Traffic flow prediction and signal adjustment method based on machine learning

Abstract

The invention discloses a traffic flow prediction and signal adjustment method based on machine learning. The method comprises the following steps: data acquisition: acquiring real-time traffic data; data processing: carrying out standardization processing on the traffic flow data and external data; machine learning model training: training the processed multi-modal data through a model; traffic flow prediction: predicting traffic flow change in a period of time in the future according to the current multi-modal data; and signal adjustment: adjusting the traffic signal of each intersection based on the predicted future traffic flow. According to the method, deep learning training is carried out on multi-modal data including traffic flow, weather, emergencies and the like, so that the model can more accurately capture the trend and rule of traffic flow change, and the prediction accuracy is remarkably improved. The system can flexibly adjust the signal period and the green light duration according to the real-time data, so that the traffic signal control has better adaptability, the instantaneous fluctuation of the traffic flow can be effectively dealt with, and the overall efficiency of the traffic system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and in particular to a traffic flow prediction and signal adjustment method based on machine learning. Background Art

[0002] In modern urban traffic management, with the acceleration of urbanization, the surge in traffic flow has become an important factor affecting urban traffic efficiency and safety. How to effectively predict and adjust traffic flow to reduce congestion, improve traffic efficiency and ensure traffic safety has become a technical problem that needs to be solved urgently.

[0003] Traditional traffic flow prediction methods mostly rely on historical data and simple statistical models. These methods often fail to fully consider the impact of real-time traffic conditions and external factors (such as weather, accidents, etc.), resulting in inaccurate prediction results. In addition, with the increasing complexity of the traffic system, simple signal control strategies have gradually become incapable of responding to changes in traffic flow in a timely manner, further exacerbating traffic congestion. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a method for traffic flow prediction and signal adjustment based on machine learning, comprising:

[0007] Data collection: collect real-time traffic data, including traffic volume, vehicle speed, and vehicle queue length; collect traffic-related data through network information, including weather conditions, emergencies, and time characteristics;

[0008] Data processing: clean the collected raw data and standardize the traffic flow data and external data;

[0009] Machine learning model training: The processed multimodal data is trained through the model;

[0010] Traffic flow prediction: Use the trained machine learning model to predict traffic flow changes in the future based on current multimodal data. The prediction time window can be dynamically adjusted according to demand;

[0011] Signal adjustment: Based on the predicted future traffic flow, the traffic signals at each intersection are dynamically adjusted by optimizing the traffic signal cycle, green light duration, and red light switching timing.

[0012] As a preferred solution of the traffic flow prediction and signal adjustment method based on machine learning described in the present invention, the multimodal fusion prediction formula is as follows:

[0013]

[0014] in represents the traffic flow data of the i-th traffic monitoring point at time t; represents the traffic status data of the i-th traffic monitoring point at time t; represents the emergency data of the i-th traffic monitoring point at time t; represents the holiday characteristic data of the i-th traffic monitoring point at time t; Indicates the current traffic light status of the i-th traffic monitoring point; α i is the weight item; is the time decay factor, indicating that the impact of historical data on the current prediction gradually decays over time, and β is a parameter used to control the weight of historical data.

[0015] As a preferred solution of the traffic flow prediction and signal adjustment method based on machine learning described in the present invention, wherein: The specific formula is as follows:

[0016]

[0017] As a preferred solution of the method for traffic flow prediction and signal adjustment based on machine learning described in the present invention, the optimization formula for adjusting the state of the traffic light based on the multimodal fusion prediction formula is as follows:

[0018]

[0019] Where S(t) represents the current signal light cycle; γ is the control parameter, reflecting the adjustment range of the signal light; Y thresh is the traffic flow threshold.

[0020] As a preferred solution of the traffic flow prediction and signal adjustment method based on machine learning described in the present invention, wherein: the model training is carried out through the loss function, and α i Adjustment is made, and the loss function is specifically as follows:

[0021]

[0022] where Y i is the real traffic flow, Predict traffic flow for the model.

[0023] As a preferred solution of the method for traffic flow prediction and signal adjustment based on machine learning described in the present invention, the model training is carried out by training through a reward function, and γ is adjusted. The reward function is specifically as follows:

[0024]

[0025] in is the average length of the vehicle queue in a specific time period; W total is the fleet waiting time; λ is the weight parameter.

