:

[0045] S31, constructing a condition decision function based on historical traffic data and setting a decision score threshold, the condition decision function being used to determine whether the tidal lane needs to be opened;

[0046] S32, inputting the predicted vehicle flow data of each lane of the next period into the condition decision function to obtain a decision score of the next period;

[0047] S33, the preset opening condition is that the decision score is greater than the decision score threshold.

[0048] In the specific implementation process, first, based on the historical traffic data, for any lane, first subtract the vehicle flow data of the next period from the vehicle flow data of the current period, and take the ratio of the difference value to the vehicle flow data of the current period as the predicted flow growth rate; second, for any lane, take the absolute value of the difference between the lane occupancy rate of the current period and the lane occupancy rate of any adjacent lane in the same direction as the adjacent lane load difference; finally, normalize the historical queuing time data and the historical queuing quantity data.

[0049] Further, the weights of the predicted flow growth rate, the adjacent lane load difference, the historical queuing time data, and the historical queuing quantity data are set respectively. It should be noted that the weight is a decimal number from 0 to 1, and the sum of the weights is 1. The plurality of weights in the embodiment are set by the implementer, and in other implementation processes, a neural network can be used to determine the optimal weight, or different weights can be set in different dates, for example, in the morning peak or evening peak, the weights of the predicted flow growth rate, the adjacent lane load difference, the historical queuing quantity data, and the historical queuing time data are 0.3, 0.1, 0.3, and 0.3 respectively; in the non-morning peak / evening peak, the weights of the predicted flow growth rate, the adjacent lane load difference, the historical queuing quantity data, and the historical queuing time data are 0.3, 0.1, 0.2, and 0.4 respectively.

[0050] Further, the constructed conditional decision function has the right side of the above-mentioned predicted traffic growth rate, adjacent lane load difference, historical queuing time data and historical queuing quantity data multiplied by corresponding weights first, and then all the products are added, and the left side is the decision score.

[0051] It should be noted that the embodiment can set the decision score threshold based on the historical decision scores of each lane in the historical traffic data and the historical processing scheme, or the decision score threshold can be set by the implementer.

[0052] Further, the traffic flow data of each lane in the next cycle is input into the conditional decision function to obtain the decision score of the next cycle, and when the decision score is greater than the decision score threshold, the tidal lane is opened.

[0053] It should be noted that in the embodiment, the tidal lane control system needs to specifically control the opening number of the lane and the change of the traffic guidance.

[0054] In the specific implementation process, the predicted traffic flow data of each lane is subtracted from the current traffic capacity of each lane, and the ratio of the difference value to the theoretical traffic capacity of a single tidal lane is used as the opening number. It should be noted that if the ratio is a decimal number, the decimal number is rounded up, and the maximum cannot exceed the maximum opening number (the number of lanes). The current traffic capacity and the theoretical traffic capacity of a single tidal lane are preset values, and the units are the same as those of the traffic flow data, both of which are vehicles / minute.

[0055] Further, the opening time and opening scale of each lane are determined based on the tidal lane resource library, and a control instruction is output.

[0056] Exemplarily, the control instruction can be: at a certain time, open 2 south-to-north tidal lanes at a certain intersection for 90 minutes.

[0057] In one embodiment, the tidal lane control method of the embodiment can further include the following steps: in the case that the opening state of the tidal lane changes, sending a prompt information to a management terminal.

[0058] The management terminal can be a tidal lane control system or a device in communication connection with the tidal lane control system.

[0059] The tidal flow lane control method proposed in this application firstly acquires real-time vehicle queuing data of the target area, which can accurately determine the congestion level and traffic efficiency of the current traffic situation; secondly, based on historical traffic flow data and current cycle data, lane-level traffic flow prediction for the next cycle can effectively improve the lag of existing fixed schedules and manual observation methods; furthermore, based on real-time queuing data and prediction results, the opening time and scale of the tidal flow lane are determined, which can ensure that the control strategy is based on actual traffic demand and avoid congestion aggravation caused by response delay.

