Traffic control method and system, computing device and medium

Abstract

The invention relates to a traffic control method and system, computing equipment and a medium. The method comprises the following steps: acquiring first traffic information in a target area in real time; inputting the first traffic information into a trained target regression model to obtain a corresponding signal lamp switching period; controlling a signal lamp according to the signal lamp switching period, and acquiring second traffic information of the target area in the signal lamp switching period; inputting the second traffic information into a trained target classification model, and obtaining a first target traffic condition of a vehicle at each intersection at a preset moment in the signal lamp switching period in the target area; and comparing the first target traffic condition with the traffic condition at the same moment in the signal lamp switching period, and adjusting the signal lamp switching period. According to the method, the condition of traffic jam can be greatly improved, so that the traffic jam rate can be effectively reduced.

Description

Background Art

[0002] With the expansion of urban construction in my country, the number of vehicles has increased day by day, which has brought great pressure to road traffic and traffic jams often occur. The traditional traffic signal control method is usually based on switching signal lights at fixed time intervals. When traffic congestion occurs, traffic commanders will conduct on-site command to clear vehicles. However, with the acceleration of urbanization and the increase in traffic flow, traffic demand has become increasingly complex. The traditional traffic control method that combines fixed time intervals with on-site command by traffic commanders places a heavy burden on traffic commanders. Therefore, the prior art proposes a method for traffic control based on vehicle flow, which predicts vehicle flow and adjusts traffic strategies according to the predicted vehicle flow. However, the traffic strategy obtained in this way has little improvement on traffic congestion, resulting in a high traffic congestion rate. Summary of the invention

[0003] In order to overcome the problem that traditional traffic strategies have little effect on improving traffic congestion, resulting in a high traffic congestion rate, the present invention provides a traffic control method, system, computing device and medium.

[0004] In a first aspect, in order to solve the above technical problems, the present invention provides a traffic control method, comprising:

[0005] Obtain the first traffic information in the target area in real time;

[0006] Inputting the first traffic information into the trained target regression model to obtain a corresponding signal light switching cycle;

[0007] Controlling the signal light according to the signal light switching cycle, and acquiring second traffic information of the target area within the signal light switching cycle;

[0008] Inputting the second traffic information into the trained target classification model to obtain a first target traffic condition of vehicles at each intersection in the target area at a preset time within the signal light switching cycle;

[0009] The first target traffic condition is compared with the traffic condition at the same time in the signal light switching cycle, and the signal light switching cycle is adjusted.

[0010] In a second aspect, in order to solve the above technical problems, the present invention provides a traffic control system, comprising:

[0011] An acquisition module, used for acquiring first traffic information in a target area in real time;

[0012] A first prediction module, used for inputting the first traffic information into a trained target regression model to obtain a corresponding signal light switching cycle;

[0013] A control module, used to control the signal light according to the signal light switching cycle, and obtain second traffic information of the target area within the signal light switching cycle;

[0014] A second prediction module is used to input the second traffic information into the trained target classification model to obtain a first target traffic condition of vehicles at each intersection in the target area at a preset time within the signal light switching cycle;

[0015] An adjustment module is used to compare the first target traffic condition with the traffic condition at the same time in the signal light switching cycle, and adjust the signal light switching cycle.

[0016] In a third aspect, in order to solve the above-mentioned technical problem, the present invention provides a computing device, including a memory, a processor, and a program stored in the memory and running on the processor, and when the processor executes the program, the steps of a traffic control method as described above are implemented.

[0017] In a fourth aspect, in order to solve the above-mentioned technical problem, the present invention provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes the steps of a traffic control method as described above.

[0018] The beneficial effects of the present invention are:

[0019] By acquiring the first traffic information in the target area, the traffic conditions in the target area are acquired. The first traffic information is then input into the trained target regression model for prediction, and the traffic light switching cycle corresponding to the traffic conditions in the target area is obtained. The traffic lights are then controlled according to the predicted traffic light switching cycle. The second traffic information of the target area within the traffic light switching cycle is acquired, and the traffic conditions under the traffic light switching cycle are acquired. The second traffic information is then input into the trained target classification model for prediction, and the first target traffic conditions at each intersection corresponding to the traffic conditions under the traffic light switching cycle are obtained. The first target traffic conditions are then compared with the traffic conditions at the same time within the traffic light switching cycle, so as to dynamically adjust the traffic light switching cycle.

