Traffic signal light timing method and system

Abstract

The present invention discloses a traffic light timing method and system, and the method includes the following steps: S1, configuring a traffic light management system to obtain the current travel time and congestion data of traffic lights at each intersection; S2, comparing the congestion data with the current congestion data threshold of the traffic lights at the intersection set by the system, and judging whether it exceeds the current congestion data threshold of the traffic lights at the intersection set by the system, if it exceeds, entering step S3, if not, returning to step S1; S3, inputting the current travel time and congestion data of the traffic lights at the intersection into a trained neural network model, and obtaining the optimal timing of the current travel time of the traffic lights at the intersection; S4, gradually adjusting the travel time of the traffic lights at the intersection according to the optimal timing and the current travel time. The present invention solves the problem that the existing traffic signal controller dynamically matches the optimal signal timing scheme without considering the congestion situation of vehicles passing through the next intersection.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology, and in particular, relates to a traffic signal light timing method and system. Background Art

[0002] Urban road networks cover a wide range and have a complex and ever-changing operating environment. With the rapid development of urban transportation, the number of motor vehicles has increased rapidly, and the incidence of traffic accidents has increased rapidly. Traffic congestion on urban roads has become a major constraint on the development of the national economy and society, causing countless losses. For urban roads, the problem is mainly concentrated at regional intersections. Therefore, how to improve the traffic capacity of regional intersections has become the primary issue in solving urban road network congestion.

[0003] In the Chinese authorized invention patent with patent number CN201910552516.X, a method for dynamically matching the optimal signal timing scheme based on a traffic signal controller is disclosed. The specific steps of the method are as follows: S1. Planning and building a traffic flow-signal scheme database; S2. Predicting the congestion coefficient; S3. Traffic signal scheme experts formulate signal light timing schemes and store them in the traffic flow-signal scheme database; S4. Perform rule judgment on the real-time traffic flow in any time interval and match the optimal signal light timing scheme for the next time interval; S5. Package the acquired signal light timing scheme into a scheme that can be recognized by the signal machine platform and send it to the signal machine platform to achieve real-time dynamic adjustment of traffic lights at traffic intersections. The technical problem to be solved by the present invention is how to collect traffic flow information on the road in real time, optimize the scheme timing, predict the congestion index, propose corresponding signal timing schemes for intersection traffic, and adjust traffic flow.

[0004] The defect of the prior art is that the above patent proposes a corresponding signal timing plan for intersection traffic by predicting the congestion coefficient to regulate traffic flow; but it lacks consideration of vehicles passing through the current intersection and the next intersection, and directly switches the timing plan, which easily causes congestion at other intersections. Summary of the invention

[0005] In view of the problem that the existing traffic signal controller dynamically matches the optimal signal timing plan without considering the congestion of vehicles passing through the current intersection and the next intersection, and directly switching the timing plan easily causes congestion at other intersections, the present invention provides a traffic signal timing method and system.

[0006] In order to achieve the above technical objectives, the technical solution adopted by the present invention is as follows:

[0007] A traffic signal light timing method, comprising the steps of:

[0008] S1. Configure the traffic light management system to obtain the current traffic time and congestion level data of traffic lights at each intersection;

[0009] S2, comparing the congestion degree data with the current congestion degree data threshold of the traffic light at the intersection set by the system, and determining whether it exceeds the current congestion degree data threshold of the traffic light at the intersection set by the system, if it exceeds, proceeding to step S3, if not, returning to step S1, and continuing to obtain the current travel time and congestion degree data of the traffic light at the intersection;

[0010] S3, inputting the current travel time and congestion level data of the traffic lights at the intersection into the trained neural network model to obtain the optimal timing of the current travel time of the traffic lights at the intersection;

[0011] S4. Gradually adjust the traffic light time at the intersection based on the optimal timing and current travel time.

[0012] Furthermore, the detailed process of configuring the traffic light management system is as follows:

[0013] S101. Obtain the current traffic time of traffic lights at each intersection and image data of vehicles congested at the intersection through the authorization of the traffic control department data system;

[0014] S102, processing the image data of vehicles congested at the intersection by an image analysis method to obtain the number of vehicles in the image data of vehicles congested at the intersection;

[0015] S103, counting the number of vehicles merging into the current intersection at a preset period (within a certain period of time);

[0016] S104, dividing the number of merging vehicles by a preset period to obtain an average speed of merging vehicles at the merging entrance of the current intersection within a preset period;

[0017] S105, calculating the average speed of vehicles passing through the current intersection within a preset period in the same manner as step S104;

[0018] S106, calculating the ratio of the average speed of vehicles merging into the intersection at the merging entrance to the current intersection to the average speed of vehicles passing through the current intersection, and classifying the congestion level according to the ratio.

