Single-lane meeting safety analysis method based on Internet of Vehicles

Abstract

The invention discloses a single-lane vehicle meeting safety analysis method based on the Internet of Vehicles, and relates to the technical field of intelligent traffic, and the method comprises the steps: collecting original data, and transmitting the original data to a roadside edge calculation unit; dividing the original data into a forward driving vehicle data set and a reverse driving vehicle data set according to the driving direction by using a roadside edge calculation unit, establishing a trajectory prediction model according to an improved Kalman filtering algorithm, predicting the driving trajectory and the longitudinal distance value of two vehicles driving in opposite directions, and calculating a minimum safe meeting distance threshold value; judging the longitudinal spacing value and the minimum safe meeting distance threshold value, and sending a meeting risk coefficient to an Internet of Vehicles road test board; and receiving the meeting risk coefficient through an Internet of Vehicles road test board, converting the meeting risk coefficient into an acousto-optic early warning signal, and sending meeting risk early warning information to the vehicle by using a roadside edge calculation unit. According to the invention, through data acquisition, dynamic risk assessment and diversified early warning mechanisms, the vehicle meeting safety is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a single-lane meeting safety analysis method based on vehicle networking. Background Art

[0002] Traditional safety measures for single-lane passing mainly rely on static management methods such as road signs, markings, and sight distance requirements, or use simple vehicle detectors to obtain traffic flow parameters. However, these traditional methods have many limitations, such as incomplete information acquisition, poor real-time performance, and a single warning mechanism. In addition, most existing vehicle safety warning systems rely on sensors carried by single vehicles for obstacle detection and trajectory prediction, but in terms of perception range, data accuracy, and communication timeliness, it is difficult to meet the strict safety requirements in single-lane passing scenarios.

[0003] In recent years, with the development of edge computing technology, combining it with the Internet of Vehicles system has become a new idea to solve the safety problem of single-lane passing. However, the existing edge computing architecture still faces a series of challenges when dealing with highly dynamic traffic scenarios. First, the existing vehicle trajectory prediction algorithm has low accuracy when dealing with complex working conditions such as sudden speed changes and sharp turns, and it is difficult to meet safety requirements; second, the existing safe passing distance threshold setting is too simple and fails to fully consider the influence of multiple factors such as road geometry, vehicle model differences, and road conditions; third, the passing risk assessment method lacks systematicity and dynamic adaptability, making it difficult to achieve accurate graded warnings and personalized avoidance strategies. In addition, the delay problem in data transmission and processing in the existing system further restricts the real-time and effectiveness of warning information.

[0004] In response to the above problems, the present invention proposes a single-lane passing safety analysis method based on the Internet of Vehicles. The method collects vehicle data by deploying an Internet of Vehicles road test board, combines the roadside edge computing unit for real-time analysis and processing, and establishes an improved trajectory prediction model and dynamic risk assessment mechanism, aiming to achieve intelligent management and active warning of single-lane passing safety. Summary of the invention

[0005] In view of the problems that existing single-lane passing safety assurance measures have incomplete information acquisition, poor real-time performance and a single early warning mechanism, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to improve the safety in single-lane meeting scenarios through the combination of vehicle networking and edge computing technology, achieve accurate trajectory prediction, dynamic risk assessment and real-time warning, thereby effectively solving the limitations of traditional methods under complex working conditions.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a single-lane meeting safety analysis method based on the Internet of Vehicles, which includes collecting raw data of vehicles traveling in a single lane through an Internet of Vehicles road test board, and transmitting the raw data to a roadside edge computing unit; using the roadside edge computing unit to divide the raw data into a forward traveling vehicle data group and a reverse traveling vehicle data group according to the driving direction, and establishing a trajectory prediction model according to an improved Kalman filtering algorithm; based on the trajectory prediction model, predicting the driving trajectory and longitudinal spacing value of two vehicles traveling in opposite directions, and calculating the minimum safe meeting distance threshold of the two vehicles; judging the longitudinal spacing value and the minimum safe meeting distance threshold through the roadside edge computing unit, when the longitudinal spacing value is less than the minimum safe meeting distance threshold, calculating the meeting risk coefficient based on the driving speed, and sending the meeting risk coefficient to the Internet of Vehicles road test board; receiving the meeting risk coefficient through the Internet of Vehicles road test board and converting it into an acoustic and optical warning signal, and using the roadside edge computing unit to send meeting risk warning information to the vehicle.