[0020] In this way, the target regression model is used to predict and obtain a preliminary traffic strategy, which is a signal light switching cycle. After the signal light is controlled according to the signal light switching cycle, the target classification model is used to predict again, and the first target traffic condition is compared with the traffic condition at the same time in the predicted signal light switching cycle, so as to verify the preliminary traffic strategy, and adjust the signal light switching cycle according to the test result, so that the adjusted signal light switching cycle is more in line with the current traffic condition in the target area, so as to greatly improve the traffic congestion and effectively reduce the traffic congestion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of a traffic control method provided by the present invention;

[0022] Figure 2 A schematic structural diagram of a traffic control system provided by the present invention. DETAILED DESCRIPTION

[0023] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0024] A traffic control method, system, computing device and medium according to an embodiment of the present invention are described below in conjunction with the accompanying drawings.

[0025] like Figure 1 As shown, the embodiment of the present disclosure provides a traffic control method, including:

[0026] Step S101, obtaining first traffic information in a target area in real time.

[0027] It can be understood that the traffic information represents the traffic situation.

[0028] Step S102: input the first traffic information into the trained target regression model to obtain the corresponding traffic light switching cycle.

[0029] In some embodiments, the target regression model includes any one of a target linear regression, a target decision tree regression, a target random forest regression, and a target XGBoos.

[0030] The signal light switching cycle is the duration of each color light of each signal light. For example, the signal lights in the target area include a first signal light and a second signal light, and the predicted signal light switching cycle is: the first signal light has a red light of 30 seconds, a green light of 90 seconds, and a yellow light of 3 seconds; the second signal light has a red light of 30 seconds, a green light of 45 seconds, and a yellow light of 3 seconds. That is, the target regression model in this embodiment is used to switch the duration of each color light.

[0031] Step S103: Control the traffic light according to the traffic light switching cycle, and obtain second traffic information of the target area within the traffic light switching cycle.

[0032] Step S104: input the second traffic information into the trained target classification model to obtain the first target traffic condition of vehicles at each intersection in the target area at a preset time within the traffic light switching cycle.

[0033] In some embodiments, the target classification model includes any one of a target logistic regression, a target decision tree classification, a target random forest classification, and a target neural network.

[0034] The traffic conditions represent the traffic order. For example, the traffic order includes first priority traffic, second priority traffic, third priority traffic, fourth priority traffic, and so on. It is understandable that there are usually four intersections at the same location, that is, the corresponding traffic order includes first priority traffic, second priority traffic, third priority traffic, and fourth priority traffic.

[0035] Step S105: Compare the first target traffic condition with the traffic condition at the same time in the signal light switching cycle, and adjust the signal light switching cycle.

[0036] A traffic control method provided by an embodiment of the present disclosure is adopted to obtain the first traffic information in the target area, thereby obtaining the traffic conditions in the target area. The first traffic information is then input into the trained target regression model for prediction, and the traffic light switching cycle corresponding to the traffic conditions in the target area is obtained. The traffic light is then controlled according to the predicted traffic light switching cycle. The second traffic information of the target area within the traffic light switching cycle is obtained, thereby obtaining the traffic conditions under the traffic light switching cycle. The second traffic information is then input into the trained target classification model for prediction, and the first target traffic conditions at each intersection corresponding to the traffic conditions under the traffic light switching cycle are obtained. The first target traffic conditions are then compared with the traffic conditions at the same time within the traffic light switching cycle, thereby dynamically adjusting the traffic light switching cycle. In this way, the target regression model is used for prediction, and a preliminary traffic strategy is obtained as the traffic light switching cycle. After the traffic light is controlled according to the traffic light switching cycle, prediction is performed again through the target classification model, and the first target traffic condition is compared with the predicted traffic condition at the same time in the signal light switching cycle, so as to verify the preliminary traffic strategy, and adjust the signal light switching cycle according to the test results, so that the adjusted signal light switching cycle is more in line with the current traffic conditions in the target area, thereby greatly improving the traffic congestion situation and effectively reducing the traffic congestion rate.

