Expressway traffic risk prediction method and device, storage medium and program product

Abstract

This disclosure provides a method, apparatus, storage medium, and program product for predicting highway traffic risks. The method includes: constructing a cascaded model based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data for a target area; predicting traffic risks based on the cascaded model; and implementing traffic control measures for the target area based on the traffic risks. The cascaded model includes two ARIMA models and one CA model. This method combines the advantages of ARIMA and CA models to more comprehensively predict future traffic risks and implement traffic control measures based on the traffic risk results, thereby reducing the incidence of traffic accidents and congestion and improving the level of highway traffic management.

Description

Technical Field

[0001] This disclosure relates to the field of road traffic technology, and in particular to a method, device, storage medium, and program product for predicting highway traffic risks. Background Technology

[0002] With the improvement of road infrastructure and economic development, vehicle use has become increasingly common, but correspondingly, traffic pressure has also increased. To reduce traffic accidents and congestion, traffic risk prediction is a good preventive measure. However, in the current highway management, risk prediction mainly relies on human experience, simple statistical methods, and sometimes a single machine learning model. These methods are not only inefficient but also have limited accuracy, making it difficult to effectively guide traffic. Summary of the Invention

[0003] In view of this, the present disclosure provides a method, apparatus, storage medium, and program product for predicting highway traffic risks, which can combine the advantages of ARIMA and CA models to more comprehensively assess traffic risks and conduct traffic control based on the traffic risk results, thereby improving the level of highway traffic management.

[0004] In a first aspect, the present disclosure provides a method for predicting highway traffic risks, employing the following technical solution: A cascaded model is constructed based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data of the target area; Traffic risks are predicted based on the cascaded model, and traffic control is implemented in the target area based on the traffic risks. The cascaded model includes two ARIMA models and one CA model; the historical traffic flow data is a record of the number of vehicles passing through the target area in the past time period; the historical vehicle attribute data is a record of the vehicle characteristics of the target area in the past time period; the road network data is the road characteristics of the target area; and the environmental data is the natural environment characteristics and traffic environment characteristics of the target area.

[0005] Optionally, the step of constructing a cascaded model based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data of the target area includes: A first ARIMA model is constructed and optimized based on the historical traffic flow data, and future traffic flow data is predicted based on the optimized first ARIMA model. A second ARIMA model is constructed and optimized based on the historical vehicle attribute data, and future vehicle attribute data is predicted based on the optimized second ARIMA model. Based on the future traffic flow data, the future vehicle attribute data, the road network data, the traffic rules, the environmental data, and the first CA model, the first traffic situation in the target area is predicted; Based on the first traffic situation, the first ARIMA model, the second ARIMA model, and the first CA model are fine-tuned to obtain the third ARIMA model, the fourth ARIMA model, and the second CA model. The third ARIMA model, the fourth ARIMA model, and the second CA model form a cascaded model.

[0006] Optionally, the step of constructing and optimizing the first ARIMA model based on the historical traffic flow data includes: Based on the historical traffic flow data, a first time series graph, a first autocorrelation function, and a first partial autocorrelation function are constructed. The first parameter is obtained based on the first time series plot, the first autocorrelation function, and the first partial autocorrelation function, and the first ARIMA model is constructed based on the first parameter. The first parameter is tuned, and the first ARIMA model is optimized based on the tuned first parameter.

[0007] Optionally, the step of constructing and optimizing the second ARIMA model based on the historical vehicle attribute data includes: A second time series graph, a second autocorrelation function, and a first partial autocorrelation function are constructed based on the historical vehicle attribute data. The second parameters are obtained based on the second time series plot, the second autocorrelation function, and the second partial autocorrelation function, and the second ARIMA model is constructed based on the second parameters. The second parameter is tuned, and the second ARIMA model is optimized based on the tuned second parameter.

