Traffic event dynamic risk prediction method and system, electronic device and storage medium

Abstract

The embodiment of the application provides a traffic event dynamic risk prediction method, system, electronic equipment and storage medium, the method comprises: determining a vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior, wherein the vehicle driving behavior is determined by a preset association relationship between a plurality of vehicle driving data;Obtain first data;The event output type is fused with the first data and the first result is output;In response to the first result, output at least one traffic event dynamic risk prediction result corresponding to the first result. The embodiment of the application can realize comprehensive perception of traffic events, significantly enhance the intelligent reasoning ability of complex scenes, so that the traffic event dynamic risk prediction is more accurate and reliable.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation systems, and in particular to a method, system, electronic device, and storage medium for dynamic risk prediction of traffic incidents. Background Technology

[0002] In recent years, with the rapid expansion of highway networks and the increase in the number of motor vehicles, traffic accidents have become frequent, especially those caused by improper driving, posing a serious threat to public safety. To reduce losses caused by traffic accidents, intelligent transportation systems have become a research hotspot.

[0003] CN115841252A discloses "A Highway Driving Risk Assessment Method Based on Predicted Trajectory," which obtains vehicle location information through a network platform and predicts the vehicle's trajectory based on this information. This allows for the determination of whether there is a potential risk in the interaction between the vehicle and other vehicles, and then a risk warning is issued based on the risk level. However, existing technologies lack accuracy in assessing complex scenarios and cannot perform collision risk assessments, making the judgment of highway traffic incidents unreliable.

[0004] CN114267173A discloses "a method, device and equipment for multi-source data fusion of spatiotemporal characteristics of highways", which realizes the fusion of various multi-source heterogeneous sensor data, retains the unique temporal and spatial characteristics of traffic data, and constructs holographic traffic data to provide travel services for safety and efficiency. It determines the risk level according to the traffic flow status classification standard of the service level of highways in my country. However, considering only the traffic flow status makes the judgment standard singular and it is not practical for the more complex and changeable real driving environment.

[0005] The existing technologies mentioned above typically rely on a single data source (such as a camera or sensor) or a single risk level assessment standard for traffic events, resulting in limited understanding of complex traffic scenarios and difficulty in achieving accurate judgment. Furthermore, the lack of efficient reasoning capabilities prevents deep fusion and comprehensive analysis of multimodal data, limiting analysis to single-modal or simple rule-based scenarios, leading to low intelligence and poor accuracy. This is particularly evident in joint reasoning and dynamic risk planning based on multi-source data; simultaneously, existing technologies lack comprehensive modeling of vehicle trajectories, driving behavior, and environmental factors, making it difficult to identify accident risks in advance and quantify risk levels. In complex dynamic scenarios (such as extreme weather and high-speed driving), existing systems struggle to handle vehicle trajectory prediction and collision risk assessment, failing to achieve comprehensive perception of traffic events and rendering dynamic risk prediction for traffic events unreliable. Summary of the Invention

[0006] In view of this, embodiments of this application provide a method, system, electronic device, and storage medium for dynamic risk prediction of traffic incidents, in order to at least partially solve the above-mentioned problems.

[0007] According to a first aspect of the embodiments of this application, a method for dynamic risk prediction of traffic incidents is provided. The method includes: determining vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior, wherein the vehicle driving behavior is determined by a preset correlation relationship between multiple vehicle driving data; acquiring first data; performing data fusion of the event output type and the first data to output a first result; and outputting at least one dynamic risk prediction result of traffic incidents corresponding to the first result in response to the first result.

[0008] According to a second aspect of the embodiments of this application, a dynamic risk prediction system for traffic incidents is provided, comprising: an acquisition module, configured to acquire vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior, wherein the vehicle driving behavior is determined by a preset correlation relationship between multiple vehicle driving data; further configured to acquire first data; and a data processing module, configured to perform data fusion based on the event output type and the first data to output a first result, and in response to the first result to output at least one dynamic risk prediction result for traffic incidents corresponding to the first result.

[0009] According to a third aspect of the present application, an electronic device is provided, comprising: a processor and a memory storing a program, wherein the program includes instructions that, when executed by the processor, cause the processor to perform the traffic event dynamic risk prediction method as described in the first aspect.