[0037] It is understandable that by extracting and analyzing the traffic conditions of each section in the target area (i.e., the traffic information of the target area), and making predictions, it is possible to determine which sections in the target area have greater traffic pressure. Based on this traffic information, the timing of each signal light in the target area (i.e., the switching cycle of the signal light) is dynamically adjusted to give priority to diverting vehicles in the direction of high traffic flow. For example, when the traffic flow of a certain section is abnormally concentrated, the system can automatically extend the green light time in that direction, thereby reducing traffic congestion. This not only alleviates the traffic pressure of a single section, but also optimizes the traffic mobility of the entire area and prevents regional traffic paralysis caused by local congestion. For another example, the intersection is a key node for traffic scheduling in the area, affecting the traffic flow in multiple directions. By analyzing the traffic flow, vehicle waiting time, and congestion index of the intersection, the switching cycle of the signal light can be intelligently adjusted. For another example, during peak hours, a certain intersection may be subject to high traffic flow in multiple directions. At this time, the signal lights of multiple intersections can be adjusted so that the traffic flow passes smoothly in the form of a "green wave belt", reducing the waiting time of vehicles at each intersection. This dynamic traffic light control based on real-time data can not only improve the traffic efficiency of a single intersection, but also optimize the traffic flow of the entire area and avoid frequent stagnation and congestion of vehicles in the area.

[0038] In this way, by comprehensively analyzing the traffic information of multiple sections and intersections in the target area, global traffic scheduling in the target area can be achieved. For example, on the main road of a district, if a section is severely congested, the traffic lights of the surrounding sections can be adjusted to adjust the traffic flow, divert some vehicles to secondary roads or branches with better traffic conditions, and relieve the pressure on the main road. In addition, by adjusting the signal light cycle of certain intersections in advance, vehicles can pass in an orderly manner to avoid congestion spreading to a wider area. This global scheduling can not only improve the traffic operation efficiency of a single district, but also coordinate the traffic flow between districts and reduce the frequency of regional traffic bottlenecks and congestion.

[0039] Preferably, the first traffic information includes: time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters; real-time acquisition of the first traffic information in the target area includes: real-time acquisition of time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters in the target area.

[0040] In this way, by evaluating the traffic conditions from aspects such as time, space, environment, pedestrians and vehicle traffic, the traffic conditions in the target area can be determined more accurately.

[0041] In some embodiments, time parameters include: time, holidays, and working days. Spatial parameters include: intersection lane occupancy rate, traffic density, and intersection queue length. Environmental parameters include: weather conditions, temperature, and special events. Among them, special events represent events that affect vehicle traffic. Pedestrian parameters include: the number of pedestrians on the zebra crossing. Vehicle traffic parameters include: traffic volume and average vehicle speed.

[0042] Preferably, comparing the first target traffic condition with the traffic condition at the same time in the signal light switching cycle and adjusting the signal light switching cycle includes: comparing the first target traffic condition with the traffic condition at the same time in the signal light switching cycle to obtain a comparison result. If the comparison result is different, adjusting the signal light switching cycle to control the traffic condition at the same time after adjustment to be consistent with the first target traffic condition. Controlling the signal light according to the adjusted signal light switching cycle.

[0043] It is understandable that if the comparison results are the same, the signal light switching period is not adjusted.

[0044] In this way, when the comparison results are different, it means that the traffic conditions within the signal light switching cycle still need to be adjusted. Therefore, by adjusting the signal light switching cycle to control the traffic conditions at the same time after adjustment to be consistent with the first target traffic conditions, and controlling the signal light according to the adjusted signal light switching cycle, at this time, vehicles at each intersection in the target area pass according to the first target traffic conditions, which can further improve the traffic congestion in the target area, thereby further reducing the traffic congestion rate.

[0045] It is understandable that the traffic condition includes the traffic order. That is, the first target traffic condition includes the traffic order; the traffic condition within the signal light switching cycle also includes the traffic order. Among them, the traffic order includes first priority traffic, second priority traffic, third priority traffic, fourth priority traffic, and so on. It is understandable that there are usually 4 intersections at the same location, that is, the corresponding traffic order includes first priority traffic, second priority traffic, third priority traffic, and fourth priority traffic.