[0019] Furthermore, the current congestion data threshold of the traffic light at the intersection set by the system is the average of the congestion levels of all next intersections that the current intersection can lead to and the sum of the congestion levels of all previous intersections corresponding to the entrance of the current intersection, which is used as the current congestion data threshold of the traffic light at the intersection set by the system.

[0020] Furthermore, the trained neural network model includes an input layer, a convolutional layer, and an output layer;

[0021] During training, the current traffic light time at the intersection, the average speed of vehicles entering the intersection at the current intersection, and the average speed of vehicles passing through the intersection are used as input data for the input layer;

[0022] The convolution layer is preset with a convolution function. The input data is processed by the convolution function of the convolution layer, and the output value of the function processing result is transmitted to the output layer;

[0023] The output value of the output layer is the optimal timing of the current traffic light at the intersection;

[0024] By inputting relevant historical data of the input layer and the output layer, the transformed convolution function of the convolution layer is trained.

[0025] Furthermore, the specific steps for gradually adjusting the traffic light time at intersections are as follows:

[0026] S401, subtract the optimal timing of the current travel time of the traffic light at the intersection from the current travel time, and take the absolute value of the difference;

[0027] S402, dividing the absolute value by the number of directions of the current road at the intersection, to obtain the data value of the current travel time and gradually increase the travel time;

[0028] S403, the passing time of the traffic light at the intersection is gradually increased according to a preset period to the optimal timing of the current passing time of the traffic light at the intersection, and the data value of the passing time obtained in step S402 is increased each time within the preset period.

[0029] A traffic light timing system, comprising a traffic light current passing time acquisition unit, a congestion degree data calculation unit, a congestion degree data over-limit judgment unit, an optimal timing calculation unit and a traffic light passing time gradual adjustment unit at an intersection;

[0030] The current traffic light passing time acquisition unit obtains the current traffic light passing time at each intersection through the authorization of the traffic control department data system;

[0031] The congestion degree data calculation unit calculates the ratio of the average speed of vehicles merging into the intersection at the confluence entrance to the average speed of vehicles passing through the intersection, and divides the congestion degree level according to the ratio;

[0032] A unit for judging whether the congestion degree data exceeds a limit, comparing the congestion degree data with a current congestion degree data threshold of a traffic light at an intersection set by the system, and judging whether the current congestion degree data threshold of a traffic light at an intersection set by the system is exceeded;

[0033] The optimal timing calculation unit inputs the current travel time and congestion level data of the traffic lights at the intersection into the trained neural network model to obtain the optimal timing of the current travel time of the traffic lights at the intersection;

[0034] The traffic light passing time at the intersection is gradually adjusted by the unit, which gradually adjusts the traffic light passing time at the intersection based on the optimal timing and the current passing time.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] By analyzing the current congestion level of traffic lights at intersections and using a neural network model to obtain the optimal timing, the traffic lights at intersections are gradually adjusted to clear the current traffic congestion. This prevents the lack of consideration of the congestion of vehicles passing through the current intersection and the next intersection, and directly switching the timing plan, which may easily lead to congestion problems at other intersections. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is an overall flow chart of a traffic signal light timing method according to an embodiment of the present invention;

[0038] Figure 2 The figure is a structural block diagram of a traffic light timing system in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0040] like Figure 1 As shown, this embodiment provides a traffic signal light timing method, comprising the steps of:

[0041] S1. Configure the traffic light management system to obtain the current traffic time and congestion level data of traffic lights at each intersection;

[0042] S2, comparing the congestion degree data with the current congestion degree data threshold of the traffic light at the intersection set by the system, and determining whether it exceeds the current congestion degree data threshold of the traffic light at the intersection set by the system, if it exceeds, proceeding to step S3, if not, returning to step S1, and continuing to obtain the current travel time and congestion degree data of the traffic light at the intersection;

[0043] S3, inputting the current travel time and congestion level data of the traffic lights at the intersection into the trained neural network model to obtain the optimal timing of the current travel time of the traffic lights at the intersection;

[0044] S4. Gradually adjust the traffic light time at the intersection based on the optimal timing and current travel time.