[0060] It should be noted that the above embodiments do not consider the impact of weather factors on traffic flow prediction. In existing technologies, when weather is introduced as a feature vector into the prediction model, it is usually treated only as a static classification label (such as "light rain" or "moderate rain"). However, this method cannot accurately describe the degree of impact of weather on traffic conditions. For example, the impact of "three hours of continuous heavy rain" on traffic conditions is significantly different from that of "light rain that has just stopped for 10 minutes". The former will lead to continuously limited traffic capacity and increasingly severe congestion; while the latter may only cause temporary congestion, and the congestion can be quickly resolved.

[0061] Based on this, this embodiment introduces a "weather impact factor" that can quantify the dynamic impact of weather on traffic flow, thereby improving the prediction accuracy of the model and the robustness of the tidal lane control method in this embodiment under different weather conditions.

[0062] Based on the above embodiments, Figure 2 This is a flowchart of the tidal lane control method provided in Embodiment 2 of this application. Figure 2 For based on Figure 1 The preferred embodiment of the corresponding tidal lane control method is referred to Figure 2 The tidal lane control method may include the following steps:

[0063] S21. Obtain real-time vehicle queuing data for the target area. The vehicle queuing data shall include at least queuing time data and queuing number data.

[0064] S22. Determine the weather influencing factors for the current cycle;

[0065] S23. Determine at least one feature vector based on historical traffic flow data, and train a GBDT model based on the feature vector. The feature vector includes weather influencing factors.

[0066] S24. Input the vehicle queuing data of the current period and the weather influencing factors of the current period into the trained GBDT model to obtain the traffic flow data of each lane in the next period.

[0067] S25, in the case where the tidal lane meets the preset opening condition, determining the opening time and opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane, and opening the tidal lane based on the opening time and opening scale.

[0068] The GBDT model is a gradient boosting decision tree.

[0069] It can be understood that the embodiment increases the weather influence factor on the basis of the above-mentioned embodiment, and trains the GBDT model by taking the weather influence factor as a feature vector.

[0070] In the specific implementation process, in step S22, the weather influence factor of the current period can be determined by using the following formula (1) :

[0071]

[0072] wherein, represents a preset adjustment constant, represents the time corresponding to the current period, represents the time when the weather of the current period starts, represents a preset decay factor constant, and the unit is time, represents an exponential function.

[0073] It should be noted that, can be set according to the persistence of different weather types (for example, when it is raining heavily, = 90 minutes), the value is between 0 and 1, 0 represents that the weather has no negative impact (for example: sunny, cloudy), and the closer the value is to 1, the stronger the negative impact of the weather (for example: heavy rain, heavy fog, ice and snow weather), and can be set by the implementer.

[0074] It can be understood that, on the basis of the above-mentioned embodiment, the embodiment introduces a dynamic decay weather influence factor, and integrates the dynamic weather factor into the GBDT prediction model, so that the model can accurately learn the differentiated influence of different weather conditions on traffic, and improve the robustness of the prediction model under different weather scenarios. Further, based on the prediction result calibrated by the weather factor, the tidal lane control decision can be made, so that the system can still make a control strategy highly matched with the traffic flow change trend under different weather conditions, so that the tidal lane control can maintain the optimal control effect under various weather conditions.

[0075] On the basis of the above-mentioned embodiment, Figure 3 is a structure block diagram of a tidal lane control device according to an embodiment of the present application, as shown in Figure 3As shown, the tidal lane control device 300 can include a data acquisition module 310, a traffic flow prediction module 320, and a tidal lane control module 330.

[0076] The data acquisition module is configured to acquire vehicle queuing data of a target region in real time, the vehicle queuing data comprising at least queuing time data and queuing quantity data; the traffic flow prediction module is configured to predict traffic flow data of each lane in a next period based on historical traffic flow data and the vehicle queuing data of a current period; and the tidal lane control module is configured to determine an opening time and an opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted traffic flow data of each lane in the case where the tidal lane meets a preset opening condition, and open the tidal lane based on the opening time and the opening scale.

[0077] In some embodiments, the traffic flow prediction module 320 can be specifically configured to determine at least one feature vector based on the historical traffic flow data, train a preset neural network based on the feature vector, wherein the feature vector is used to represent the traffic condition of the target region; and input the vehicle queuing data of the current period into the trained preset neural network to obtain the traffic flow data of each lane in the next period.

