C-V2X-based vehicle driving early warning method and system

Abstract

The invention discloses a C-V2X-based vehicle driving early warning method and system, relates to the technical field of auxiliary driving, and solves the problems that in the prior art, a traditional TTC model only considers the speed and distance of a vehicle, is not comprehensive enough and does not consider factors such as acceleration, brake pressure and steering angle. According to the method, a multi-factor collision time model and a collision probability model are constructed by collecting multi-source data such as vehicle states, environmental perception and traffic signals; compared with a traditional TTC model, the method comprehensively considers key factors such as the acceleration, the brake pressure and the steering angle, and the collision risk is evaluated more accurately. And the collision probability model further calculates the collision probability through a machine learning algorithm by using the data, so that a more reliable decision basis is provided for a vehicle driving auxiliary system. Finally, early warning signals of different levels are generated according to the collision risk, a driver is reminded to take measures in time, the accident risk is effectively reduced, and the driving safety and the traffic efficiency are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of driver assistance technology, and in particular to a vehicle driving warning method and system based on C-V2X. Background Technology

[0002] C-V2X (Cellular Vehicle-to-Everything) is an advanced vehicle-to-everything (V2X) communication technology that enables high-speed, low-latency communication between vehicles (V2V), vehicles and infrastructure (V2I), vehicles and pedestrians (V2P), and vehicles and networks (V2N) through cellular network infrastructure. This technology provides robust support for vehicle driver assistance systems, particularly in collision prediction and warning. With C-V2X, vehicles can receive real-time information from surrounding vehicles and infrastructure, such as speed, location, and direction of travel, allowing them to anticipate potential collision risks and issue warnings. For example, when a vehicle approaches a vehicle braking suddenly ahead, the C-V2X system can quickly transmit this information to vehicles behind, alerting the driver to take braking measures and effectively reducing the probability of a collision. Furthermore, C-V2X technology can be integrated with autonomous driving systems to further enhance vehicle active safety performance, laying a solid foundation for the development of intelligent transportation.

[0003] While C-V2X technology offers numerous advantages for vehicle driving assistance, it still has some shortcomings in collision risk prediction. Traditional time-of-collision (TTC) models only consider vehicle speed and relative distance. While this method can predict collision risk to some extent, its calculations are often incomplete and inaccurate. For example, TTC models fail to adequately consider key factors such as vehicle acceleration, braking pressure, and steering angle, which are crucial for collision risk assessment in real-world driving scenarios. Acceleration reflects the vehicle's acceleration or deceleration trend, braking pressure reflects the intensity of braking, and steering angle directly affects the vehicle's trajectory. In complex traffic environments, relying solely on speed and distance to assess collision risk can lead to misjudgments, especially during emergency braking, rapid acceleration, or sudden lane changes. Therefore, traditional TTC models have significant limitations in collision risk prediction and cannot meet the high-precision warning requirements of modern intelligent transportation systems.

[0004] Therefore, a vehicle driving warning method and system based on C-V2X is needed. Summary of the Invention

[0005] To address the shortcomings of existing traditional TTC models, which only consider vehicle speed and distance and lack comprehensiveness by neglecting factors such as acceleration, braking pressure, and steering angle, resulting in inadequate collision risk prediction, this invention provides a C-V2X-based vehicle driving warning method and system. This method improves the reliability and effectiveness of collision warning systems by introducing a more comprehensive collision risk assessment model that considers multi-dimensional data including vehicle motion state, driver behavior, and environmental factors. The specific technical solution is as follows: A vehicle driving warning method based on C-V2X includes the following steps: Collect data, including vehicle status data, environmental perception data, and traffic signal data; A multi-factor collision time model is constructed based on multi-source data, as detailed below: In the formula, It is the distance between the two cars. It is relative velocity. It is relative acceleration. It is the acceleration influence coefficient; Based on vehicle status data, environmental perception data, traffic signal data, and multi-source data, a multi-factor collision time model is constructed and the output is used to build and train a collision probability model. Different levels of warning signals are generated based on the level of collision risk.