[0008] Optionally, the step of fine-tuning the first ARIMA model, the second ARIMA model, and the first CA model based on the first traffic situation to obtain the third ARIMA model, the fourth ARIMA model, and the second CA model includes: Based on the first traffic situation, the first parameter and the second parameter are finely adjusted to obtain the third ARIMA model and the fourth ARIMA model; The first CA model is fine-tuned based on the third ARIMA model and the fourth ARIMA model to obtain the second CA model.

[0009] Optionally, the prediction of traffic risk based on the cascaded model includes: The second traffic situation in the target area is simulated based on the cascaded model; Based on the second traffic situation and the prediction model, the traffic risk is predicted.

[0010] Optionally, it also includes: The historical traffic flow data and the historical vehicle attribute data are cleaned, time-seriesified, stationarity checked, trend identified, and seasonally adjusted. The future traffic flow data, the future vehicle attribute data, the road network data, the traffic rules, and the environmental data are cleaned, integrated, and preprocessed.

[0011] Secondly, this disclosure also provides a highway traffic risk prediction system, which adopts the following technical solution: The building module is used to construct a cascaded model based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules and environmental data of the target area; The prediction module is used to predict traffic risks based on the cascaded model and to implement traffic control in the target area based on the traffic risks. The cascaded model includes two ARIMA models and one CA model; the historical traffic flow data is a record of the number of vehicles passing through the target area in the past time period; the historical vehicle attribute data is a record of the vehicle characteristics of the target area in the past time period; the road network data is the road characteristics of the target area; and the environmental data is the natural environment characteristics and traffic environment characteristics of the target area.

[0012] Thirdly, this disclosure also provides a computer device, which adopts the following technical solution: The computer device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform any of the highway traffic risk prediction methods described above.

[0013] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium storing computer instructions for causing a computer to execute any of the highway traffic risk prediction methods described above.

[0014] Fifthly, embodiments of this disclosure also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0015] The highway traffic risk prediction method provided in this disclosure comprehensively considers the impact of multiple factors on traffic risk. It selects historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data to construct a cascaded model. The cascaded model combines the advantages of ARIMA and CA models. The ARIMA model can capture the trend and periodicity of time series, while the CA model can consider the impact of spatial correlation and traffic network topology on risk. By constructing a cascaded model, the advantages of both ARIMA and CA models are combined, thereby more comprehensively assessing traffic risk, increasing the accuracy of traffic risk prediction, and enabling timely traffic control measures to prevent traffic accidents and traffic congestion based on traffic risk.

[0016] The above description is merely an overview of the technical solution disclosed herein. In order to better understand the technical means of this disclosure and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the highway traffic risk prediction method provided in this embodiment of the disclosure; Figure 2 A schematic diagram illustrating the process of constructing a cascaded model provided in this embodiment of the disclosure; Figure 3 A schematic diagram of the highway traffic risk prediction system provided in this embodiment of the disclosure; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. Detailed Implementation

[0019] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0020] It should be understood that the following specific examples illustrate the implementation of this disclosure, and those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0021] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0022] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The drawings only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0023] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0024] Reference Figure 1 This disclosure provides a method for predicting highway traffic risks, including the following steps: S1: Construct a cascaded model based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data of the target area.

[0025] The data includes: historical traffic flow data, which records the number of vehicles passing through the target area over a past period, including at least one or more of the following: traffic flow at different times (e.g., hourly, daily), the number of vehicles passing through a certain point or road segment; historical vehicle attribute data, which records the characteristics of vehicles in the target area over a past period, including at least one or more of the following: vehicle density, type, speed, acceleration, etc. Different types of vehicles may have different driving characteristics, such as cars, buses, and trucks, which can affect the simulation results of traffic flow; road network data, which records the road characteristics of the target area, including at least one or more of the following: the topology, length, width, and connectivity of each road in the target area; traffic rules, which records the traffic rules of the target area, including at least one or more of the following: vehicle driving direction, speed limits, traffic signal control, etc.; and environmental data, which records the natural and traffic environmental characteristics of the target area, including at least one or more of the following: weather, visibility, traffic accidents, road conditions, etc. Weather and visibility data include both historical and forecast data, and can be obtained through weather forecasts.