[0045] Detailed process of configuring traffic light management system:

[0046] S101. Obtain the current traffic time of traffic lights at each intersection and image data of vehicles congested at the intersection through the authorization of the traffic control department data system;

[0047] S102. Process the image data of congested vehicles at the intersection by using an image analysis method to obtain the number of vehicles in the image data of congested vehicles at the intersection; the image analysis method distinguishes the congested vehicles from the road surface by a grayscale image processing method to obtain congested vehicle objects, and then summarizes the number of congested vehicle objects by a counter to obtain the number of vehicles.

[0048] S103, counting the number of vehicles merging into the current intersection at a preset period (within a certain period of time);

[0049] S104, dividing the number of merging vehicles by a preset period to obtain an average speed of merging vehicles at the merging entrance of the current intersection within a preset period;

[0050] S105, calculating the average speed of vehicles passing through the current intersection within a preset period in the same manner as step S104;

[0051] S106, calculating the ratio of the average speed of vehicles merging into the current intersection to the average speed of vehicles passing through the current intersection, and classifying the congestion level according to the ratio. The average speed of vehicles merging into the current intersection is included in the congestion level consideration to ensure more accurate congestion level measurement.

[0052] The current congestion data threshold of the traffic lights at the intersection set by the system is the average of the congestion levels of all the next intersections that the current intersection can reach and the congestion levels of all the previous intersections corresponding to the confluence of the current intersection, which is used as the current congestion data threshold of the traffic lights at the intersection set by the system. The congestion levels of all the next intersections that the current intersection can reach and the congestion levels of all the previous intersections corresponding to the confluence of the current intersection are calculated as the current congestion data threshold to ensure that the current congestion data threshold result fully considers the unblocking and inflow of the current intersection to achieve a better threshold setting, which changes according to real-time changes.

[0053] The trained neural network model includes input layer, convolution layer and output layer;

[0054] During training, the current traffic light time at the intersection, the average speed of vehicles entering the intersection at the current intersection, and the average speed of vehicles passing through the intersection are used as input data for the input layer;

[0055] The convolution layer is preset with a convolution function. The input data is processed by the convolution function of the convolution layer, and the output value of the function processing result is transmitted to the output layer;

[0056] The output value of the output layer is the optimal timing of the current traffic light at the intersection;

[0057] Through the input of historical data related to the input layer and the output layer, the transformed convolution function of the convolution layer is trained. The convolution function can initially draw a two-dimensional function image according to the input value and output value of the neural network model, preset the function formula, and then adjust the preset function formula through continuous training of historical data values ​​to achieve the output layer data value and the convolution layer convolution function The optimal timing is closest to the convolution function.

[0058] Specific steps to gradually adjust the traffic light time at intersections:

[0059] S401, subtract the optimal timing of the current travel time of the traffic light at the intersection from the current travel time, and take the absolute value of the difference;

[0060] S402, dividing the absolute value by the number of directions of the current road at the intersection, to obtain the data value of the current travel time and gradually increase the travel time;

[0061] S403, the passing time of the traffic light at the intersection is gradually increased according to a preset period to the optimal timing of the current passing time of the traffic light at the intersection, and the data value of the passing time obtained in step S402 is increased each time within the preset period.

[0062] like Figure 2 The traffic light timing system comprises a traffic light current passing time acquisition unit, a congestion degree data calculation unit, a congestion degree data over-limit judgment unit, an optimal timing calculation unit and a traffic light passing time gradual adjustment unit at the intersection;

[0063] The current traffic light passing time acquisition unit obtains the current traffic light passing time at each intersection through the authorization of the traffic control department data system;

[0064] The congestion degree data calculation unit calculates the ratio of the average speed of vehicles merging into the intersection at the confluence entrance to the average speed of vehicles passing through the intersection, and divides the congestion degree level according to the ratio;

[0065] A unit for judging whether the congestion degree data exceeds a limit, comparing the congestion degree data with a current congestion degree data threshold of a traffic light at an intersection set by the system, and judging whether the current congestion degree data threshold of a traffic light at an intersection set by the system is exceeded;

[0066] The optimal timing calculation unit inputs the current travel time and congestion level data of the traffic lights at the intersection into the trained neural network model to obtain the optimal timing of the current travel time of the traffic lights at the intersection;

[0067] The traffic light passing time at the intersection is gradually adjusted by the unit, which gradually adjusts the traffic light passing time at the intersection based on the optimal timing and the current passing time.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] By analyzing the current congestion level of traffic lights at intersections and using a neural network model to obtain the optimal timing, the traffic lights at intersections are gradually adjusted to clear the current traffic congestion. This prevents the lack of consideration of the congestion of vehicles passing through the current intersection and the next intersection, and directly switching the timing plan, which may easily lead to congestion problems at other intersections.