[0078] In some embodiments, the tidal lane control module 330 can be specifically configured to construct a conditional decision function based on the historical traffic flow data and set a decision score threshold, the conditional decision function being used to determine whether the tidal lane needs to be opened; input the predicted traffic flow data of each lane in the next period into the conditional decision function to obtain a decision score in the next period; and the preset opening condition is that the decision score is greater than the decision score threshold.

[0079] In some embodiments, the traffic flow prediction module 320 can be specifically configured to classify weather data and determine weather influence factors of each type of weather data, wherein the weather influence factors represent the influence degree of weather on traffic flow.

[0080] In some embodiments, the tidal lane control module 330 can be specifically configured to determine a weather influence factor of the current period; input the vehicle queuing data of the current period and the weather influence factor of the current period into the trained GBDT model to obtain the traffic flow data of each lane in the next period.

[0081] In some embodiments, the tidal lane control module 330 can be specifically configured to send prompt information to a management terminal in the case where there is a change in the opening state of the tidal lane.

[0082] It should be noted that details not disclosed in the tidal lane control device of the present embodiment are referred to the details disclosed in the embodiments of the tidal lane control method in the present specification, which will not be described here.

[0083] On the basis of the above-mentioned embodiments, Figure 4 FIG. 1 is a schematic structural diagram of a tidal lane control device according to an embodiment of the present application, and Figure 4 As shown in the figure, the electronic device can include a processor 410, a communications interface 420, a memory 430, and a communications bus 440, wherein the processor 410, the communications interface 420, and the memory 430 complete communications with each other through the communications bus 440. The processor 410 can invoke a logical instruction in the memory 430 to execute a tidal lane control method, which includes: acquiring vehicle queuing data of a target area in real time, the vehicle queuing data including at least queuing time data and queuing quantity data; predicting vehicle flow data of each lane in a next period based on historical vehicle flow data and the vehicle queuing data of a current period; determining an opening time and an opening scale of a tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane in the case that the tidal lane meets a preset opening condition, and opening the tidal lane based on the opening time and the opening scale.

[0084] In addition, the logical instruction in the memory 430 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or in part or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0085] On the basis of the above-mentioned embodiments, in another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program being stored on a non-transitory computer-readable storage medium, and the computer program being executable by a processor to enable a computer to perform a tidal lane control method provided by any of the above-mentioned methods, the method comprising: acquiring vehicle queuing data of a target area in real time, the vehicle queuing data at least comprising queuing time data and queuing quantity data; predicting vehicle flow data of each lane in a next period based on historical vehicle flow data and the vehicle queuing data of a current period; and determining an opening time and an opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane, and opening the tidal lane based on the opening time and the opening scale, in a case where the tidal lane meets a preset opening condition.

[0086] On the basis of the above-mentioned embodiments, in another aspect, the present application also provides a non-transitory computer-readable storage medium, which stores a computer program, the computer program being executable by a processor to implement a tidal lane control method provided by any of the above-mentioned methods, the method comprising: acquiring vehicle queuing data of a target area in real time, the vehicle queuing data at least comprising queuing time data and queuing quantity data; predicting vehicle flow data of each lane in a next period based on historical vehicle flow data and the vehicle queuing data of a current period; and determining an opening time and an opening scale of the tidal lane based on the vehicle queuing data of the current period and the predicted vehicle flow data of each lane, and opening the tidal lane based on the opening time and the opening scale, in a case where the tidal lane meets a preset opening condition.