[0006] Preferably, the vehicle status data includes vehicle braking data, vehicle steering data, and vehicle acceleration data, wherein: In addition to position, speed, and direction of travel, vehicle braking data also includes vehicle braking pressure and braking frequency data. Vehicle steering data includes the vehicle's steering angle and steering speed; Vehicle acceleration data includes both longitudinal and lateral acceleration.

[0007] Preferably, the environmental sensing data includes meteorological data and road condition data, wherein: Meteorological data is mainly obtained through meteorological sensors on the vehicle or from roadside units, including real-time meteorological data such as rainfall, wind speed, and visibility. Road condition data includes the degree of slipperiness and icing of the road, obtained through vehicle tire sensors or from roadside units.

[0008] Preferably, the traffic signal data includes traffic light status data and traffic flow data, wherein: Traffic light status data is obtained through V2I communication to obtain the real-time status and remaining time of traffic lights.

[0009] Traffic flow data is obtained from roadside units or cloud platforms, including real-time traffic flow information such as vehicle volume and speed distribution.

[0010] Preferably, the process for obtaining the collision probability model is as follows: Data collection: In addition to acquiring vehicle status data, environmental perception data, and traffic signal data, it also acquires driver driving habit data and vehicle hardware performance data; Feature extraction includes multi-factor collision time features, driver driving habit features, and vehicle hardware performance features; Data labeling: Labeling includes collision event labeling and non-collision event labeling. Normal driving data is labeled as non-collision events, and accident records are labeled as collision events. The model is trained based on the dataset formed after feature extraction and data annotation to obtain a collision probability model, and the trained collision probability model is deployed to the vehicle or cloud platform.

[0011] Preferably, the driver's driving habit data includes the frequency of rapid acceleration, the frequency of rapid braking, the frequency of lane changing, speeding behavior, and following distance. Among them, the frequency of rapid acceleration is the frequency of rapid acceleration by the driver, the frequency of rapid braking is the frequency of rapid braking by the driver, the frequency of lane changing is the frequency of lane changing by the driver, the frequency of speeding behavior is the frequency of speeding by the driver, and the following distance is the average following distance between the driver and the vehicle in front. Vehicle hardware performance data includes at least braking performance, acceleration performance, steering performance, and vehicle lifespan. Braking performance is represented by braking distance and braking time obtained through braking tests. Acceleration performance is represented by acceleration time and maximum acceleration obtained through acceleration tests. Steering performance is represented by steering sensitivity and maximum steering angle obtained through steering tests. Vehicle lifespan is assessed and quantified by vehicle mileage and years of use.

[0012] Preferably, the multi-factor collision time characteristics are represented by a multi-factor collision time model; A driver's driving habits include at least the following: frequency of rapid acceleration Emergency braking frequency Lane change frequency Frequency of speeding and average following distance .

[0013] Vehicle hardware performance characteristics include at least: braking distance Acceleration time Steering sensitivity and vehicle life indicators Among them, vehicle life indicators are expressed by at least one of mileage or years of use.

[0014] A C-V2X-based vehicle driving warning system, applied to the method described above, includes: Vehicle-side equipment includes a GNSS positioning module, a sensor module, a C-V2X communication module, and a data processing and decision-making module. The GNSS positioning module acquires the vehicle's real-time position and speed. The sensor module, including cameras and millimeter-wave radar, is used to perceive the surrounding environment. The C-V2X communication module communicates with other vehicles (V2V) and roadside units (V2I). The data processing and decision-making module fuses sensor data and received C-V2X information to determine the presence of collision risks. Roadside units: Roadside units are deployed in key traffic areas, such as intersections and highway entrances and exits, to sense traffic conditions and broadcast early warning information; Cloud platform: Used to receive data uploaded by vehicles, perform big data analysis, and optimize traffic flow.

[0015] A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the C-V2X-based vehicle driving warning method as described above.