[0070] The above is a detailed introduction to a traffic signal timing method and system provided by the present application. The description of the specific embodiment is only used to help understand the method and its core idea of ​​the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A traffic light timing method, characterized in that: Includes steps: S1. Configure the traffic light management system to obtain the current traffic time and congestion level data of traffic lights at each intersection; S2, comparing the congestion degree data with the current congestion degree data threshold of the traffic light at the intersection set by the system, and determining whether it exceeds the current congestion degree data threshold of the traffic light at the intersection set by the system, if it exceeds, proceeding to step S3, if not, returning to step S1, and continuing to obtain the current travel time and congestion degree data of the traffic light at the intersection; S3, inputting the current travel time and congestion level data of the traffic lights at the intersection into the trained neural network model to obtain the optimal timing of the current travel time of the traffic lights at the intersection; S4. Gradually adjust the traffic light time at the intersection based on the optimal timing and current travel time.

2. A method according to claim 1, characterized in that: Detailed process of configuring traffic light management system: S101. Obtain the current traffic time of traffic lights at each intersection and image data of vehicles congested at the intersection through the authorization of the traffic control department data system; S102, processing the image data of vehicles congested at the intersection by an image analysis method to obtain the number of vehicles in the image data of vehicles congested at the intersection; S103, counting the number of vehicles merging into the current intersection at a preset period; S104, dividing the number of merging vehicles by a preset period to obtain an average speed of merging vehicles at the merging entrance of the current intersection within a preset period; S105, calculating the average speed of vehicles passing through the current intersection within a preset period in the same manner as step S104; S106, calculating the ratio of the average speed of vehicles merging into the intersection at the merging entrance to the current intersection to the average speed of vehicles passing through the current intersection, and classifying the congestion level according to the ratio.

3. A method according to claim 1, characterized in that: The current congestion level data threshold of the traffic light at the intersection set by the system is the average of the congestion level of all next intersections that can be reached from the current intersection and the sum of the congestion level of all previous intersections corresponding to the confluence of the current intersection, which is used as the current congestion level data threshold of the traffic light at the intersection set by the system.

4. A method according to claim 1, characterized in that: The trained neural network model includes input layer, convolution layer and output layer; During training, the current traffic light time at the intersection, the average speed of vehicles entering the intersection at the current intersection, and the average speed of vehicles passing through the intersection are used as input data for the input layer; The convolution layer is preset with a convolution function. The input data is processed by the convolution function of the convolution layer, and the output value of the function processing result is transmitted to the output layer; The output value of the output layer is the optimal timing of the current traffic light at the intersection; By inputting relevant historical data of the input layer and the output layer, the transformed convolution function of the convolution layer is trained.

5. A method according to claim 1, characterized in that: Specific steps to gradually adjust the traffic light time at intersections: S401, subtract the optimal timing of the current travel time of the traffic light at the intersection from the current travel time, and take the absolute value of the difference; S402, dividing the absolute value by the number of directions of the current road at the intersection, to obtain the data value of the current travel time and gradually increase the travel time; S403, the passing time of the traffic light at the intersection is gradually increased according to a preset period to the optimal timing of the current passing time of the traffic light at the intersection, and the data value of the passing time obtained in step S402 is increased each time within the preset period.

6. A traffic light timing system, characterized in that: It includes a traffic light current travel time acquisition unit, a congestion degree data calculation unit, a congestion degree data over-limit judgment unit, an optimal timing calculation unit and a traffic light intersection travel time gradual adjustment unit; The current traffic light passing time acquisition unit obtains the current traffic light passing time at each intersection through the authorization of the traffic control department data system; The congestion degree data calculation unit calculates the ratio of the average speed of vehicles merging into the intersection at the confluence entrance to the average speed of vehicles passing through the intersection, and divides the congestion degree level according to the ratio; A unit for judging whether the congestion degree data exceeds a limit, comparing the congestion degree data with a current congestion degree data threshold of a traffic light at an intersection set by the system, and judging whether the current congestion degree data threshold of a traffic light at an intersection set by the system is exceeded; The optimal timing calculation unit inputs the current travel time and congestion level data of the traffic lights at the intersection into the trained neural network model to obtain the optimal timing of the current travel time of the traffic lights at the intersection; The traffic light passing time at the intersection is gradually adjusted by the unit, which gradually adjusts the traffic light passing time at the intersection based on the optimal timing and the current passing time.

Images