[0009] As a preferred solution of the single-lane vehicle-to-vehicle network-based safety analysis method described in the present invention, wherein: the vehicle-to-vehicle risk warning information includes a deceleration amplitude and a vehicle-to-vehicle avoidance strategy; the original data includes position coordinates, driving speed and driving direction; the method for converting the sound and light warning signal is to convert the vehicle-to-vehicle risk coefficient into a graded warning signal according to a preset risk level classification rule, and trigger a sound and light warning device, wherein the sound and light warning device includes a variable information sign and an audio alarm; based on the vehicle-to-vehicle risk coefficient, the driving speed and the longitudinal spacing value, the recommended deceleration amplitude is calculated by the roadside edge calculation unit, and a vehicle-to-vehicle avoidance strategy is formulated according to the lane width and road conditions; the deceleration amplitude and the vehicle-to-vehicle avoidance strategy are packaged to form vehicle-to-vehicle risk warning information, and the vehicle-to-vehicle risk warning information is sent to the vehicle-to-vehicle terminals of the two vehicles traveling in opposite directions through the vehicle-to-vehicle network communication protocol, wherein the vehicle-to-vehicle terminal receives the vehicle-to-vehicle risk warning information and issues a warning prompt to the driver through the in-vehicle display screen and voice broadcast system.

[0010] As a preferred solution of the single-lane passing safety analysis method based on the vehicle network of the present invention, wherein: the calculation method of the passing risk coefficient is to set the dynamic evaluation cycle of the roadside edge calculation unit, and compare the longitudinal spacing value and the minimum safe passing distance threshold in real time; when it is detected that the longitudinal spacing value is less than the minimum safe passing distance threshold, the roadside edge calculation unit extracts the respective driving speeds of the two vehicles traveling in opposite directions, and establishes a passing risk calculation model; when it is detected that the longitudinal spacing value is greater than or equal to the minimum safe passing distance threshold, the roadside edge calculation unit continues to monitor the longitudinal spacing value change trend of the two vehicles; if the safe distance is maintained for a continuous preset sampling period, the passing risk monitoring state is released; the relative speed of the two vehicles and the difference between the longitudinal spacing value and the minimum safe passing distance threshold are used as inputs to the passing risk calculation model to calculate the passing risk coefficient; the passing risk coefficient is sent to the vehicle network road test board of the corresponding road section through the vehicle network communication protocol, and the vehicle network road test board performs warning signal conversion after receiving the passing risk coefficient.

[0011] As a preferred solution of the single-lane passing safety analysis method based on the Internet of Vehicles of the present invention, wherein: the calculation method of the minimum safe passing distance threshold is, based on the prediction result of the trajectory prediction model, the vehicles in the forward driving vehicle data group and the reverse driving vehicle data group are paired for the opposite-way passing by the roadside edge calculation unit; according to the result of the opposite-way passing pairing, the position coordinates and the driving speed are extracted for each pair of two vehicles traveling in opposite directions; according to the projection of the position coordinates in the same coordinate system, the straight-line distance between the center points of the two vehicles is calculated, and combined with the road direction, the straight-line distance is projected to the direction of the center line of the road to obtain the longitudinal spacing value of the two vehicles; based on the longitudinal spacing value, the vehicle model characteristics, driving speed, road curvature and road surface conditions of the two vehicles are calculated by the roadside edge calculation unit, and the specific formula is as follows:

[0012]

[0013] Among them, D safe is the minimum safe passing distance threshold, v1 is the current speed of the first vehicle, v2 is the current speed of the second vehicle, a1 is the maximum deceleration of the first vehicle, a2 is the maximum deceleration of the second vehicle, γ is the road condition coefficient, θ is the road curvature angle, R c is the road curvature radius, η(ξ) is the vehicle model characteristic function, ξ is the vehicle model characteristic parameter, k is the sensitivity coefficient, μ is the standard vehicle model parameter, and ξ0 is the vehicle model reference value.

[0014] As a preferred solution of the single-lane passing safety analysis method based on the vehicle networking according to the present invention, wherein: the method for establishing the trajectory prediction model is to classify and process the original data according to the driving direction, and divide the original data with the driving direction toward the vehicle networking road test board into a forward driving vehicle data group; at the same time, the original data with the driving direction away from the vehicle networking road test board is divided into a reverse driving vehicle data group; vehicle motion state matrices are established for the forward driving vehicle data group and the reverse driving vehicle data group respectively; an improved Kalman filter algorithm is used to perform data smoothing processing on the vehicle motion state matrix to establish a trajectory prediction model, wherein the improved Kalman filter algorithm introduces an adaptive noise covariance matrix.