[0010] According to a fourth aspect of the embodiments of this application, a computer storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the traffic event dynamic risk prediction method as described in the first aspect.

[0011] In summary, the embodiments of this application achieve comprehensive perception of traffic events through data fusion from multiple data sources, significantly enhancing the intelligent reasoning capability for complex scenarios, effectively improving the system's judgment accuracy and reliability, thereby making intelligent perception and dynamic risk prediction of highway traffic events more accurate and reliable. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0013] Figure 1 This is a flowchart illustrating the dynamic risk prediction method for traffic incidents according to an exemplary embodiment of this application.

[0014] Figure 2 A schematic diagram of the structure of a traffic incident dynamic risk prediction system, which is an exemplary embodiment of this application;

[0015] Figure 3 This is a schematic diagram of the structure of an electronic device that is an exemplary embodiment of this application. Detailed Implementation

[0016] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the technical solutions in the embodiments of this application will be clearly and thoroughly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art should fall within the protection scope of the embodiments of this application.

[0017] The specific implementation of the embodiments of this application will be further described below with reference to the accompanying drawings.

[0018] See Figure 1 The traffic incident dynamic risk prediction method in this embodiment mainly includes the following steps:

[0019] Step S102: Determine the vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior.

[0020] In this embodiment, vehicle driving behavior is determined by a preset correlation between multiple vehicle driving data.

[0021] Understandably, when vehicle driving data satisfies a preset correlation between at least one or more sets of vehicle driving data, at least one vehicle driving behavior is determined. The preset correlation between the vehicle driving data can be defined as one condition or a combination of multiple condition items. In some embodiments, the specific content of the preset correlation between the vehicle driving data can be arbitrarily adjusted according to actual needs, and is not limited here.

[0022] Example 1: Vehicle driving behavior is determined based on a preset correlation between multiple vehicle driving data. Existing technologies have relatively simple criteria for determining vehicle driving behavior, such as acquiring vehicle driving status through road surface sensors. More specifically, vehicle driving status is captured by cameras and / or infrared sensors installed on the road surface. However, when sudden events occur on the road surface, it is impossible to make a correct judgment on vehicle driving behavior. For example, when the vehicle in front is involved in an accident, the following vehicle needs to avoid it, which may involve illegal lane changes or failure to use turn signals as required by traffic regulations. Therefore, the methods for acquiring vehicle driving data in this application include, but are not limited to: acquiring vehicle driving data through a vehicle networking platform; acquiring vehicle driving data through vehicle sensors; and acquiring vehicle driving data through a third-party platform. Those skilled in the art will understand that average vehicle driving data can be obtained through various publicly available platforms, including but not limited to the public platforms of local traffic control authorities.

[0023] The preset relationships between the driving data of each vehicle may include: AND logical relationship, OR logical relationship.

[0024] For example, vehicle driving data includes, but is not limited to: vehicle driving status data, such as: average speed of the vehicle during driving, engine speed for economical driving, rapid acceleration / 1000 km / h, rapid braking / 1000 km / h, number / frequency of lane changes, frequency of gear shifting, etc.; vehicle headlight status (e.g., high beams on / off, low beams on / off, turn signals on / off, etc.); vehicle idling time; vehicle-side function status data, such as: handbrake on / off, parking gear on / off, doors on / off, windows on / off, seat belt status, etc.; and vehicle driving environment data, such as: road type and road surface conditions. The judgment conditions for vehicle driving behavior in this application are determined based on the preset correlation between various vehicle driving data, which provides greater reliability in complex driving environments.

[0025] Example 2, exemplarily, in some preferred embodiments, the road surface condition is obtained through weather data analysis from a third-party platform. For example, if the current weather at the location of the target vehicle's road surface is rainy, the road surface condition is determined to be slippery / or poses an accident risk; if the current weather at the location of the target vehicle's road surface is sunny, the road surface condition is determined to be dry / or safe.

[0026] Furthermore, the association between various vehicle driving data related to vehicle driving behavior is set, the logical relationship between various vehicle driving data is edited, the preset association between various vehicle driving data is obtained, and the judgment conditions for vehicle driving behavior are generated based on the preset association between various vehicle driving data.