[0026] As a preferred solution of the traffic flow prediction and signal adjustment method based on machine learning described in the present invention, it also includes a system interactive interface: real-time display of traffic flow, prediction results and traffic light adjustment status, managers can view the traffic conditions at different intersections through the interface, and the system provides an interface to connect with the traffic signal control system, and managers can manually intervene in the control of traffic lights through the interface.

[0027] In a second aspect, an embodiment of the present invention further provides a traffic flow prediction and signal adjustment system based on machine learning, including:

[0028] Data collection module: It is used to collect traffic flow data in real time through external devices, and to collect external traffic-related data such as weather conditions, emergencies and time characteristics through API and crawler technology;

[0029] Data processing module: cleans and standardizes the collected raw data;

[0030] Machine learning model training module: performs model training on processed multimodal data;

[0031] Traffic flow prediction module: Use the trained model to predict future traffic flow, organize and output the prediction results;

[0032] Signal conditioning module: optimizes the cycle setting of traffic signals according to the predicted traffic flow;

[0033] System interactive interface: displays traffic flow, prediction results, and current traffic light status, while allowing managers to manually intervene.

[0034] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of a traffic flow prediction and signal adjustment method based on machine learning as described in the first aspect of the present invention is implemented.

[0035] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of a traffic flow prediction and signal adjustment method based on machine learning as described in the first aspect of the present invention.

[0036] Beneficial effects of the present invention:

[0037] Through deep learning training of multimodal data including traffic flow, weather, emergencies, etc., the model can more accurately capture the trends and patterns of traffic flow changes, significantly improving the accuracy of predictions. The system can flexibly adjust the signal cycle and green light duration based on real-time data, making traffic signal control more adaptable, effectively responding to instantaneous fluctuations in traffic flow, and improving the overall efficiency of the transportation system. Based on the predicted traffic flow information, the system can reasonably allocate the green light duration of traffic lights to ensure smooth traffic, reduce congestion at intersections, and thus improve the overall road capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0039] Figure 1 A method flow chart of a traffic flow prediction and signal adjustment method based on machine learning proposed by the present invention;

[0040] Figure 2 This is a system architecture diagram of a traffic flow prediction and signal adjustment system based on machine learning proposed by the present invention;

[0041] Figure 3 This is a data comparison chart in Example 2 of the present invention. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0045] Example 1

[0046] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a method for traffic flow prediction and signal adjustment based on machine learning, comprising:

[0047] Data collection: collect real-time traffic data, including traffic volume, speed, and length of vehicle queues, through various traffic monitoring equipment such as road sensors, cameras, and vehicle systems; collect traffic-related data through network information, including weather conditions, emergencies, and time characteristics; weather conditions include temperature, humidity, rainfall, snowfall, wind speed, and other weather data that affect traffic; emergencies include information on mobile phone road traffic accidents, construction, road closures, etc.; time characteristics consider holidays, weekends, peak hours, etc. by combining date and time;

[0048] Data processing: clean the collected raw data and standardize the traffic flow data and external data;

[0049] Machine learning model training: The processed multimodal data is trained through the model, so that it can more accurately predict future traffic volume based on traffic flow, weather, and emergencies, and adjust traffic lights;

[0050] Traffic flow prediction: Use the trained machine learning model to predict traffic flow changes in the future based on current multimodal data. The prediction time window can be dynamically adjusted according to demand;

[0051] Signal adjustment: Based on the predicted future traffic flow, the traffic signals at each intersection are dynamically adjusted by optimizing the traffic signal cycle, green light duration, and red light switching timing.

[0052] Among them, the multimodal fusion prediction formula is as follows:

[0053]

[0054] in represents the traffic flow data of the i-th traffic monitoring point at time t; Represents the traffic condition data of the i-th traffic monitoring point at time t, such as temperature, humidity, rainfall, etc.; Represents the emergency data of the i-th traffic monitoring point at time t, such as accidents, construction, etc.; represents the holiday characteristic data of the i-th traffic monitoring point at time t; Indicates the current traffic light status of the i-th traffic monitoring point; α i is a weight term, which is used to indicate the influence of different data modes on traffic flow prediction; is the time decay factor, which means that the influence of historical data on the current prediction gradually decays over time. β is a parameter used to control the weight of historical data. A larger β value means that the model pays more attention to the latest traffic flow data, while a smaller β allows historical data to have a longer-term impact on the prediction. Integration is used to process traffic flow changes in future time periods and predict the total flow or change trend in a certain time period [t, t+Δt]. By integrating the time τ, the system can capture the dynamic characteristics in the time series.