[0015] As a preferred solution of the single-lane meeting safety analysis method based on the Internet of Vehicles of the present invention, the specific formula of the trajectory prediction model is as follows:

[0016]

[0017] Among them, P(t) is the predicted position value at time t, α i is the adaptive weight coefficient, β is the speed sensitivity parameter, v i is the current speed of the i-th vehicle, μ is the speed reference value, n is the number of sampling points, is the position x of the i-th vehicle i The rate of change is the instantaneous velocity, λ is the time attenuation factor, τ is the integral variable, and R(t) is the residual correction term.

[0018] As a preferred solution of the single-lane passing safety analysis method based on the Internet of Vehicles described in the present invention, wherein: the method for obtaining the original data is to scan the vehicles traveling in the single lane through the Internet of Vehicles road test board to collect vehicle data; based on the vehicle data, the position coordinates of the vehicles traveling in the single lane are obtained according to the laser ranging principle, and the driving speed of the vehicles traveling in the single lane is calculated according to the changes in the position coordinates at adjacent moments; the driving direction of the vehicles traveling in the single lane is determined by comparing the change trends of the position coordinates at adjacent moments; the position coordinates, the driving speed and the driving direction are used as original data through the Internet of Vehicles road test board, and the original data are transmitted to the roadside edge computing unit through the Internet of Vehicles communication protocol, wherein the roadside edge computing unit receives the original data transmitted by several Internet of Vehicles road test boards.

[0019] In a second aspect, an embodiment of the present invention provides a single-lane meeting safety analysis system based on the Internet of Vehicles, which includes: an acquisition module, which is used to collect raw data of vehicles traveling in a single lane through a road test board of the Internet of Vehicles, and transmit the raw data to a roadside edge computing unit; a model building module, which is used to use the roadside edge computing unit to divide the raw data into a forward driving vehicle data group and a reverse driving vehicle data group according to the driving direction, and establish a trajectory prediction model according to an improved Kalman filtering algorithm; a prediction module, based on the trajectory prediction model, predicts the driving trajectory and longitudinal spacing value of two vehicles traveling in opposite directions, and calculates the minimum safe meeting distance threshold of the two vehicles; a judgment module, which is used to judge the longitudinal spacing value and the minimum safe meeting distance threshold through the roadside edge computing unit, and when the longitudinal spacing value is less than the minimum safe meeting distance threshold, calculates the meeting risk coefficient based on the driving speed, and sends the meeting risk coefficient to the road test board of the Internet of Vehicles; a meeting risk warning module, which is used to receive the meeting risk coefficient through the road test board of the Internet of Vehicles and convert it into an acoustic and optical warning signal, and use the roadside edge computing unit to send the meeting risk warning information to the vehicle.

[0020] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the single-lane meeting safety analysis method based on the Internet of Vehicles as described in the first aspect of the present invention are implemented.

[0021] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the single-lane meeting safety analysis method based on the Internet of Vehicles as described in the first aspect of the present invention are implemented.

[0022] The beneficial effects of the present invention are as follows: vehicle data is collected through the Internet of Vehicles road test board to solve the delay and error problems existing in traditional manual monitoring and achieve real-time and accuracy of data collection; vehicle data is classified and processed and trajectory predicted through an improved Kalman filter algorithm, and an adaptive noise covariance matrix is ​​introduced to improve prediction accuracy; a dynamic minimum safe meeting distance threshold calculation model is established in combination with multi-dimensional parameters such as vehicle model characteristics, speed and road conditions to achieve a forward-looking assessment of meeting risks; a dynamic risk assessment mechanism is established by comparing the longitudinal spacing value with the safety threshold in real time to ensure the system's timely discovery of potential dangers; and a clear response plan is provided to the driver through multi-dimensional sound and light warnings and avoidance strategy guidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0024] Figure 1 This is a flow chart of a single-lane meeting safety analysis method based on the Internet of Vehicles according to Example 1. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0026] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0027] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0028] Example 1

[0029] Reference Figure 1 , which is the first embodiment of the present invention, and provides a single-lane meeting safety analysis method based on the Internet of Vehicles, including:

[0030] S1: Collect the original data of vehicles traveling in a single lane through the Internet of Vehicles road test board, and transmit the original data to the roadside edge computing unit.

[0031] Specifically, the method for obtaining the original data is to scan vehicles traveling on a single lane through the Internet of Vehicles road test board to collect vehicle data.

[0032] It should be noted that Internet of Vehicles road test boards are deployed every 50-100 meters along a single lane. The Internet of Vehicles road test boards are installed on one side of the road using poles, with an installation height of 4-6 meters above the ground. They are equipped with high-precision millimeter-wave radar sensors and high-definition visual sensors, and vehicle data is collected through dual-sensor fusion. The signal coverage range of the Internet of Vehicles road test boards is 100 meters in front and behind them, and there is an overlapping area of ​​20-30 meters in the signal coverage range of adjacent road test boards.