[0027] For example, in some embodiments, the preset association relationship can be set according to the following formula:

[0028]

[0029] Where St is the predicted position, S0 is the initial position, V0 is the initial velocity, a is the acceleration, and t is the time.

[0030] Specifically, in some embodiments, the location information of the target vehicle appearing under the camera is obtained through a third-party platform, such as a monitoring platform for the road surface where the target vehicle is traveling. For example, if the target vehicle is at point S0, the camera can be named "KXX+NN". More specifically, in road surveying, the mileage marker "KXX+NN" represents the horizontal distance from the centerline of the route along the curve to the starting point of the route (where the starting point is represented as "K0+000m"). It is also called the route centerline marker. To facilitate the marking of the position and length of the route centerline, a mileage marker is set at regular intervals along the centerline starting from the starting point of the route, serving as the basis for measuring the longitudinal and transverse sections of the route. "KNN" indicates how many kilometers away from the start of the highway, and "NN" is the distance from the next road sign relative to the starting point of the route.

[0031] For example, in some embodiments, the calculation method is as follows: K is the first letter of the English word Kilometers. K0+752 means that the distance from the starting point to this marker is 752 meters. K1+232 means that the distance from the starting point to this marker is 1232 meters. This is used as one of the bases for determining the vehicle's travel distance.

[0032] In other embodiments, the distance between two points can also be obtained based on road topology. Specifically, road topology refers to a highway connecting point A to point B, such as the Chengdu-Chongqing Expressway. The arrangement of the cameras along the road surface where the target vehicle is traveling is named "KXX+NN".

[0033] It is understandable that the specific details of the preset relationships between vehicle driving data can be adjusted arbitrarily according to actual needs, and no restrictions are imposed here.

[0034] For example, in some embodiments, the event output types include, but are not limited to: stopping in the emergency lane, speeding, illegal lane changing, and riding across the emergency lane.

[0035] Furthermore, based on the preset correlation between the driving data of each vehicle, the driving behavior of the target vehicle is determined. In some embodiments, the driving behavior includes, but is not limited to, normal driving and abnormal driving. Further still, in some embodiments, abnormal driving includes, but is not limited to, speeding, driving in the wrong lane, and driving in the wrong direction. For example, in some embodiments, the method for determining abnormal driving includes, but is not limited to: if any vehicle driving data differs from the average vehicle driving data, the vehicle driving behavior is determined to be abnormal. For example, if the target vehicle's speed is 120 km / h, and the average vehicle speed at the location of the road where the target vehicle is traveling is 90 km / h, the vehicle driving behavior of the target vehicle is determined to be abnormal based on the above vehicle driving data. In other embodiments, for example, abnormal driving behavior of the target vehicle is determined by obtaining photo information and / or video stream information of the target vehicle during its driving process through sensors on the road surface where the target vehicle is traveling. For example, the target vehicle crosses lanes (e.g., occupies the adjacent lane), or the target vehicle's driving direction differs from the driving direction of the road surface where the target vehicle is traveling (e.g., driving in the wrong direction, driving not in a straight line, etc.).

[0036] Among them, the vehicle driving behavior is associated with any event type or a combination of multiple event types, which can be adjusted arbitrarily according to actual needs, and there are no restrictions here.

[0037] Furthermore, by way of example, in some embodiments the event output type also includes the following information: timestamp information (e.g., "5:32:02"), location information (e.g., "C segment"), and vehicle characteristic information (e.g., vehicle type information, license plate information, etc.).

[0038] Step S104: Obtain the first data.

[0039] In existing technologies, data collection is relatively simple. Commonly used data collection content includes video stream data of the road surface where the vehicle is traveling and / or data collected by the vehicle-mounted data collector. Since the data collection content is too simple, it cannot output more reliable judgment results. Therefore, the first data in this application includes, but is not limited to, the following: vehicle speed data, vehicle lane traffic flow data, vehicle lane density data, vehicle trajectory data, vehicle speed data, weather data, and map information data.