[0055] Furthermore, The specific formula is as follows:

[0056]

[0057] Here, the logarithmic function of X2(t) reflects the nonlinear impact of weather on traffic flow changes. Emergency events and traffic light status are multiplied together to reflect the priority of emergency events on signal adjustment. Holiday data X4(t) is introduced through the denominator to indicate its longer-term impact on traffic flow.

[0058] Furthermore, the optimization formula for adjusting the traffic light status based on the multimodal fusion prediction formula is as follows:

[0059]

[0060] Where S(t) represents the current signal light cycle, such as the traffic light time; γ is the control parameter, reflecting the adjustment range of the signal light; Y thresh It is the threshold of traffic flow, which is used to measure whether the traffic light cycle needs to be adjusted. The value range of the traffic light cycle adjustment S(t+Δt) is a positive number, usually within a reasonable time range such as 10 seconds to 120 seconds, to ensure the actual operability of traffic signal control.

[0061] Furthermore, the model training is carried out through the loss function, i Adjustment is made, and the loss function is as follows:

[0062]

[0063] where Y i is the real traffic flow, To predict traffic flow for the model, we use traffic flow data for training. By minimizing the loss function, the model can predict αi Adjustments are made to achieve higher prediction accuracy.

[0064] Furthermore, the model training is carried out through the reward function to adjust γ. The reward function is as follows:

[0065]

[0066] in is the average length of the vehicle queue in a specific time period; W total is the waiting time of the fleet; λ is the weight parameter. The system simulates the traffic environment and tries different signal light adjustment strategies and different γ values ​​based on real-time traffic flow and signal light status. It also strengthens the optimal signal light adjustment range through reward mechanisms such as reducing the waiting time of vehicles.

[0067] Furthermore, it also includes a system interactive interface: real-time display of traffic flow, prediction results and traffic light adjustment status. Managers can view the traffic conditions at different intersections through the interface, and the system provides an interface to connect to the traffic signal control system, and managers can manually intervene in the control of traffic lights through the interface.

[0068] This embodiment also provides a traffic flow prediction and signal adjustment system based on machine learning, which specifically includes:

[0069] Data collection module: It is used to collect traffic flow data in real time through external devices, and to collect external traffic-related data such as weather conditions, emergencies and time characteristics through API and crawler technology;

[0070] Data processing module: cleans and standardizes the collected raw data;

[0071] Machine learning model training module: performs model training on processed multimodal data;

[0072] Traffic flow prediction module: Use the trained model to predict future traffic flow, organize and output the prediction results;

[0073] Signal conditioning module: optimizes the cycle setting of traffic signals according to the predicted traffic flow;

[0074] System interactive interface: displays traffic flow, prediction results, and current traffic light status, while allowing managers to manually intervene.

[0075] This embodiment also provides a computer device, which is suitable for a traffic flow prediction and signal adjustment method based on machine learning, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement a traffic flow prediction and signal adjustment method based on machine learning as proposed in the above embodiment.

[0076] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0077] The present embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for traffic flow prediction and signal adjustment based on machine learning as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0078] In summary, the present invention proposes a method for traffic flow prediction and signal adjustment based on machine learning. By performing deep learning training on multimodal data including traffic flow, weather, emergencies, etc., the model can more accurately capture the trends and laws of traffic flow changes, significantly improving the accuracy of predictions. The system can flexibly adjust the signal cycle and green light duration according to real-time data, so that traffic signal control has better adaptability, can effectively respond to instantaneous fluctuations in traffic flow, and improve the overall efficiency of the transportation system. Based on the predicted traffic flow information, the system can reasonably allocate the green light duration of the signal light to ensure smooth traffic, reduce congestion at intersections, and thus improve the overall road capacity.