[0016] A processor for running a program, wherein the program, when running, executes the C-V2X-based vehicle driving warning method as described above.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention integrates multiple information sources, including vehicle status data, environmental perception data, and traffic signal data, to construct a more comprehensive multi-factor time-of-collision (TTC) model, and further develops a collision probability model based on this. This comprehensive approach can overcome the shortcomings of traditional TTC models and more accurately assess collision risk. Vehicle status data provides real-time vehicle motion information, environmental perception data reflects road and weather conditions, and traffic signal data helps vehicles understand traffic conditions in advance. The fusion of these data allows the collision time model to comprehensively consider key factors such as acceleration, braking pressure, and steering angle, thereby more accurately predicting collision risk. The collision probability model further utilizes this data, using machine learning algorithms to calculate the probability of a collision, providing a more reliable decision-making basis for vehicle driving assistance systems. Finally, different levels of warning signals are generated based on the level of collision risk, which can promptly remind drivers to take appropriate measures, effectively reducing the possibility of accidents and improving driving safety and traffic efficiency. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0019] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0022] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] In one embodiment of the present invention, a vehicle driving warning method based on C-V2X is provided, such as... Figure 1 As shown, it includes the following steps: Step 1: Collect data, including vehicle status data, environmental perception data, and traffic signal data; The vehicle status data includes vehicle braking data, vehicle steering data, and vehicle acceleration data.

[0025] In addition to position, speed, and direction of travel, vehicle braking data also includes braking pressure and braking frequency. Braking pressure and braking frequency are collected separately because these data can help determine whether the vehicle is in an emergency braking situation.

[0026] Vehicle steering data includes the vehicle's steering angle and steering speed, and is mainly used to determine whether the vehicle is changing lanes or turning.

[0027] Vehicle acceleration data includes the vehicle's longitudinal and lateral accelerations, which are measured by accelerometers. Vehicle acceleration data can be used to predict the vehicle's trajectory.

[0028] Environmental perception data includes meteorological data and road condition data.

[0029] Meteorological data is primarily obtained through real-time meteorological sensors on vehicles or from roadside units, including rainfall, wind speed, and visibility. Under severe weather conditions, the warning system can adjust warning thresholds based on this data.

[0030] Road condition data includes the degree of wetness and iciness of the road, which can be obtained through vehicle tire sensors or from roadside units.

[0031] Traffic signal data includes traffic light status data and traffic flow data.

[0032] Traffic light status data: Real-time status (red light, green light, yellow light) and remaining time of traffic lights are obtained through V2I communication.

[0033] Traffic flow data: Real-time traffic flow information, including vehicle flow and speed distribution, is obtained from roadside units or cloud platforms.

[0034] Step 2: Construct a multi-factor collision time (TTC) model and a collision probability model based on multi-source data; For example, traditional TTC models only consider vehicle speed and distance. This embodiment introduces more factors (such as acceleration, braking pressure, steering angle, etc.) to more accurately assess collision risk, as detailed below: In the formula, It is the distance between the two cars. It is relative velocity. It is relative acceleration. It is the acceleration influence coefficient.

[0035] For example, the process of obtaining the collision probability model is as follows: S1: Data Acquisition: In addition to acquiring the data collected in step one, driver driving habit data and vehicle hardware performance data are also acquired.

[0036] The driver driving habit data includes the frequency of rapid acceleration, the frequency of rapid braking, the frequency of lane changing, speeding behavior, and following distance. The rapid acceleration frequency is the frequency of rapid acceleration by the driver, the rapid braking frequency is the frequency of rapid braking by the driver, the lane changing frequency is the frequency of lane changing by the driver, the speeding behavior is the frequency of speeding by the driver, and the following distance is the average following distance between the driver and the vehicle in front.

[0037] Vehicle hardware performance data includes at least braking performance, acceleration performance, steering performance, and vehicle lifespan. Braking performance is represented by braking distance and braking time obtained through braking tests; acceleration performance is represented by acceleration time and maximum acceleration obtained through acceleration tests; steering performance is represented by steering sensitivity and maximum steering angle obtained through steering tests; and vehicle lifespan is assessed and quantified by vehicle mileage and years of use.

[0038] S2: Feature extraction: including multi-factor collision time features, driver driving habit features, and vehicle hardware performance features.

[0039] Among them, the multi-factor collision time characteristics are represented by the multi-factor collision time (TTC) model; A driver's driving habits include at least the following: frequency of rapid acceleration Emergency braking frequency Lane change frequency Frequency of speeding and average following distance .