[0046] In some embodiments, the four intersections of the intersection are the northbound intersection, the eastbound intersection, the southbound intersection and the westbound intersection. It is predicted that at 10 o'clock, the traffic order of the four intersections corresponding to the intersection in the first target traffic situation is: the northbound intersection has the first priority, the eastbound intersection has the second priority, the southbound intersection has the third priority, and the westbound intersection has the fourth priority. If, at the same time as predicted, the traffic order of the four intersections corresponding to the intersection in the signal light switching cycle is: the northbound intersection has the first priority, the eastbound intersection has the second priority, the southbound intersection has the third priority, and the westbound intersection has the fourth priority. It can be seen that if the comparison results of the two are the same, the signal light switching cycle will not be adjusted. If, at the same time as predicted, the traffic order of the four intersections corresponding to the intersection in the signal light switching cycle is: the northbound intersection has the second priority, the eastbound intersection has the first priority, the southbound intersection has the third priority, and the westbound intersection has the fourth priority. It can be seen that the comparison results of the two are different, so the traffic light switching cycle is adjusted so that the traffic conditions indicated by the adjusted traffic light switching cycle are first priority at the northbound intersection, second priority at the eastbound intersection, third priority at the southbound intersection, and fourth priority at the westbound intersection, so as to be consistent with the first target traffic conditions at the same time.

[0047] Preferably, the signal light is controlled according to the adjusted signal light switching cycle, including: controlling the signal light according to the adjusted signal light switching cycle and continuously for a preset time. Obtaining the third traffic information of the target area within the preset time, inputting the third traffic information into the target regression model, and obtaining a new signal light switching cycle. Controlling the signal light according to the new signal light switching cycle, and obtaining the fourth traffic information of the target area within the new signal light switching cycle. Inputting the fourth traffic information into the target classification model, obtaining the second target traffic condition of vehicles at each intersection in the target area at a preset time within the new signal light switching cycle. Compare the second target traffic condition with the traffic condition at the same time within the new signal light switching cycle, and adjust the new signal light switching cycle.

[0048] The traffic lights are controlled according to the adjusted traffic light switching cycle and continue for a preset time. At this time, vehicles at each intersection in the target area pass according to the first target traffic condition, thereby effectively reducing the traffic congestion rate. Since the traffic situation changes in real time, in order to ensure the improvement of the traffic congestion situation, the third traffic information of the target area within the preset time is obtained and predicted through the target regression model, so that the new traffic light switching cycle can be predicted according to the third traffic information corresponding to the first target traffic condition, and the traffic lights are controlled using the new traffic light switching cycle, so that the traffic conditions of vehicles at each intersection pointed to by the new traffic light switching cycle are more in line with the current traffic conditions. The new traffic light switching cycle is then verified by the second target traffic condition, and the new traffic light switching cycle is adjusted to further improve the traffic congestion situation, so as to effectively reduce the traffic congestion rate.

[0049] In some embodiments, the preset duration includes the preset duration of the first priority passage, the preset duration of the second priority passage, the preset duration of the third priority passage, and the preset duration of the fourth priority passage. For example, the preset durations corresponding to the first target passage conditions of the four intersections of the intersection are: the first preset duration of the first priority passage is 90s, the second preset duration of the second priority passage is 60s, the preset duration of the third priority passage is 30s, and the preset duration of the fourth priority passage is 30s. By controlling the signal lights according to the aforementioned preset durations, traffic congestion at the intersection can be improved.

[0050] Preferably, the target regression model is trained and obtained in the following manner: a plurality of historical traffic information in the target area and a historical traffic light switching cycle corresponding to each historical traffic information are collected to obtain a first training set. The first training set is input into a preset regression model for training to obtain a target regression model. In this way, by using historical traffic information and corresponding historical traffic light switching cycles for training, the obtained target regression model can more accurately predict the real-time traffic light switching cycle corresponding to the real-time traffic information, thereby facilitating the reduction of traffic congestion rate.

[0051] It can be understood that the historical traffic information used for training is traffic information with a lower traffic congestion rate, so that the traffic congestion rate predicted by the target regression model obtained through training is also lower.