[0079] Example 2

[0080] Reference Figure 3 , which is the second embodiment of the present invention. This embodiment is different from the first embodiment in that, in order to verify its beneficial effects, experimental comparison data between the present invention and the prior art are provided.

[0081] This embodiment records data in a selected time period, processes the collected raw data, trains the model, uses the trained model to predict traffic flow, and adjusts traffic lights according to the prediction results. The relevant data is recorded and compared with the prior art, as follows:

[0082] index Prior art Summary of the invention Average traffic volume (vehicles / hour) 800 1000 Average waiting time (seconds) 45 30 Average speed (km / h) 30 40 Number of traffic accidents 5 2

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for traffic flow prediction and signal adjustment based on machine learning, characterized in that: include: Data collection: collect real-time traffic data, including traffic volume, speed, and vehicle queue length; Collect traffic-related data through network information, including weather conditions, emergencies and time characteristics; Data processing: clean the collected raw data and standardize the traffic flow data and external data; Machine learning model training: The processed multimodal data is trained through the model; Traffic flow prediction: Use the trained machine learning model to predict traffic flow changes in the future based on current multimodal data. The prediction time window can be dynamically adjusted according to demand; Signal adjustment: Based on the predicted future traffic flow, the traffic signals at each intersection are dynamically adjusted by optimizing the traffic signal cycle, green light duration, and red light switching timing.

2. The method for traffic flow prediction and signal adjustment based on machine learning according to claim 1, characterized in that: The multimodal fusion prediction formula is as follows: in represents the traffic flow data of the i-th traffic monitoring point at time t; represents the traffic status data of the i-th traffic monitoring point at time t; represents the emergency data of the i-th traffic monitoring point at time t; represents the holiday characteristic data of the i-th traffic monitoring point at time t; Indicates the current traffic light status of the i-th traffic monitoring point; α i is the weight item; is the time decay factor, indicating that the impact of historical data on the current prediction gradually decays over time, and β is a parameter used to control the weight of historical data.

3. The method for traffic flow prediction and signal adjustment based on machine learning according to claim 2, characterized in that: Said The specific formula is as follows:

4. The method for traffic flow prediction and signal adjustment based on machine learning according to claim 3, characterized in that: The optimization formula for adjusting the state of the traffic light based on the multimodal fusion prediction formula is as follows: Where S(t represents the current signal light cycle; γ is the control parameter, reflecting the adjustment range of the signal light; Y th resh is the traffic flow threshold.

5. The method for traffic flow prediction and signal adjustment based on machine learning according to claim 4, characterized in that: The model training is carried out through the loss function, i Adjustment is made, and the loss function is specifically as follows: where Y i is the real traffic flow, Predict traffic flow for the model.

6. The method for traffic flow prediction and signal adjustment based on machine learning according to claim 5, characterized in that: The model training is performed by training through a reward function to adjust γ, and the reward function is specifically as follows: in is the average length of the vehicle queue in a specific time period; W total is the fleet waiting time; λ is the weight parameter.

7. The method for traffic flow prediction and signal adjustment based on machine learning according to claim 6, characterized in that: It also includes a system interactive interface: real-time display of traffic flow, prediction results and traffic light adjustment status. Managers can view traffic conditions at different intersections through the interface, and the system provides an interface to connect to the traffic signal control system, through which managers can manually intervene in the control of traffic lights.

8. A machine learning-based traffic flow prediction and signal adjustment system, based on a machine learning-based traffic flow prediction and signal adjustment method according to any one of claims 1 to 7, characterized in that: Specifically include: Data collection module: It is used to collect traffic flow data in real time through external devices, and to collect external traffic-related data through API and crawler technology; Data processing module: cleans and standardizes the collected raw data; Machine learning model training module: performs model training on processed multimodal data; Traffic flow prediction module: Use the trained model to predict future traffic flow, organize and output the prediction results; Signal conditioning module: optimizes the cycle setting of traffic signals according to the predicted traffic flow; System interactive interface: displays traffic flow, prediction results, and current traffic light status, while allowing managers to manually intervene.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a traffic flow prediction and signal adjustment method based on machine learning are implemented in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a traffic flow prediction and signal adjustment method based on machine learning according to any one of claims 1 to 7 are implemented.

Images