[0040] Vehicle hardware performance characteristics include at least: braking distance Acceleration time Steering sensitivity and vehicle life indicators Among them, vehicle life indicators are expressed by at least one of mileage or years of use.

[0041] That is, the feature vector obtained after feature extraction is represented as follows: S3: Data Labeling: Labeling includes collision event labeling and non-collision event labeling. Normal driving data is labeled as non-collision events, and accident records are labeled as collision events. For example, 1 indicates a collision occurred, and 0 indicates no collision occurred. For example, the label vector is represented as: .

[0042] In addition to accident record data, the data marked as collision events also includes simulation data, that is, simulated collision accident data.

[0043] For example, the following are the specific steps for obtaining simulation data based on CarSim software (which has vehicle dynamics model, sensor model and data export tool installed by default) to generate simulation data for training collision probability model: S301: Select a scene template and edit the scene; For example, in CarSim, select "New Project," then choose a template suitable for a traffic scenario, such as "City Road" or "Highway." In the scene editor, add road elements (such as straight roads, curves, and intersections). Add traffic signs and traffic lights. Add other vehicles (target vehicles) and pedestrians, and set the physical properties of the road (such as the coefficient of friction).

[0044] S302: Configure the vehicle and sensors, including selecting the vehicle model and configuring the sensors; For example, select a vehicle model from the vehicle library and place it in the scene; vehicle models include cars, trucks, etc. For example, sensors, including cameras, radar, and lidar, are added to the vehicle model. In addition, sensor parameters need to be configured, including: setting the resolution, field of view, and sampling frequency of the camera; setting the detection range, angular resolution, and sampling frequency of the radar; and setting the point cloud density, scanning range, and sampling frequency of the lidar.

[0045] S303: Set simulation parameters, including simulation time, vehicle behavior, environmental variables, and collision events; For example, the total simulation run time is set to 30 seconds, and driving behaviors are set for the vehicle and the target vehicle. Driving behaviors include constant speed driving, acceleration, braking, and lane changing. In addition, weather conditions and lighting conditions are set. Weather conditions include sunny, rainy, and foggy days, and lighting conditions include day and night. Potential collision events are set in the scene, such as the vehicle in front braking suddenly and a pedestrian crossing the road.

[0046] S304: Start the simulation and collect data.

[0047] For example, clicking the "Run Simulation" button allows the vehicle to drive in a virtual scene. During the simulation, the vehicle's motion status data is recorded in real time, including position, speed, acceleration, braking pressure, and steering angle; data collected by sensors are also recorded, including camera images, radar point clouds, and lidar point clouds; environmental variables and the occurrence of collision events are also recorded, including weather and lighting.

[0048] S305: Export the data and convert it into a format suitable for training the collision probability model; For example, data can be exported as follows: In CarSim, select the "Output Settings" menu; select the data type to be exported, including vehicle status data, sensor data, and collision event data; set the export format, including CSV, MATLAB, Excel, etc.; specify the export path, and click the "Export" button.

[0049] For example, converting the data into a format suitable for training a collision probability model includes: merging vehicle state data and sensor data into a single data frame; labeling collision events as 1 (collision occurred) or 0 (no collision occurred); and normalizing the data to normalize all feature values ​​to the range [0, 1].

[0050] S4: Model selection and training. Details are as follows: S401: Model selection: For example, in this embodiment, a deep learning model is used; practically, a convolutional neural network (CNN) is selected.

[0051] It should be understood that, in this step, those skilled in the art can adaptively select a model for training based on the actual application scenario and the actual data collected, and this application does not impose any limitations.

[0052] S402: After selecting the model, normalize all feature values ​​to the range of [0, 1] and divide the dataset into training set and test set, for example, the ratio is 80% (training set) and 20% (test set).

[0053] S403: Train the model using the training set data, tune the model's hyperparameters to optimize performance, validate the model's performance using the test set data, and evaluate model metrics, including accuracy, recall, and F1 score. S404: Evaluate the model's performance using the confusion matrix, ROC curve, and AUC value. Adjust the model structure or hyperparameters based on the evaluation results to further optimize the model's performance. Use cross-validation methods (such as K-fold cross-validation) to evaluate the model's stability and generalization ability.