[0026] The cascaded model includes two ARIMA models and one CA model. The ARIMA model is an Autoregressive Integrated Moving Average Model, and the CA model is a Cellular Automaton Model.

[0027] S2: Based on the cascaded model, predict traffic risks and implement traffic control in the target area according to the traffic risks.

[0028] This disclosed method for predicting highway traffic risks comprehensively considers the impact of multiple factors on traffic risk. It employs a cascaded model constructed from historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data, combining two ARIMA models and one CA model. This allows for modeling and predicting traffic risks from different perspectives, improving prediction accuracy and stability. The ARIMA model captures the trends and periodicity of time series data, while the CA model considers the impact of spatial correlation and traffic network topology on risk. The complementary nature of the ARIMA and CA models leads to a more comprehensive assessment of traffic risks and increases the accuracy of traffic risk prediction.

[0029] Based on traffic risks, timely traffic control measures can be taken, including route optimization, traffic signal adjustment, and traffic restrictions, to reduce the likelihood of traffic accidents, improve road traffic efficiency, and thus improve the highway traffic environment and promote the sustainable development of highways.

[0030] In summary, the implementation of such precise traffic risk prediction and guidance measures can make more effective use of traffic management resources, provide a scientific basis for highway traffic operation, improve the level of highway traffic management, reduce accident handling costs and time costs caused by traffic congestion, thereby achieving the goals of resource optimization and cost saving.

[0031] Optionally, before building the cascaded model, historical traffic flow data and historical vehicle attribute data undergo data cleaning, time series processing, stationarity checks, trend identification, and seasonal adjustment. Data cleaning includes handling missing values, outliers, isolated points, and erroneous data; time series processing organizes the data into a time series format, ensuring the data is arranged in chronological order; stationarity checks verify the stability of the time series data, and if not, adjustments are needed; trend identification uses statistical methods or visualization tools to identify trends in the data, including linear and non-linear trends, for further trend adjustment; seasonal adjustment, if the data shows seasonal variations, eliminates these periodic effects through differencing or seasonal decomposition to stabilize the data.

[0032] Reference Figure 2 The flowchart illustrating the construction process of a cascading model is shown. "Constructing a cascading model" includes the following steps: S21: Construct and optimize the first ARIMA model based on historical traffic flow data, and predict future traffic flow data based on the optimized first ARIMA model; S22: Construct and optimize the second ARIMA model based on historical vehicle attribute data, and predict future vehicle attribute data based on the optimized second ARIMA model; S23: Predict the first traffic situation in the target area based on future traffic flow data, future vehicle attribute data, road network data, traffic rules, environmental data, and the first CA model; S24: Based on the first traffic situation, fine-tune the first ARIMA model, the second ARIMA model, and the first CA model to obtain the third ARIMA model, the fourth ARIMA model, and the second CA model. The third ARIMA model, the fourth ARIMA model, and the second CA model form a cascaded model.

[0033] In S21, the first ARIMA model consists of three parts: AR1 (autoregressive), I1 (difference), and MA1 (moving average). The first parameter of the first ARIMA model includes the main parameters of these three parts. The main parameter of the AR1 part is the first autoregressive order p1, the main parameter of the I1 part is the first difference order d1, and the main parameter of the MA1 part is the first moving average order q1.

[0034] Based on historical traffic flow data, a first time series graph, a first autocorrelation function (ACF), and a first partial autocorrelation function (PFC) are constructed. The first parameter is initially determined based on these parameters. Specifically, the time series graph is examined to determine if it is stationary. If stationary, d1 is set to zero. If not stationary, an appropriate d1 value is selected for differencing. If the time series remains non-stationary, the d1 value is adjusted until it becomes stationary. The first lag term exceeding the confidence interval after truncation of the PFC is identified; this lag term's index is the initial p1 value. The first lag term exceeding the confidence interval after truncation of the ACF is also identified; this lag term's index is the initial q1 value.