[0033] Furthermore, based on the vehicle data, the position coordinates of the vehicle traveling in a single lane are obtained according to the laser ranging principle, and the driving speed of the vehicle traveling in the single lane is calculated according to the changes in the position coordinates at adjacent times; the driving direction of the vehicle traveling in the single lane is determined by comparing the changing trends of the position coordinates at adjacent times.

[0034] Furthermore, the location coordinates, driving speed and driving direction are taken as raw data through the Internet of Vehicles road test board, and the raw data are transmitted in real time to the roadside edge computing unit through the Internet of Vehicles communication protocol, wherein the roadside edge computing unit receives the raw data transmitted by several Internet of Vehicles road test boards; the raw data includes the location coordinates, driving speed and driving direction.

[0035] S2: Using the roadside edge computing unit, the raw data is divided into a forward driving vehicle data group and a reverse driving vehicle data group according to the driving direction, and a trajectory prediction model is established according to an improved Kalman filter algorithm.

[0036] Specifically, the method for establishing the trajectory prediction model is to classify and process the original data according to the driving direction, and divide the original data with the driving direction toward the vehicle networking road test board into the forward driving vehicle data group; at the same time, the original data with the driving direction away from the vehicle networking road test board is divided into the reverse driving vehicle data group.

[0037] Furthermore, a vehicle motion state matrix is ​​established for the forward driving vehicle data group and the reverse driving vehicle data group respectively; an improved Kalman filter algorithm is used to perform data smoothing on the vehicle motion state matrix to establish a trajectory prediction model;

[0038] It should be noted that the row vectors of the vehicle motion state matrix represent the sequence of sampling time points, and the column vectors represent the vehicle's motion state parameters; the improved Kalman filter algorithm introduces an adaptive noise covariance matrix; the trajectory prediction model predicts the subsequent motion trajectory of the vehicle in the forward driving vehicle data group and the reverse driving vehicle data group respectively.

[0039] Furthermore, the specific formula of the trajectory prediction model is as follows:

[0040]

[0041] Among them, P(t) is the predicted position value at time t, α i is the adaptive weight coefficient, β is the speed sensitivity parameter, v i is the current speed of the i-th vehicle, μ is the speed reference value, n is the number of sampling points, is the position x of the i-th vehicle i The rate of change is the instantaneous velocity, λ is the time attenuation factor, τ is the integral variable, and R(t) is the residual correction term.

[0042] S3: Based on the trajectory prediction model, the driving trajectories and longitudinal spacing values ​​of the two vehicles traveling towards each other are predicted, and a minimum safe passing distance threshold of the two vehicles is calculated.

[0043] Specifically, the method for calculating the minimum safe meeting distance threshold is to pair the vehicles in the forward-traveling vehicle data group and the reverse-traveling vehicle data group for meeting each other through the roadside edge calculation unit based on the prediction result of the trajectory prediction model.

[0044] It should be noted that if the position coordinates of a vehicle in the forward-traveling vehicle data group are within 100 meters in front of the vehicle in the reverse-traveling vehicle data group, the vehicle will be paired for oncoming vehicles; if the minimum longitudinal spacing value of the predicted trajectories of the two vehicles in the next 2 seconds is less than 50 meters, the two vehicles will be marked as potential oncoming vehicles; if the driving speeds of the two vehicles are both greater than 30km / h, and the relative speed is greater than 50km / h, the two vehicles will be included in the priority monitoring list; if the predicted trajectories of the two vehicles show that they will meet on the same road section, the two vehicles will be determined as oncoming vehicles. Pairing relationship; if a vehicle meets the pairing conditions with multiple oncoming vehicles at the same time, the vehicle with the smallest longitudinal spacing value is preferentially selected as its oncoming vehicle pairing object; if the angle between the driving directions of the two vehicles is between 150° and 210°, it is determined that they meet the angle requirements for oncoming vehicles; if both vehicles are on a curved section of the road, and the radius of the curve is less than 200 meters, the oncoming vehicle pairing priority of the two vehicles is increased.

[0045] Furthermore, according to the result of the oncoming vehicle pairing, the position coordinates and the driving speed are extracted for each pair of two vehicles traveling in opposite directions; the position coordinates are projected in the same coordinate system to calculate the straight-line distance between the center points of the two vehicles, and this straight-line distance is projected to the direction of the road center line in combination with the road direction to obtain the longitudinal spacing value of the two vehicles.