[0054] S5: Deploy the trained model to the vehicle or cloud platform.

[0055] For example, during model training, the logistic regression model is represented as follows: In the formula, Represents the model weights. b This indicates the bias term.

[0056] For example, the loss function is expressed as follows: For example, gradient descent can be used to optimize model parameters.

[0057] Step 3: Generate different levels of warning signals based on the level of collision risk, and dynamically adjust the warning thresholds according to real-time weather conditions, road conditions, and traffic flow. In one embodiment of the present invention, a C-V2X-based vehicle driving warning system is provided, comprising: Vehicle-side equipment includes a GNSS positioning module, a sensor module, a C-V2X communication module, and a data processing and decision-making module. The GNSS positioning module is used to acquire the vehicle's real-time position and speed. The sensor module includes cameras, millimeter-wave radar, etc., to perceive the surrounding environment. The C-V2X communication module is used to communicate with other vehicles (V2V) and roadside units (V2I). The data processing and decision-making module fuses sensor data and received C-V2X information to determine whether there is a collision risk.

[0058] For example, the vehicle periodically collects data such as its position, speed, acceleration, braking pressure, and steering angle, and also gathers meteorological and road condition data through its sensors. Vehicles periodically broadcast and receive vehicle status data via C-V2X modules. Additionally, vehicles receive traffic signal status and road hazard information broadcast by roadside units. The vehicle then fuses the sensor data with the received C-V2X data and uses machine learning algorithms to analyze the multi-source data to calculate collision risk.

[0059] Roadside Units (RSUs) are deployed in key traffic areas, such as intersections and highway entrances / exits, to sense traffic conditions and broadcast early warning information.

[0060] For example, the roadside testing unit collects data such as traffic flow, traffic light status, and road hazard conditions through devices such as cameras and radar. It processes the collected traffic data in real time, generates early warning information, and broadcasts it to vehicles via C-V2X.

[0061] For example, an improved TTC model and a collision probability model are used to assess collision risk, and warning thresholds are dynamically adjusted based on real-time weather and road conditions.

[0062] Cloud platform: Used to receive data uploaded by vehicles, perform big data analysis, and optimize traffic flow.

[0063] In summary, this invention integrates multiple information sources, including vehicle status data, environmental perception data, and traffic signal data, to construct a more comprehensive multi-factor time-of-collision (TTC) model, and further develops a collision probability model based on this. This comprehensive approach can overcome the shortcomings of traditional TTC models and more accurately assess collision risk. Vehicle status data provides real-time vehicle motion information, environmental perception data reflects road and weather conditions, and traffic signal data helps vehicles understand traffic conditions in advance. The fusion of these data allows the collision time model to comprehensively consider key factors such as acceleration, braking pressure, and steering angle, thereby more accurately predicting collision risk. The collision probability model further utilizes this data, using machine learning algorithms to calculate the probability of a collision, providing a more reliable decision-making basis for vehicle driving assistance systems. Finally, different levels of warning signals are generated based on the level of collision risk, which can promptly remind drivers to take appropriate measures, effectively reducing the possibility of accidents and improving driving safety and traffic efficiency.

[0064] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. 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 implementations should not be considered beyond the scope of the invention.

[0065] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical functional division. In actual implementation, there may be other division methods, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored.

[0066] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0067] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A vehicle driving warning method based on C-V2X, characterized in that, Includes the following steps: Collect data, including vehicle status data, environmental perception data, and traffic signal data; A multi-factor collision time model is constructed based on multi-source data, as detailed below: In the formula, It is the distance between the two cars. It is relative velocity. It is relative acceleration. It is the acceleration influence coefficient; Based on vehicle status data, environmental perception data, traffic signal data, and multi-source data, a multi-factor collision time model is constructed and the output is used to build and train a collision probability model. Different levels of warning signals are generated based on the level of collision risk.

2. The vehicle driving warning method based on C-V2X according to claim 1, characterized in that, Vehicle status data includes vehicle braking data, vehicle steering data, and vehicle acceleration data, among which: In addition to position, speed, and direction of travel, vehicle braking data also includes vehicle braking pressure and braking frequency data. Vehicle steering data includes the vehicle's steering angle and steering speed; Vehicle acceleration data includes both longitudinal and lateral acceleration.