[0035] Fine-tuning of p1, d1, and q1 yields multiple different combinations of (p1, d1, q1). For each combination, multiple first ARIMA models are fitted. These models are then evaluated using the Akaike Information Criterion (AIC) to obtain multiple first AIC values. Similarly, the Bayesian Information Criterion (BIC) is used to evaluate these models, resulting in multiple first BIC values. The first ARIMA model with the smallest first AIC and / or first BIC value is selected. A white noise test is then used to assess whether the residual sequence of the selected first ARIMA model exhibits white noise. If the residual sequence does not exhibit white noise, the three parameters p1, d1, and q1 of the first ARIMA model are further adjusted until the residual sequence exhibits white noise.

[0036] The first ARIMA model is used to predict future traffic flow data, which includes the predicted traffic flow values ​​for each area of ​​the target region within a preset time period. The preset time period is set to one week.

[0037] In S22, the second ARIMA model consists of three parts: AR2, I2, and MA2. The second parameters of the second ARIMA model include the main parameters of these three parts. The main parameter of the AR2 part is the second autoregressive order p2, the main parameter of the I2 part is the second difference order d2, and the main parameter of the MA1 part is the second moving average order q2.

[0038] A second time series plot, a second autocorrelation function, and a second partial autocorrelation function are constructed based on historical vehicle attribute data. The second parameter is initially determined based on these data. Specifically, the time series is checked to determine if it is stationary; if stationary, d2 is set to zero; if not stationary, an appropriate d2 value is selected for differencing. If the time series remains non-stationary, the d2 value is adjusted until it becomes stationary. The first lag term exceeding the confidence interval after truncation of the second partial autocorrelation function is identified; this lag term's index is the initial p2 value. The first lag term exceeding the confidence interval after truncation of the second autocorrelation function is also identified; this lag term's index is the initial q2 value.

[0039] Fine-tuning of p2, d2, and q2 yields multiple different combinations of (p2, d2, q2). For each combination, multiple second ARIMA models are fitted. These models are evaluated using the Akaike Information Criterion to obtain multiple second AIC values. Similarly, multiple second BIC values ​​are obtained by evaluating these models using the Bayesian Information Criterion. The second ARIMA model with the smallest second AIC and / or second BIC values ​​is selected. A white noise test is used to evaluate whether the residual sequence of the selected second ARIMA model exhibits white noise. If the residual sequence does not exhibit white noise, the second ARIMA model has captured the data structure well. If the residual sequence does not exhibit white noise, the p2, d2, and q2 parameters of the second ARIMA model are further adjusted until the residual sequence exhibits white noise.

[0040] The second ARIMA model is used to predict future vehicle attribute data, which includes the predicted vehicle attribute values ​​for each region in the target area within a preset time period.

[0041] Optionally, data cleaning, data integration, and data preprocessing can be performed on future traffic flow data, future vehicle attribute data, road network data, traffic rules, and environmental data. Data integration involves standardizing the formats of different data sets and then consolidating them into a unified dataset for subsequent processing. Data preprocessing includes spatialization, normalization, and standardization of the data.

[0042] In S23, based on future traffic flow data, future vehicle attribute data, road network data, traffic rules, and environmental data, the cells, cell neighbors, and state transition rules of the first CA model are determined.

[0043] The method for determining cells is as follows: a preliminary simulation of the target area is performed using road network data, future traffic flow data, and environmental data. The simulated target area is divided into multiple units, each of which is a cell. These cells form the space of the first CA model. For example, a small segment of a lane is considered as a cell.

[0044] Among these, road network data determines the length, width, topology, and connectivity of roads, allowing for the simulation of various roads within the target area. Future vehicle attribute data determines the number of vehicles and their attributes, such as vehicle type and speed, on each cell. Future traffic flow data provides vehicle density information for various areas over a specific time period, thus influencing the initial cell state distribution. Environmental data considers the impact of weather and road conditions on traffic behavior, specifically affecting vehicle speed and driving patterns, influencing not only the cell itself but also its state transition rules.