[0046] Furthermore, based on the longitudinal spacing value, the roadside edge calculation unit calculates the minimum safe passing distance threshold of the two vehicles by taking into account the vehicle type characteristics, driving speed, road curvature and road surface conditions of the two vehicles. The specific formula is as follows:

[0047]

[0048] Among them, D safe is the minimum safe passing distance threshold, v1 is the current speed of the first vehicle, v2 is the current speed of the second vehicle, a1 is the maximum deceleration of the first vehicle, a2 is the maximum deceleration of the second vehicle, γ is the road condition coefficient, θ is the road curvature angle, R cis the road curvature radius, η(ξ) is the vehicle model characteristic function, ξ is the vehicle model characteristic parameter, k is the sensitivity coefficient, μ is the standard vehicle model parameter, and ξ0 is the vehicle model reference value.

[0049] S4: The longitudinal spacing value and the minimum safe meeting distance threshold are judged by the roadside edge calculation unit. When the longitudinal spacing value is less than the minimum safe meeting distance threshold, a meeting risk coefficient is calculated based on the driving speed, and the meeting risk coefficient is sent to the Internet of Vehicles road test board.

[0050] Specifically, the method for calculating the meeting risk coefficient is to set a dynamic evaluation cycle of the roadside edge calculation unit and compare the longitudinal spacing value with the minimum safe meeting distance threshold in real time.

[0051] Furthermore, when it is detected that the longitudinal spacing value is less than the minimum safe meeting distance threshold, the roadside edge calculation unit extracts the respective driving speeds of the two vehicles traveling in opposite directions and establishes a meeting risk calculation model; when it is detected that the longitudinal spacing value is greater than or equal to the minimum safe meeting distance threshold, the roadside edge calculation unit continues to monitor the change trend of the longitudinal spacing value of the two vehicles; if the safe distance is maintained for a continuous preset sampling period, the meeting risk monitoring state is released.

[0052] Furthermore, if the longitudinal spacing value is less than the minimum safe meeting distance threshold for the first time, the corresponding vehicle pair will be marked as a warning state; if the longitudinal spacing value is less than 80% of the minimum safe meeting distance threshold, the oncoming risk factor will be increased by 0.2; if the longitudinal spacing value is less than the minimum safe meeting distance threshold for three consecutive sampling cycles, the continuous monitoring mode will be started; if the longitudinal spacing value is less than the minimum safe meeting distance threshold and the relative speed of the two vehicles is greater than 60km / h, the oncoming risk factor will be increased by 0.3; if the longitudinal spacing value suddenly decreases rapidly and the rate of change exceeds 2m / s, the emergency warning mechanism will be immediately triggered; if the longitudinal spacing value fluctuates within the range of 80%-120% of the minimum safe meeting distance threshold, this vehicle pair will be included in the key monitoring list.

[0053] It should be noted that the oncoming risk calculation model is a comprehensive assessment model based on the motion state and environmental parameters of two vehicles traveling in opposite directions. This model uses the relative speed of the two vehicles, the ratio of the longitudinal spacing value to the safety threshold, and the acceleration change trend as the main assessment factors, while introducing environmental factors such as road curvature and road friction coefficient and vehicle model characteristic parameters as correction factors. After each assessment factor is normalized, the final risk coefficient is obtained through weighted calculation.

[0054] Specifically, the relative speed of the two vehicles, the difference between the longitudinal spacing value and the minimum safe meeting distance threshold are used as inputs to the meeting risk calculation model to calculate the meeting risk coefficient; the meeting risk coefficient is sent to the Internet of Vehicles road test board of the corresponding road section through the Internet of Vehicles communication protocol, and the Internet of Vehicles road test board converts the early warning signal after receiving the meeting risk coefficient.

[0055] S5: Receive the oncoming vehicle risk coefficient through the Internet of Vehicles road test board and convert it into an acoustic and visual warning signal, and use the roadside edge computing unit to send oncoming vehicle risk warning information to the vehicle.

[0056] Specifically, the risk coefficient of the vehicle approaching is converted into a graded warning signal according to the preset risk level classification rules, and the sound and light warning device is triggered, wherein the sound and light warning device includes a variable information sign and an audio alarm; the variable information sign displays warning prompts of different colors and frequencies; the audio alarm emits warning sounds of different tones and rhythms.

[0057] It should be noted that when the oncoming risk factor is less than 0.3, the road test board displays a yellow flashing signal with a flashing frequency of 60 times / minute; when the oncoming risk factor is between 0.3-0.7, it displays an orange fast flashing signal with a flashing frequency of 120 times / minute; when the oncoming risk factor is greater than 0.7, it displays a red high-frequency flashing signal with a flashing frequency of 180 times / minute. The display screen of the sound and light warning signal adopts a high-brightness LED display module, with a visible distance of not less than 200 meters during the day and not less than 300 meters at night.