3. The vehicle driving warning method based on C-V2X according to claim 1, characterized in that, Environmental perception data includes meteorological data and road condition data, among which: Meteorological data is mainly obtained through meteorological sensors on the vehicle or from roadside units, including real-time meteorological data such as rainfall, wind speed, and visibility. Road condition data includes the degree of slipperiness and icing of the road, obtained through vehicle tire sensors or from roadside units.

4. The vehicle driving warning method based on C-V2X according to claim 1, characterized in that, Traffic signal data includes signal light status data and traffic flow data, among which: Traffic light status data is obtained through V2I communication to obtain the real-time status and remaining time of traffic lights; Traffic flow data is obtained from roadside units or cloud platforms, including real-time traffic flow information such as vehicle volume and speed distribution.

5. A vehicle driving warning method based on C-V2X according to claim 1, characterized in that, The process of obtaining the collision probability model is as follows: Data collection: In addition to acquiring vehicle status data, environmental perception data, and traffic signal data, it also acquires driver driving habit data and vehicle hardware performance data; Feature extraction includes multi-factor collision time features, driver driving habit features, and vehicle hardware performance features; Data labeling: Labeling includes collision event labeling and non-collision event labeling. Normal driving data is labeled as non-collision events, and accident records are labeled as collision events. The model is trained based on the dataset formed after feature extraction and data annotation to obtain a collision probability model, and the trained collision probability model is deployed to the vehicle or cloud platform.

6. A vehicle driving warning method based on C-V2X according to claim 5, characterized in that, Driver driving habit data includes frequency of rapid acceleration, frequency of rapid braking, frequency of lane change, speeding behavior, and following distance. Among them, the frequency of rapid acceleration is the frequency of rapid acceleration by the driver, the frequency of rapid braking is the frequency of rapid braking by the driver, the frequency of lane change is the frequency of lane change by the driver, the frequency of speeding behavior is the frequency of speeding by the driver, and the following distance is the average following distance between the driver and the vehicle in front. Vehicle hardware performance data includes at least braking performance, acceleration performance, steering performance, and vehicle lifespan. Braking performance is represented by braking distance and braking time obtained through braking tests. Acceleration performance is represented by acceleration time and maximum acceleration obtained through acceleration tests. Steering performance is represented by steering sensitivity and maximum steering angle obtained through steering tests. Vehicle lifespan is assessed and quantified by vehicle mileage and years of use.

7. A vehicle driving warning method based on C-V2X according to claim 1, characterized in that, The characteristics of multi-factor collision time are represented by a multi-factor collision time model; A driver's driving habits include at least the following: frequency of rapid acceleration Emergency braking frequency Lane change frequency Frequency of speeding and average following distance ; Vehicle hardware performance characteristics include at least: braking distance Acceleration time Steering sensitivity and vehicle life indicators Among them, vehicle life indicators are expressed by at least one of mileage or years of use.

8. A vehicle driving warning system based on C-V2X, characterized in that, The method applied to any one of claims 1 to 7 includes: Vehicle-side equipment includes a GNSS positioning module, a sensor module, a C-V2X communication module, and a data processing and decision-making module. The GNSS positioning module acquires the vehicle's real-time position and speed. The sensor module, including cameras and millimeter-wave radar, is used to perceive the surrounding environment. The C-V2X communication module communicates with other vehicles and roadside units. The data processing and decision-making module fuses sensor data and received C-V2X information to determine the presence of collision risks. Roadside units: Roadside units are deployed in key traffic areas, such as intersections and highway entrances and exits, to sense traffic conditions and broadcast early warning information; Cloud platform: Used to receive data uploaded by vehicles, perform big data analysis, and optimize traffic flow.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the C-V2X-based vehicle driving warning method according to any one of claims 1 to 7.

10. A processor, characterized in that, The processor is used to run a program, wherein the program executes the C-V2X-based vehicle driving warning method according to any one of claims 1 to 7 when it runs.

Images