[0052] Furthermore, multiple historical traffic information in the target area are collected and the historical traffic light switching cycle corresponding to each historical traffic information is collected to obtain a first training set, including: collecting multiple historical traffic information in the target area and the historical traffic light switching cycle corresponding to each historical traffic information, preprocessing each historical traffic information, and using the preprocessed data and the corresponding historical traffic light switching cycle as the first training set.

[0053] In some embodiments, historical traffic data includes: time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters. Time parameters include: time, holidays, weekdays. Space parameters include: intersection lane occupancy rate, traffic density. Environmental parameters include: weather, temperature, visibility. Pedestrian parameters include: number of pedestrians on zebra crossings. Vehicle traffic parameters include: traffic volume, average vehicle speed.

[0054] Specifically, dynamic videos of vehicles and pedestrians can be collected through traffic cameras installed at various intersections, and analyzed to determine the vehicle traffic parameters, pedestrian parameters and space parameters of the intersection. Traffic sensors buried in the road can collect the number, speed and type of vehicles, and analyze to determine the vehicle traffic parameters of the channel. The vehicle's driving speed and location information can be extracted from the vehicle's GPS data through the vehicle's GPS system, and analyzed to determine the vehicle's traffic parameters. Weather data, such as temperature, humidity, rainfall, weather type, etc., can be collected through meteorological sensors, and analyzed to determine environmental parameters.

[0055] Furthermore, each historical traffic information is preprocessed, including: preprocessing each historical traffic information by means of data cleaning, data filtering, data normalization and data digitization.

[0056] In some embodiments, data cleaning includes denoising and missing value processing. Denoising: removing abnormal or invalid data points. For example, excluding extreme values ​​in vehicle speed data (such as abnormally high speeds) or erroneous vehicle counts. For GPS data, eliminate positioning errors caused by signal interference. Missing value processing: completing or interpolating missing data. For example, in vehicle flow data, if the value of a certain time period is missing, it can be completed based on the average value of the previous and next time periods, or linear interpolation can be used to restore data continuity.

[0057] In some embodiments, data filtering includes time filtering and scene filtering. Among them, time filtering: according to specific application requirements, filter out key period data related to traffic signal optimization. For example, extract traffic data during peak hours in the morning and evening, and ignore low traffic periods at night. Scene filtering: select data for specific scenes. For example, focus on traffic data on rainy days, foggy days, etc. to study the impact of bad weather on traffic flow and traffic control.

[0058] In some embodiments, data normalization includes data standardization and data alignment. Among them, numerical standardization: standardize data from different sources (such as the number of vehicles, vehicle speed, traffic density, etc.) so that each feature is comparable at the same scale. For example, the vehicle speed data is normalized so that the numerical range of the vehicle speed is between 0 and 1. Data alignment: align asynchronous data collected from different data sources so that they are analyzed within the same time window. For example, align the video data collected by the traffic camera every second with the location data recorded by the GPS system every minute to unify the time scale.

[0059] In some embodiments, data digitization: non-numerical data such as time parameters and environmental parameters are digitized, for example, holidays are set as 1 and non-holidays are set as 0.

[0060] Furthermore, the preprocessed data and the corresponding historical traffic light switching cycle are used as the first training set, including: fusing the preprocessed data, using the fused data as the feature part of the first training set, and using the corresponding historical traffic light switching cycle as the sample part of the first training set.

[0061] For example, for ease of understanding, pedestrian parameters and vehicle traffic parameters are not shown in this embodiment. Time parameters: hour_of_day: hour (for example, 8 represents the morning rush hour); day_of_week: day of the week (for example, 1 represents Monday); is_holiday: whether it is a holiday (1 represents yes, 0 represents no). Spatial parameters: road_segment_flow_density: traffic flow density of the road segment (for example, cars / minute); intersection_queue_length: intersection queue length (for example, number of vehicles); lane_occupat ion_rate: lane occupancy rate (for example, percentage). Environmental parameters: weather_condition: weather conditions (for example, sunny and rainy day codes); temperature: temperature (for example, degrees Celsius); special_events: special events (for example, traffic accidents, road construction signs). By fusing the above features, each data point (the feature part of the sample in the training set) contains the following input data: time parameters: [hour_of_day=8, day_of_week=1, is_holiday=0]; spatial parameters: [road_segment_flow_density=150, intersection_queue_length=20, lane_occupat ion_rate=75]; environmental parameters: [weather_condition=1, temperature=22, special_events=0]. It can be understood that data fusion of feature parts (such as constructing feature vectors as input data of feature parts, etc.) is a relatively mature existing technology, which will not be elaborated here.