[0058] Furthermore, based on the oncoming risk factor, driving speed and longitudinal spacing value, the recommended deceleration amplitude is calculated by the roadside edge calculation unit, and an oncoming avoidance strategy is formulated according to the lane width and road conditions.

[0059] Furthermore, the deceleration range and oncoming vehicle avoidance strategy are packaged into oncoming vehicle risk warning information, which is sent to the onboard terminals of the two vehicles traveling in the opposite direction through the Internet of Vehicles communication protocol.

[0060] It should be noted that the oncoming risk warning information includes the deceleration range and oncoming avoidance strategy; after receiving the oncoming risk warning information, the on-board terminal will issue a warning prompt to the driver through the in-vehicle display screen and voice broadcast system.

[0061] Furthermore, the present embodiment also provides a single-lane meeting safety analysis system based on the Internet of Vehicles, including: a collection module, which is used to collect raw data of vehicles traveling in a single lane through a road test board of the Internet of Vehicles, and transmit the raw data to a roadside edge computing unit; a model building module, which is used to use the roadside edge computing unit to divide the raw data into a forward-traveling vehicle data group and a reverse-traveling vehicle data group according to the driving direction, and establish a trajectory prediction model according to an improved Kalman filter algorithm; a prediction module, which predicts the driving trajectory and longitudinal spacing value of two vehicles traveling in opposite directions based on the trajectory prediction model, and calculates the minimum safe meeting distance threshold of the two vehicles; a judgment module, which is used to judge the longitudinal spacing value and the minimum safe meeting distance threshold through the road side edge computing unit, and when the longitudinal spacing value is less than the minimum safe meeting distance threshold, calculate the meeting risk coefficient based on the driving speed, and send the meeting risk coefficient to the road test board of the Internet of Vehicles; a meeting risk warning module, which is used to receive the meeting risk coefficient through the road test board of the Internet of Vehicles and convert it into an acoustic and optical warning signal, and use the road side edge computing unit to send the meeting risk warning information to the vehicle.

[0062] This embodiment also provides a computer device, which is applicable to the single-lane passing safety analysis method based on the Internet of Vehicles, and includes a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the single-lane passing safety analysis method based on the Internet of Vehicles as proposed in the above embodiment.

[0063] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0064] The present embodiment also provides a storage medium having a computer program stored thereon, which implements the following steps when executed by a processor: collecting raw data of vehicles traveling in a single lane through a vehicle networking road test board, and transmitting the raw data to a roadside edge computing unit; using the roadside edge computing unit to divide the raw data into a forward traveling vehicle data group and a reverse traveling vehicle data group according to the traveling direction, and establishing a trajectory prediction model according to an improved Kalman filter algorithm; based on the trajectory prediction model, predicting the traveling trajectories and longitudinal spacing values ​​of two vehicles traveling in opposite directions, and calculating a minimum safe passing distance threshold of the two vehicles; judging the longitudinal spacing value and the minimum safe passing distance threshold through the roadside edge computing unit, and when the longitudinal spacing value is less than the minimum safe passing distance threshold, calculating a passing risk coefficient based on the traveling speed, and sending the passing risk coefficient to the vehicle networking road test board; receiving the passing risk coefficient through the vehicle networking road test board and converting it into an acoustic and optical warning signal, and using the roadside edge computing unit to send a passing risk warning information to the vehicle.

[0065] In summary, the present invention collects vehicle data through the Internet of Vehicles road test board, solves the delay and error problems existing in traditional manual monitoring, and realizes the real-time and accuracy of data collection; classifies and processes vehicle data and predicts trajectories through an improved Kalman filtering algorithm, and introduces an adaptive noise covariance matrix to improve prediction accuracy; combines multi-dimensional parameters such as vehicle model characteristics, speed and road conditions to establish a dynamic minimum safe meeting distance threshold calculation model to achieve a forward-looking assessment of meeting risks; establishes a dynamic risk assessment mechanism by comparing the longitudinal spacing value with the safety threshold in real time to ensure the system's timely detection of potential dangers; and provides drivers with clear response plans through multi-dimensional sound and light warnings and avoidance strategy guidance.