[0045] Cells are labeled with their states: 0 represents no vehicle, meaning the cell is idle; 1 represents a low-speed vehicle, meaning there is a low-speed vehicle in the cell; and 2 represents a high-speed vehicle, meaning there is a high-speed vehicle in the cell. Other states can be added as needed, such as congested state and free flow state.

[0046] The method for determining cell neighbors is as follows: the road network data provides the connection relationship and direction of each road. Based on the connection relationship and direction of each road, neighboring cells are assigned to the cell. For example, if the cell is the left side of a crossroads, the cells in the up, down, and right directions are assigned as cell neighbors to this cell.

[0047] The method for determining state transition rules is as follows: State transition rules are defined considering factors such as future vehicle attribute data, traffic rules, environmental data, and the states of cell neighbors. These rules affect how vehicles move between cells, and as vehicles move from one state to another, they influence the cell's state. For example, when the vehicle density is greater than a first threshold, the cell transitions from a high-speed vehicle state to a low-speed vehicle state; when the vehicle speed is greater than a second threshold, the vehicle transitions from a low-speed vehicle state to a high-speed vehicle state. If a cell is currently in a car-free state, whether a vehicle will enter in the next moment depends on the state of its neighbors. For example, if a cell's neighbor is in a low-speed vehicle state, the probability of the cell transitioning to a low-speed vehicle state in the next moment is 0.6; if a cell's neighbor is in a high-speed vehicle state, the probability of the cell transitioning to a high-speed vehicle state in the next moment is 0.8.

[0048] Furthermore, this disclosure also introduces a transition matrix. Starting from the initial state, the propagation and distribution changes of traffic flow in the target area are simulated multiple times based on the transition matrix. In this process, the third parameter in the first CA model is adjusted and optimized in combination with the actual traffic flow characteristics and road structure. The third parameter includes the size of the cell division, the range of cell neighbors, and the state transition rules.

[0049] Based on the defined cells, cell neighbors, and state transition rules, the behavior of vehicles within the target area is simulated. At each time step, the state of each vehicle, including its position and speed, is updated according to the state transition rules. Furthermore, the simulation considers behaviors such as acceleration, deceleration, and lane changes of the vehicles to reflect the initial traffic conditions in the target area in the future.

[0050] In S24, based on the simulated first traffic situation, it is determined whether abnormal congestion occurs within the cell. If abnormal congestion occurs, the region where the cell is located is identified. To reflect this local congestion in the prediction of the first ARIMA model, a correction value is added to the traffic flow prediction value for that region, resulting in a new traffic flow prediction value that reflects the impact of congestion on traffic flow. The first parameter is then fine-tuned based on the new traffic flow prediction value, and the final third ARIMA model is obtained based on the fine-tuned first parameter. The prediction results of the third ARIMA model are more consistent with the actual situation. If no abnormal congestion occurs, there is no need to fine-tune the first parameter, and the first ARIMA model is used as the final third ARIMA model. Abnormal congestion refers to a sudden and significant increase in traffic flow in a certain region or road segment, leading to severe traffic congestion and preventing smooth vehicle passage.

[0051] Vehicle attribute data is analyzed from the first traffic situation in the target area. When the difference between the analyzed vehicle attribute data and the future vehicle attribute data is greater than the preset value, the second parameter is adjusted based on the analyzed vehicle attribute data, thereby achieving the purpose of fine-tuning the second ARIMA model and obtaining the final fourth ARIMA model.

[0052] The third ARIMA model is used to re-predict future traffic flow data, and the fourth ARIMA model is used to re-predict future vehicle attribute data. The third parameter is then fine-tuned based on the new future traffic flow data and the new future vehicle attribute data to obtain the final second CA model. The third ARIMA model, the fourth ARIMA model, and the second CA model form a cascaded model.