[0040] More specifically, in some embodiments, vehicle speed data, vehicle lane traffic flow data, and vehicle lane density data are acquired through traffic sensors, including but not limited to forward-facing cameras, surround-view cameras, in-vehicle monitoring cameras, millimeter-wave radar, etc.

[0041] Example 3 involves acquiring vehicle trajectory data and vehicle speed data through vehicle network data; more specifically, it involves acquiring weather data and map information data through a third-party platform. For example, in some embodiments, the weather information may include one or more of the following: visibility, rainfall, wind speed, etc. The third-party platforms for acquiring weather data include, but are not limited to: the China Meteorological Data Network, NOAA (National Oceanic and Atmospheric Administration) global ground station observation data, etc.

[0042] Example 4, exemplarily, in some embodiments, the map information also includes road topology, such as the number of lanes, slope, curvature, etc., of the road surface on which the target vehicle travels. The third-party platforms for acquiring the map information data include, but are not limited to, platforms such as geospatial data clouds and the Chinese Academy of Sciences data cloud.

[0043] Step S106: Perform data fusion between the event output type and the first data to output the first result.

[0044] The first data obtained in step S104 is preprocessed. Further preprocessing includes, but is not limited to: removing outliers and redundant data from the first data; and performing spatiotemporal alignment on data of different modalities (e.g., video frames, numerical data, and text information). Compared to existing technologies, this application's dynamic risk prediction of traffic incidents based on multimodal data has higher reliability and accuracy. Because the data involved is relatively complex, the output format after data preprocessing is more uniform, facilitating subsequent data processing and thus improving data processing speed.

[0045] Example 5, exemplarily, as those skilled in the art will understand, the preprocessing of the first data includes: standardizing the first data and outputting it as text information; in other embodiments, the standardization of the first data outputs information in JSON format.

[0046] Furthermore, in some embodiments, the first data preprocessing result is used as the first sub-processing result. Even further, the event type obtained in step S102 and the first sub-processing result are input together into a large model for multimodal data fusion to further obtain the first data output result. The large model includes, but is not limited to, GPT-4 (Generative Pre-trained Transformer4), Transformer, and deep seek.

[0047] Step S108: In response to the first result, output at least one dynamic risk prediction result for a traffic event corresponding to the first result.

[0048] In some embodiments, at least one traffic event dynamic risk prediction result corresponding to the first output result is the target vehicle traffic event dynamic risk prediction level.

[0049] More specifically, the risk levels indicated by the first result obtained in step S106 are edited, and further, the risk level classification results are obtained based on the first result in step S106, which serve as the dynamic risk prediction results for the traffic incident.

[0050] Example 6, exemplarily, in some embodiments, the dynamic risk prediction level of traffic incidents is classified according to the following: if the relative distance between the target vehicle and the third-party vehicle in the first result is greater than the safe distance, the risk level is classified as low risk; if the relative distance between the target vehicle and the third-party vehicle in the first result is less than or equal to the safe distance, the risk level is classified as medium risk; if the relative distance between the target vehicle and the third-party vehicle in the first result is equal to or less than the alarm distance, the risk level is classified as high risk. Specifically, in some embodiments, the safe distance can be set to 500 meters, and the alarm distance can be set to 200 meters. More specifically, the distance between the target vehicle and the third-party vehicle can be obtained according to Equation 2. The specific content can be arbitrarily adjusted according to actual needs, and is not limited here.

[0051] d = |S_target_vehicle - S_third_vehicle |, (Equation 2)

[0052] Where d is the straight-line distance between the target vehicle and the third-party vehicle, and S is the location of the vehicle.

[0053] The risk level output based on the relative distance between the target vehicle and the third-party vehicle in the first result is used as the first response result. Further, the second response outputs a second response result based on a preset impact factor. The second response result is compared with the first response result, and a higher risk level is output. That is, the risk level is divided based on the relative distance between the target vehicle and the third-party vehicle in the first result, and a risk level with a higher relative danger level is output. The preset impact factor is obtained based on the first data.