[0066] Example 2

[0067] Referring to Table 1, which is a second embodiment of the present invention, this embodiment provides a single-lane meeting safety analysis method based on the Internet of Vehicles. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0068] Specifically, the section from K45+000 to K48+000 of a provincial highway S203 was tested. This section is a typical two-way single lane in a mountainous area, with a length of 3 kilometers, a road width of 6.5 meters, a design speed of 40 kilometers per hour, and many continuous S-shaped curves along the line, with a minimum circular curve radius of 125 meters. The test was carried out under sunny and rainy conditions for a total of 7 days. A vehicle network road test board was installed every 75 meters on the test section, with a total of 40 installed, and the installation height was uniformly 4.8 meters. Each road test board is equipped with a 77GHz millimeter wave radar and an 8-megapixel high-definition camera, and a dual-sensor data fusion algorithm is used to achieve precise vehicle positioning, with a positioning accuracy better than ±0.1 meters; the road test board communicates with the roadside edge computing unit in real time through the 5G network, and the data transmission delay is less than 10 milliseconds; the test vehicles include 10 small buses, medium-sized buses and large trucks, and all vehicles are equipped with on-board terminal equipment.

[0069] Furthermore, the vehicle trajectory is predicted through an improved Kalman filter algorithm, with the prediction time window set to 3 seconds and the sampling frequency set to 20Hz. The system adopts an adaptive dynamic evaluation mechanism. When a potential oncoming risk is detected, the evaluation cycle is automatically shortened from 1 second to 0.2 seconds. At curves, the system considers road curvature and sight distance factors and appropriately increases the safety threshold. For different weather conditions, the system automatically adjusts the road condition coefficient, with a value of 0.2 on sunny days and 0.4 on rainy days. When an oncoming risk is detected, the system issues a warning through an audio-visual warning device and pushes personalized avoidance suggestions to the vehicle terminal.

[0070] Furthermore, as shown in Table 1, in terms of prediction accuracy, the system's trajectory prediction error for different models under sunny conditions is controlled within 0.21 meters, which is 58% higher than the 0.5-meter error of the traditional prediction method. In particular, the prediction accuracy for small passenger cars reaches 0.15 meters, thanks to the adaptive noise covariance matrix introduced by the improved Kalman filter algorithm; the system response delay is generally maintained within 15 milliseconds, and the response delay for large trucks is only 15 milliseconds even under rainy conditions, which is much better than the industry average of 30 milliseconds. In terms of risk recognition rate, the system's recognition rate under sunny conditions is as high as 95.8%, and the recognition rate of large trucks under rainy conditions is maintained at more than 90.6%, which is significantly higher than the average level of 80% of the existing system; at the same time, the system's false alarm rate is effectively controlled, only 2.1%-2.5% under sunny conditions, and slightly improved under rainy conditions but still controlled within 3.8%, which shows that the system has high reliability.

[0071] Table 1 Experimental data table

[0072] Test object ID Vehicle Type Weather conditions Prediction accuracy Response Delay Risk identification rate False Positive Rate Safety improvement rate Vehicle-001 Small passenger car sunny 0.15m 8ms 95.8％ 2.1％ 89.5％ Vehicle-002 Medium Bus sunny 0.18m 9ms 94.6％ 2.3％ 88.7％ Vehicle-003 Large truck sunny 0.21m 11ms 93.9％ 2.5％ 87.9％ Vehicle-004 Small passenger car rain 0.19m 12ms 92.5％ 3.2％ 86.4％ Vehicle-005 Medium Bus rain 0.22m 13ms 91.8％ 3.5％ 85.8％ Vehicle-006 Large truck rain 0.25m 15ms 90.6％ 3.8％ 84.6％

[0073] Specifically, the safety improvement rate of different models under different weather conditions reached more than 84%, among which the safety improvement rate of small passenger cars under sunny conditions was as high as 89.5%, which fully demonstrated the significant effect of the system in practical application. The data also showed that the system performance would fluctuate with the size of the model and weather conditions, but the fluctuation range always remained within an acceptable range, reflecting the stability and adaptability of the system.

[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A single lane meeting safety analysis method based on vehicle networking, characterized by: include, The original data of vehicles traveling in a single lane is collected through the Internet of Vehicles road test board, and the original data is transmitted to the roadside edge computing unit; Using the roadside edge computing unit, the raw data is divided into a forward-traveling vehicle data group and a reverse-traveling vehicle data group according to the driving direction, and a trajectory prediction model is established according to an improved Kalman filter algorithm; Based on the trajectory prediction model, the driving trajectories and longitudinal spacing values ​​of two vehicles traveling towards each other are predicted, and a minimum safe passing distance threshold of the two vehicles is calculated; The longitudinal spacing value and the minimum safe meeting distance threshold are judged by the roadside edge calculation unit, and when the longitudinal spacing value is less than the minimum safe meeting distance threshold, a meeting risk coefficient is calculated based on the driving speed, and the meeting risk coefficient is sent to the vehicle networking road test board; The oncoming vehicle risk coefficient is received by the Internet of Vehicles road test board and converted into an acoustic and optical warning signal, and the oncoming vehicle risk warning information is sent to the vehicle using a roadside edge computing unit.