[0062] In some embodiments, the preset regression model includes any one of linear regression, decision tree regression, random forest regression, and XGBoost.

[0063] Preferably, collecting the historical traffic light switching cycle corresponding to each historical traffic information includes: determining whether there is an on-site command at each intersection corresponding to each historical traffic information. If there is an on-site command, determining the corresponding historical traffic light switching cycle based on the on-site command. If there is no on-site command, determining the actual traffic light switching cycle of the intersection as the historical traffic light switching cycle of the corresponding intersection.

[0064] It is understandable that on-site command refers to the on-site command situation of the traffic commander in the target area. Usually, when the traffic commander conducts on-site command, the vehicle driver will pass according to the on-site command situation. That is, the passage directed by the traffic commander on-site will take precedence over the passage indicated by the traffic light. In this way, by determining whether there is on-site command at each intersection corresponding to the historical traffic information, the actual passage situation of the vehicle driver can be determined, so as to facilitate the determination of the real historical traffic light switching cycle. That is, the on-site command situation is used as the historical traffic light cycle.

[0065] Furthermore, the target regression model is obtained by training in the following manner: multiple historical traffic information in the training set is adjusted, and the corresponding signal light switching cycle is simulated. The adjusted multiple historical traffic information and the simulated corresponding signal light switching cycle are used as a new first training set. The new first training set is used to train the current target regression model to obtain a new target regression model. The new target regression model is used as the final target model.

[0066] It is understandable that the corresponding signal light switching cycle is artificially simulated based on historical traffic information. In this way, the target regression model is trained by adjusting the parameters of the first training set, so that the data predicted by the obtained new target regression model is more accurate.

[0067] Preferably, the target regression model is trained and obtained in the following manner: a plurality of historical traffic information in the target area and the traffic conditions corresponding to each historical traffic information are collected to obtain a second training set. The second training set is input into a preset classification model for training to obtain a target classification model. In this way, the target classification model obtained can more accurately predict the real-time traffic conditions corresponding to the real-time traffic information by using the historical traffic information and the corresponding traffic conditions for training, thereby facilitating the verification of the signal light switching cycle to further reduce the traffic congestion rate.

[0068] Furthermore, the target classification model is obtained by training in the following manner: multiple historical traffic information in the training set is adjusted, and corresponding traffic conditions are simulated. The adjusted multiple historical traffic information and the simulated corresponding traffic conditions are used as a new second training set. The new second training set is used to train the current target classification model to obtain a new target classification model. The new target classification model is used as the final target model.

[0069] It is understandable that the corresponding traffic conditions are artificially simulated based on historical traffic information. In this way, by adjusting the parameters of the second training set to train the target classification model, the data predicted by the obtained new target classification type is made more accurate.

[0070] In some embodiments, the target regression model and the target classification model are used to predict and obtain the signal light switching cycle and traffic conditions, and the signal lights in the target area are controlled accordingly. The target regression model and the target classification model are then trained and adjusted according to the controlled traffic conditions (i.e., traffic information), so that the data predicted by the target regression model and the target classification model are more in line with the actual traffic conditions in the target area, so that the control strategy of the signal lights can be dynamically adjusted according to the actual traffic conditions in the target area, so as to further improve the traffic congestion situation and reduce the traffic congestion rate.

[0071] Preferably, a traffic control method further includes: if there is an abnormality in the signal light switching cycle and / or the traffic condition of the first target, a warning prompt is issued to a preset server.

[0072] In this way, by sending an early warning to the preset server when there is an abnormality in the signal light switching cycle and / or the first target traffic condition, it is convenient to remind the staff to handle it in time, thereby ensuring the traffic safety of the target area.