[0053] In S2, a second CA model is used to re-simulate the future traffic conditions of the target area, denoted as the second traffic scenario. The second traffic scenario is more accurate than the first traffic scenario. A prediction model is pre-built, and traffic simulation data is obtained based on the second traffic scenario of the target area. Feature engineering methods are used to extract features from the traffic simulation data to help the prediction model predict traffic risks, including traffic congestion and traffic accidents.

[0054] The prediction model uses a pre-trained convolutional neural network model, collects labeled historical traffic simulation data as sample data, and divides the sample data into training samples, test samples and validation samples according to a preset ratio. The convolutional neural network model is trained using the training samples and optimized using the test samples and validation samples.

[0055] The highway traffic risk prediction method disclosed herein uses the data predicted by the first ARIMA model and the data predicted by the second ARIMA model as the traffic simulation data source for the first CA model. This not only helps the first CA model to complete the simulation prediction of the first traffic situation, but also optimizes the first ARIMA model and the second ARIMA model based on the first traffic situation to obtain the third ARIMA model and the fourth ARIMA model. The third ARIMA model and the fourth ARIMA model can then be used to further optimize the first CA model to obtain the second CA model. This mutual optimization method can make the final simulation results of the second CA model more accurate.

[0056] Reference Figure 3 This disclosure provides a highway traffic risk prediction system, including: Module 101 is used to build a cascaded model based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules and environmental data of the target area; Prediction module 102 is used to predict traffic risks based on cascaded models and to implement traffic control in the target area based on the traffic risks. The cascaded model includes two ARIMA models and one CA model; historical traffic flow data is a record of the number of vehicles passing through the target area in the past time period; historical vehicle attribute data is a record of the vehicle characteristics of the target area in the past time period; road network data is the road characteristics of the target area; and environmental data is the natural environment characteristics and traffic environment characteristics of the target area.

[0057] The various variations and specific examples of the highway traffic risk prediction method provided above are also applicable to the highway traffic risk prediction system provided in this disclosure. Through the foregoing detailed description of the highway traffic risk prediction method, those skilled in the art can clearly understand the implementation method of the highway traffic risk prediction system. For the sake of brevity, they will not be described in detail here.

[0058] A computer device according to embodiments of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0059] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the computer device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory, causing the computer device to perform all or part of the steps of the highway traffic risk prediction method of the foregoing embodiments of this disclosure.

[0060] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0061] like Figure 4 This is a schematic diagram of a computer device provided for an embodiment of the present disclosure. It illustrates a structural schematic diagram suitable for implementing the computer device in the embodiments of the present disclosure. Figure 4 The computer device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0062] like Figure 4 As shown, a computer device may include a processor (such as a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) or programs loaded from storage devices into random access memory (RAM). The RAM also stores various programs and data required for the operation of the computer device. The processor, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0063] Typically, the following devices can be connected to the I / O interface: input devices, such as sensors or visual information acquisition devices; output devices, such as displays; storage devices, such as magnetic tapes or hard drives; and communication devices. Communication devices allow the computer device to communicate wirelessly or wiredly with other devices (such as edge computing devices) to exchange data. Although Figure 4 A computer apparatus with various devices is shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or included alternatively.

[0064] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processor, all or part of the steps of the highway traffic risk prediction method of embodiments of this disclosure are performed.

[0065] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0066] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the highway traffic risk prediction methods described in the foregoing embodiments of the present disclosure are performed.

[0067] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0068] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0069] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0070] In this disclosure, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The block diagrams of devices, apparatuses, devices, and systems involved in this disclosure are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as "comprising," "including," "having," etc., are open-ended terms meaning "including but not limited to," and are used interchangeably with them. The terms "or" and "and" as used herein refer to the terms "and / or," and are used interchangeably with them unless the context clearly indicates otherwise. The term "such as" as used herein refers to the phrase "such as but not limited to," and is used interchangeably with it.