[0087] The above merely describes the preferred embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation based on the content of the specification and the drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for controlling tidal flow lanes, characterized in that, An application is made to a tidal flow lane control system, wherein the system stores historical traffic flow data and a tidal flow lane resource database, the historical traffic flow data including at least historical queuing time data and historical queuing length data, and the method includes: Real-time acquisition of vehicle queuing data in the target area, wherein the vehicle queuing data includes at least queuing time data and queuing number data; Based on the historical traffic flow data and the vehicle queuing data of the current period, predict the traffic flow data of each lane in the next period. When the tidal flow lane meets the preset opening conditions, the opening time and opening scale of the tidal flow lane are determined based on the vehicle queuing data of the current period and the predicted traffic flow data of each lane, and the tidal flow lane is opened based on the opening time and the opening scale. The preset activation conditions are determined based on the historical traffic flow data; the method further includes: A conditional decision function is constructed based on historical traffic flow data, and a decision score threshold is set. The conditional decision function is used to determine whether a tidal flow lane needs to be opened. The traffic flow data for each lane predicted for the next period is input into the conditional decision function to obtain the decision score for the next period. The preset activation condition is that the decision score is greater than the decision score threshold. The conditional decision function constructed based on historical traffic flow data includes: For any lane, acquire the lane's historical queuing time data and historical queuing quantity data, determine the lane's predicted traffic growth rate based on the traffic flow data of adjacent cycles, and determine the lane's adjacent lane load difference based on the lane's occupancy rate and the occupancy rate of adjacent lanes in the same direction. The conditional decision function for the lane is constructed based on the lane's predicted traffic growth rate, adjacent lane load difference, historical queuing time data, and historical queue count data, as well as the preset weights corresponding to the predicted traffic growth rate, adjacent lane load difference, historical queuing time data, and historical queue count data.

2. The method as described in claim 1, characterized in that, The method of predicting traffic flow data for each lane in the next period based on the historical traffic flow data and the vehicle queuing data of the current period includes: At least one feature vector is determined based on historical traffic flow data, and a preset neural network is trained based on the feature vector, wherein the feature vector is used to characterize the traffic conditions of the target area; The vehicle queuing data for the current period is input into the trained preset neural network to obtain the traffic flow data for each lane in the next period.

3. The method as described in claim 2, characterized in that, The system also stores weather data, the preset neural network is a GBDT model, the feature vector includes weather influencing factors, and before determining at least one feature vector based on historical traffic flow data, the method further includes: The weather data is classified and the weather impact factors for each type of weather data are determined, wherein the weather impact factors characterize the degree of influence of weather on traffic flow.

4. The method as described in claim 3, characterized in that, Before inputting the vehicle queuing data of the current period into the trained preset neural network to obtain the traffic flow data of each lane in the next period, the method further includes: Determine the weather influencing factors for the current cycle; The step of inputting the vehicle queuing data of the current period into a trained preset neural network to obtain the traffic flow data of each lane in the next period includes: By inputting the vehicle queuing data and weather influencing factors of the current period into the trained GBDT model, traffic flow data for each lane in the next period can be obtained.

5. The method as described in claim 1, characterized in that, The tidal lane resource database includes lane locations and lane variable directions, and the opening scale includes the number of lanes to be opened and vehicle traffic guidance.

6. The method as described in claim 1, characterized in that, The method further includes: If the opening status of the tidal lane changes, a notification message is sent to the management terminal.

7. A tidal flow lane control device, characterized in that, include: The data acquisition module is used to acquire vehicle queuing data in the target area in real time. The vehicle queuing data includes at least queuing time data and queuing number data. The traffic flow prediction module is used to predict the traffic flow data of each lane in the next period based on historical traffic flow data and vehicle queuing data of the current period. The tidal lane control module is used to determine the opening time and scale of the tidal lane based on the vehicle queuing data of the current period and the predicted traffic flow data of each lane when the tidal lane meets the preset opening conditions, and to open the tidal lane based on the opening time and the opening scale. The tidal lane control module is also used to acquire historical queuing time data and historical queuing quantity data for any lane, and determine the predicted traffic flow growth rate of the lane based on the traffic flow data of adjacent cycles of the lane, and determine the load difference between adjacent lanes of the lane based on the occupancy rate of the lane and the occupancy rate of adjacent lanes in the same direction. The conditional decision function for the lane is constructed based on the lane's predicted traffic growth rate, adjacent lane load difference, historical queuing time data, and historical queuing quantity data, as well as the preset weights corresponding to the predicted traffic growth rate, adjacent lane load difference, historical queuing time data, and historical queuing quantity data. Set a decision score threshold for the conditional decision function, which is used to determine whether the tidal lane needs to be opened; The traffic flow data for each lane predicted for the next period is input into the conditional decision function to obtain the decision score for the next period. The preset activation condition is that the decision score is greater than the decision score threshold.

8. A tidal flow lane control device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The device contains a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

Images