[0054] In some embodiments, the target vehicle driving environment is used as one of the preset influencing factors. Specifically, the dynamic risk prediction level of traffic events is further classified according to the following: if the target vehicle driving environment in the first result is sunny weather, the risk level is classified as low risk; if the target vehicle driving environment in the first result is rainy or snowy weather, the risk level is classified as medium risk; if there is abnormal data in the target vehicle driving environment in the first result, the risk level is classified as high risk; wherein, any data in the first result that far exceeds the average data is identified as abnormal data. It should be understood that the influencing factors can be preset according to actual needs, and this solution does not impose any restrictions on this.

[0055] Furthermore, by comparing the results of Level 1 response with those of Level 2 response, a risk level with a relatively higher danger level is output as the result of the dynamic risk prediction level classification for traffic incidents.

[0056] For example, if the relative distance between the target vehicle and the third-party vehicle in the first result is greater than the safe distance, the risk level of the first-level response result is classified as low risk; when the target vehicle's driving environment is one of the preset influencing factors, if the target vehicle's driving environment in the first result is rainy or snowy weather, the risk level of the second-level response result is classified as medium risk. Further comparison between the first-level response result and the second-level response result will output a risk level with a higher relative danger level, that is, the output of the traffic event dynamic risk prediction level classification result is medium risk.

[0057] In some embodiments, the dynamic risk prediction results of traffic incidents are associated with at least one emergency response plan, wherein the emergency response plan is associated with the dynamic risk prediction level of the traffic incident. It should be noted that the emergency response plan is not the focus of this invention and can be arbitrarily adjusted according to actual needs, without limitation herein.

[0058] Example 8: If the dynamic risk prediction level of a traffic incident is high, an emergency warning is sent to the command center to request the dispatch of rescue resources and nearby cameras are notified to lock the footage; if the dynamic risk prediction level of a traffic incident is medium, a warning is sent to nearby vehicle drivers via the vehicle network; if the dynamic risk prediction level of a traffic incident is low, the incident is recorded and vehicle driving behavior is continuously monitored.

[0059] This embodiment of the disclosure achieves comprehensive perception of traffic events through data fusion from multiple data sources, significantly enhancing the intelligent reasoning capability for complex scenarios and effectively improving the system's judgment accuracy and reliability, thereby making intelligent perception and dynamic risk prediction of highway traffic events more accurate and reliable.

[0060] Figure 2 A schematic diagram of the structure of a traffic incident dynamic risk prediction system, which is an exemplary embodiment of this application;

[0061] The acquisition module 201 is used to acquire vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior, wherein the vehicle driving behavior is determined by a preset association relationship between multiple vehicle driving data.

[0062] Also, it is used to obtain the first data;

[0063] Data processing module 202, based on the above, performs data fusion with the first data to output a first result, and in response to the first result, outputs at least one traffic event dynamic risk prediction result corresponding to the first result;

[0064] Another embodiment of this application provides a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the processing steps of the above-described dynamic risk prediction methods for traffic events.

[0065] See Figure 3 The diagram shown is a schematic diagram of an electronic device according to an embodiment of this application. The specific embodiments of this application do not limit the specific implementation of the electronic device.

[0066] like Figure 3 As shown, the electronic device may include: a processor 302, a communications interface 304, a memory 306, and a communications bus 308.

[0067] in:

[0068] The processor 302, communication interface 304, and memory 306 communicate with each other via communication bus 308.

[0069] Communication interface 304 is used to communicate with other electronic devices or servers.

[0070] The processor 302 is used to execute program 310, specifically to execute the relevant steps in the above-described embodiment of the dynamic risk prediction method for traffic incidents.

[0071] Specifically, program 310 may include program code that includes computer operation instructions.

[0072] Processor 302 may be a CPU, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. A smart device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0073] Memory 306 is used to store program 310. Memory 306 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0074] Program 310 may include multiple computer instructions. Specifically, program 310 may use multiple computer instructions to cause processor 302 to perform the operations corresponding to the dynamic risk prediction method for traffic events described in any of the foregoing method embodiments.

[0075] The specific implementation of each step in procedure 310 can be found in the corresponding descriptions of the steps and units in the above method embodiments, and has corresponding beneficial effects, which will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the devices and modules described above can be referred to the corresponding process descriptions in the foregoing method embodiments, and will not be repeated here.