[0071] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0072] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0073] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0074] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0075] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for predicting traffic risks on highways, characterized in that, include: A cascaded model is constructed based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data of the target area; Traffic risks are predicted based on the cascaded model, and traffic control is implemented in the target area based on the traffic risks. The cascaded model includes two ARIMA models and one CA model; the historical traffic flow data is a record of the number of vehicles passing through the target area in the past time period; the historical vehicle attribute data is a record of the vehicle characteristics of the target area in the past time period; the road network data is the road characteristics of the target area; and the environmental data is the natural environment characteristics and traffic environment characteristics of the target area.

2. The highway traffic risk prediction method according to claim 1, characterized in that, The construction of a cascaded model based on historical traffic flow data, historical vehicle attribute data, road network data, traffic rules, and environmental data of the target area includes: A first ARIMA model is constructed and optimized based on the historical traffic flow data, and future traffic flow data is predicted based on the optimized first ARIMA model. A second ARIMA model is constructed and optimized based on the historical vehicle attribute data, and future vehicle attribute data is predicted based on the optimized second ARIMA model. Based on the future traffic flow data, the future vehicle attribute data, the road network data, the traffic rules, the environmental data, and the first CA model, the first traffic situation in the target area is predicted; Based on the first traffic situation, the first ARIMA model, the second ARIMA model, and the first CA model are fine-tuned to obtain the third ARIMA model, the fourth ARIMA model, and the second CA model. The third ARIMA model, the fourth ARIMA model, and the second CA model form a cascaded model.

3. The highway traffic risk prediction method according to claim 2, characterized in that, The construction and optimization of the first ARIMA model based on the historical traffic flow data includes: Based on the historical traffic flow data, a first time series graph, a first autocorrelation function, and a first partial autocorrelation function are constructed. The first parameter is obtained based on the first time series plot, the first autocorrelation function, and the first partial autocorrelation function, and the first ARIMA model is constructed based on the first parameter. The first parameter is tuned, and the first ARIMA model is optimized based on the tuned first parameter.

4. The highway traffic risk prediction method according to claim 3, characterized in that, The construction and optimization of the second ARIMA model based on the historical vehicle attribute data includes: A second time series graph, a second autocorrelation function, and a first partial autocorrelation function are constructed based on the historical vehicle attribute data. The second parameters are obtained based on the second time series plot, the second autocorrelation function, and the second partial autocorrelation function, and the second ARIMA model is constructed based on the second parameters. The second parameter is tuned, and the second ARIMA model is optimized based on the tuned second parameter.

5. The highway traffic risk prediction method according to claim 4, characterized in that, The step of fine-tuning the first ARIMA model, the second ARIMA model, and the first CA model based on the first traffic situation to obtain the third ARIMA model, the fourth ARIMA model, and the second CA model includes: Based on the first traffic situation, the first parameter and the second parameter are finely adjusted to obtain the third ARIMA model and the fourth ARIMA model; The first CA model is fine-tuned based on the third ARIMA model and the fourth ARIMA model to obtain the second CA model.

6. The highway traffic risk prediction method according to claim 5, characterized in that, The prediction of traffic risk based on the cascaded model includes: The second traffic situation in the target area is simulated based on the cascaded model; Based on the second traffic situation and the prediction model, the traffic risk is predicted.

7. The highway traffic risk prediction method according to claim 5, characterized in that, Also includes: The historical traffic flow data and the historical vehicle attribute data are cleaned, time-seriesified, stationarity checked, trend identified, and seasonally adjusted. The future traffic flow data, the future vehicle attribute data, the road network data, the traffic rules, and the environmental data are cleaned, integrated, and preprocessed.

8. A computer device, characterized in that, The computer device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the highway traffic risk prediction method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the highway traffic risk prediction method according to any one of claims 1-7.

10. A computer program product comprising computer instructions, characterized in that, When executed by a processor, the computer instructions implement the steps of the method according to any one of claims 1 to 7.

Images