[0073] Combination Figure 2 As shown, Figure 2 A traffic control system 100 is provided. It includes: an acquisition module 101, a first prediction module 102, a control module 103, a second prediction module 104, and an adjustment module 105. Among them, the acquisition module is used to acquire the first traffic information in the target area in real time. The first prediction module is used to input the first traffic information into the trained target regression model to obtain the corresponding signal light switching cycle. The control module is used to control the traffic light according to the signal light switching cycle, and obtain the second traffic information of the target area within the signal light switching cycle. The second prediction module is used to input the second traffic information into the trained target classification model to obtain the first target traffic condition of vehicles at each intersection in the target area at a preset time within the signal light switching cycle. The adjustment module is used to compare the first target traffic condition with the traffic condition at the same time within the signal light switching cycle, and adjust the signal light switching cycle.

[0074] A traffic control system provided by an embodiment of the present disclosure is adopted, and the traffic conditions in the target area are obtained by acquiring the first traffic information in the target area. The first traffic information is then input into the trained target regression model for prediction, and the traffic light switching cycle corresponding to the traffic conditions in the target area is obtained. The traffic light is then controlled according to the predicted traffic light switching cycle. The second traffic information of the target area within the traffic light switching cycle is obtained, and the traffic conditions under the traffic light switching cycle are obtained. The second traffic information is then input into the trained target classification model for prediction, and the first target traffic conditions at each intersection corresponding to the traffic conditions under the traffic light switching cycle are obtained. The first target traffic conditions are then compared with the traffic conditions at the same time in the traffic light switching cycle, so as to dynamically adjust the traffic light switching cycle. In this way, the target regression model is used for prediction, and a preliminary traffic strategy is obtained as the traffic light switching cycle. After the traffic light is controlled according to the traffic light switching cycle, prediction is performed again through the target classification model, and the first target traffic condition is compared with the predicted traffic condition at the same time in the signal light switching cycle, so as to verify the preliminary traffic strategy, and adjust the signal light switching cycle according to the test results, so that the adjusted signal light switching cycle is more in line with the current traffic conditions in the target area, thereby greatly improving the traffic congestion situation and effectively reducing the traffic congestion rate.

[0075] Preferably, the first traffic information includes: time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters; the acquisition module is specifically used to acquire the time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters in the target area in real time.

[0076] Preferably, the adjustment module is specifically used to compare the first target traffic condition with the traffic condition at the same time within the signal light switching cycle to obtain a comparison result; if the comparison result is different, the signal light switching cycle is adjusted to control the adjusted traffic condition at the same time to be consistent with the first target traffic condition; and the signal light is controlled according to the adjusted signal light switching cycle.

[0077] Preferably, the adjustment module is specifically used to control the traffic light according to the adjusted traffic light switching cycle and continue for a preset time length; obtain the third traffic information of the target area within the preset time length, input the third traffic information into the target regression model, and obtain a new traffic light switching cycle; control the traffic light according to the new traffic light switching cycle, and obtain the fourth traffic information of the target area within the new traffic light switching cycle; input the fourth traffic information into the target classification model, and obtain the second target traffic condition of vehicles at each intersection in the target area at a preset time within the new traffic light switching cycle; compare the second target traffic condition with the traffic condition at the same time within the new traffic light switching cycle, and adjust the new traffic light switching cycle.

[0078] Preferably, a traffic control system further includes a training module. The training module is used to collect multiple historical traffic information in the target area and collect the historical traffic light switching cycle corresponding to each historical traffic information to obtain a first training set; input the first training set into a preset regression model for training to obtain a target regression model.

[0079] Preferably, the training module is specifically used to determine whether there is on-site command at each intersection corresponding to each historical traffic information; if there is on-site command, the corresponding historical traffic light switching cycle is determined based on the situation of the on-site command; if there is no on-site command, the actual traffic light switching cycle of the intersection is determined as the historical traffic light switching cycle of the corresponding intersection.

[0080] Preferably, a traffic control system further comprises an early warning module, which is used to send an early warning prompt to a preset server if there is an abnormality in the signal light switching cycle and / or the first target traffic condition.

[0081] A computing device according to an embodiment of the present invention includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, part or all of the steps of the above-mentioned traffic control method are implemented.

[0082] Among them, the computing device can be a computer, and correspondingly, its program is computer software. The above-mentioned parameters and steps in a computing device of the present invention can refer to the parameters and steps in an embodiment of a traffic control method above, and will not be repeated here.