[0076] The electronic device in this application embodiment enables comprehensive perception of traffic events, significantly enhances the intelligent reasoning ability for complex scenarios, and effectively improves the system's judgment accuracy and reliability, thereby making intelligent perception and dynamic risk prediction of highway traffic events more accurate and reliable.

[0077] This application also provides a computer-readable storage medium storing instructions for causing a machine to perform the methods described herein. Specifically, a system or apparatus equipped with a storage medium storing software program code that implements the functions of any of the embodiments described above, and enabling a computer (or CPU or MPU) of the system or apparatus to read and execute the program code stored in the storage medium.

[0078] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of this application.

[0079] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0080] This application also provides a computer program product, including computer instructions that instruct a computing device to perform any corresponding operation in the above-described plurality of method embodiments.

[0081] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this application can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this application.

[0082] The methods described in the embodiments of this application can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code downloaded over a network that is originally stored in a remote recording medium or a non-transitory machine-readable medium and will be stored in a local recording medium. Thus, the methods described herein can be processed by software stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., RAM, ROM, flash memory, etc.) capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the navigation methods described herein are implemented. Furthermore, when a general-purpose computer accesses the code used to implement the navigation methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for executing the navigation methods shown herein.

[0083] Those skilled in the art will recognize that the units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this application.

[0084] The above embodiments are only used to illustrate the embodiments of this application, and are not intended to limit the embodiments of this application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of this application. Therefore, all equivalent technical solutions also fall within the scope of the embodiments of this application, and the patent protection scope of the embodiments of this application should be defined by the claims.

Claims

1. A method for dynamic risk prediction of traffic incidents, characterized in that, The method includes: Determine vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior, wherein the vehicle driving behavior is determined by a preset association relationship between multiple vehicle driving data; Get the first data; The event output type is fused with the first data to output a first result; In response to the first result, output at least one dynamic risk prediction result for a traffic event corresponding to the first result; The dynamic risk prediction results for traffic incidents are obtained through the following steps: Edit the risk levels indicated by the first result; Based on the first result, a risk level classification result is obtained, which serves as the dynamic risk prediction result for the traffic incident. The risk level is obtained through the following steps: The system prioritizes responding to the relative distance between the target vehicle and the third-party vehicle in the first result and outputs a first response result. When the relative distance between the target vehicle and the third-party vehicle in the first result is greater than the safe distance, the risk level is classified as low risk; when the relative distance between the target vehicle and the third-party vehicle in the first result is less than or equal to the safe distance, the risk level is classified as medium risk; when the relative distance between the target vehicle and the third-party vehicle in the first result is equal to or less than the alarm distance, the risk level is classified as high risk. The secondary response outputs a second response result based on a preset impact factor. The second response result is compared with the first response result, and a higher risk level is output. The preset influence factor is obtained based on the first data.

2. The method according to claim 1, characterized in that, The following steps are used to determine vehicle driving behavior: Obtain vehicle driving data; Edit the preset relationships between the driving data of each vehicle; The vehicle driving behavior is determined based on the preset correlation between the driving data of each vehicle.

3. The method according to claim 1, characterized in that, The first data includes: vehicle speed data, vehicle lane traffic flow data, vehicle lane density data, vehicle trajectory data, weather data, and map information data.

4. The method according to claim 1, characterized in that, The step of fusing the event output type with the first data to output a first result includes: The first data is preprocessed to obtain the first sub-processing result; The event output type and the first sub-processing result are input together into the trained large model to obtain the first data output result.

5. A traffic incident dynamic risk prediction system, said system being used to execute the method as described in any one of claims 1 to 4, characterized in that, include: The acquisition module is used to acquire vehicle driving behavior and at least one event output type corresponding to the vehicle driving behavior, wherein the vehicle driving behavior is determined by a preset association relationship between multiple vehicle driving data. Also, it is used to obtain the first data; The data processing module fuses the event output type with the first data to output a first result, and in response to the first result, outputs at least one dynamic risk prediction result for a traffic event corresponding to the first result.

6. An electronic device, comprising: processor; And the memory for storing programs, The program includes instructions that, when executed by the processor, cause the processor to perform the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 4.

Images