[0083] In an embodiment of the present invention, a computer-readable storage medium is provided, in which instructions are stored. When the instructions are executed, the steps of the above-mentioned traffic control method are executed.

[0084] The computer-readable storage medium may be a transient computer-readable storage medium or a non-transitory computer-readable storage medium.

[0085] The technical solution of the embodiment of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of the embodiment of the present disclosure. The aforementioned computer-readable storage medium may be a non-transitory computer-readable storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes, or a transient computer-readable storage medium.

[0086] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code. Computer-readable storage media can be, for example, but not limited to - electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above.

[0087] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0088] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A traffic control method, characterized in that: include: Obtain the first traffic information in the target area in real time; Inputting the first traffic information into the trained target regression model to obtain a corresponding signal light switching cycle; Controlling the signal light according to the signal light switching cycle, and acquiring second traffic information of the target area within the signal light switching cycle; Inputting the second traffic information into the trained target classification model to obtain a first target traffic condition of vehicles at each intersection in the target area at a preset time within the signal light switching cycle; The first target traffic condition is compared with the traffic condition at the same time in the signal light switching cycle, and the signal light switching cycle is adjusted.

2. The method according to claim 1, characterized in that: The first traffic information includes: time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters; the real-time acquisition of the first traffic information in the target area includes: Acquire the time parameters, space parameters, environmental parameters, pedestrian parameters and vehicle traffic parameters in the target area in real time.

3. The method according to claim 1, characterized in that The comparing the first target traffic condition with the traffic condition at the same time in the signal light switching cycle and adjusting the signal light switching cycle includes: Comparing the first target traffic condition with the traffic condition at the same time in the signal light switching cycle to obtain a comparison result; If the comparison result is different, adjusting the signal light switching cycle to control the traffic condition at the same time after the adjustment to be consistent with the first target traffic condition; The signal light is controlled according to the adjusted signal light switching cycle.

4. The method according to claim 3, characterized in that: The controlling of the signal light according to the adjusted signal light switching cycle includes: The signal light is controlled according to the adjusted signal light switching cycle and lasts for a preset time; Acquire the third traffic information of the target area within the preset time period, input the third traffic information into the target regression model, and obtain a new signal light switching cycle; Controlling the signal light according to the new signal light switching cycle, and acquiring fourth traffic information of the target area within the new signal light switching cycle; Inputting the fourth traffic information into the target classification model to obtain a second target traffic condition of vehicles at each intersection in the target area at a preset time within the new signal light switching cycle; The second target traffic condition is compared with the traffic condition at the same time in the new signal light switching cycle, and the new signal light switching cycle is adjusted.

5. The method according to claim 1, characterized in that The target regression model is trained in the following way: Collecting a plurality of historical traffic information in the target area and collecting a historical traffic light switching cycle corresponding to each of the historical traffic information to obtain a first training set; The first training set is input into a preset regression model for training to obtain the target regression model.

6. The method according to claim 5, characterized in that The collecting of each historical traffic light switching cycle corresponding to the historical traffic information includes: Determine whether there is on-site control at each intersection corresponding to each of the historical traffic information; If there is the on-site command, determining the corresponding historical signal light switching cycle based on the on-site command; If there is no on-site command, the actual traffic light switching cycle of the intersection is determined as the historical traffic light switching cycle of the corresponding intersection.

7. The method according to any one of claims 1 to 6, characterized in that: Also includes: If the signal light switching cycle and / or the first target traffic condition are abnormal, an early warning prompt is sent to a preset server.

8. A traffic control system, characterized in that: include: An acquisition module, used for acquiring first traffic information in a target area in real time; A first prediction module, used for inputting the first traffic information into a trained target regression model to obtain a corresponding signal light switching cycle; A control module, used to control the signal light according to the signal light switching cycle, and obtain second traffic information of the target area within the signal light switching cycle; A second prediction module is used to input the second traffic information into the trained target classification model to obtain a first target traffic condition of vehicles at each intersection in the target area at a preset time within the signal light switching cycle; An adjustment module is used to compare the first target traffic condition with the traffic condition at the same time in the signal light switching cycle, and adjust the signal light switching cycle.

9. A computing device comprising a memory, a processor, and a program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the steps of a traffic control method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the steps of a traffic control method as claimed in any one of claims 1 to 7.

Images