2. The single-lane meeting safety analysis method based on the Internet of Vehicles according to claim 1, characterized in that: The oncoming risk warning information includes the deceleration range and oncoming avoidance strategy; the original data includes the position coordinates, driving speed and driving direction; the conversion method of the sound and light warning signal is: The risk factor of the oncoming vehicle is converted into a graded warning signal according to the preset risk level classification rules, and the sound and light warning device is triggered, wherein the sound and light warning device includes a variable information sign and an audio alarm; Based on the oncoming risk coefficient, the driving speed and the longitudinal spacing value, the roadside edge calculation unit calculates a recommended deceleration amplitude, and formulates an oncoming avoidance strategy according to the lane width and road surface conditions; The deceleration amplitude and the oncoming vehicle avoidance strategy are packaged to form oncoming vehicle risk warning information, and the oncoming vehicle risk warning information is sent to the vehicle terminals of two vehicles traveling in opposite directions through the vehicle networking communication protocol. After receiving the oncoming vehicle risk warning information, the vehicle terminals issue a warning prompt to the driver through the in-vehicle display screen and voice broadcast system.

3. The single-lane meeting safety analysis method based on the Internet of Vehicles as claimed in claim 2, characterized in that: The calculation method of the meeting risk coefficient is: Setting a dynamic evaluation cycle of the roadside edge computing unit and comparing the longitudinal spacing value with the minimum safe passing distance threshold in real time; When it is detected that the longitudinal spacing value is less than the minimum safe passing distance threshold, the roadside edge calculation unit extracts the respective driving speeds of the two vehicles traveling in opposite directions, and establishes a passing risk calculation model; when it is detected that the longitudinal spacing value is greater than or equal to the minimum safe passing distance threshold, the roadside edge calculation unit continues to monitor the change trend of the longitudinal spacing value of the two vehicles; If the safe distance is maintained for consecutive preset sampling periods, the oncoming vehicle risk monitoring state is released; The relative speed of the two vehicles, the difference between the longitudinal spacing value and the minimum safe passing distance threshold are used as inputs to the passing risk calculation model to calculate the passing risk coefficient; The oncoming vehicle risk coefficient is sent to the oncoming vehicle road test board of the corresponding road section through the oncoming vehicle communication protocol, wherein the oncoming vehicle road test board converts the oncoming vehicle risk coefficient into a warning signal after receiving the oncoming vehicle risk coefficient.

4. The single-lane meeting safety analysis method based on the Internet of Vehicles as claimed in claim 3 is characterized by: The calculation method of the minimum safe passing distance threshold is: Based on the prediction result of the trajectory prediction model, the roadside edge computing unit performs on-coming vehicle pairing for the vehicles in the forward-traveling vehicle data group and the reverse-traveling vehicle data group; According to the result of the pairing of the oncoming vehicles, the position coordinates and the driving speed are extracted for each pair of two vehicles traveling towards each other; Projecting the position coordinates in the same coordinate system, calculating the straight-line distance between the center points of the two vehicles, and projecting the straight-line distance to the center line direction of the road in combination with the road direction, to obtain the longitudinal spacing value between the two vehicles; Based on the longitudinal spacing value, the roadside edge calculation unit calculates the minimum safe passing distance threshold of the two vehicles by taking into account the vehicle type characteristics, driving speed, road curvature and road surface conditions of the two vehicles. The specific formula is as follows: Among them, D safe is the minimum safe passing distance threshold, v1 is the current speed of the first vehicle, v2 is the current speed of the second vehicle, a1 is the maximum deceleration of the first vehicle, a2 is the maximum deceleration of the second vehicle, γ is the road condition coefficient, θ is the road curvature angle, R c is the road curvature radius, η(ξ) is the vehicle model characteristic function, ξ is the vehicle model characteristic parameter, k is the sensitivity coefficient, μ is the standard vehicle model parameter, and ξ0 is the vehicle model reference value.

5. The single lane meeting safety analysis method based on the Internet of Vehicles as claimed in claim 4, characterized in that: The method for establishing the trajectory prediction model is as follows: Classify and process the raw data according to the driving direction, and classify the raw data with the driving direction toward the IoV road test board into a forward driving vehicle data group; At the same time, the original data of the vehicle network road test board whose driving direction is opposite to the vehicle network road test board is divided into a reverse driving vehicle data group; Establishing a vehicle motion state matrix for the forward-traveling vehicle data set and the reverse-traveling vehicle data set respectively; An improved Kalman filter algorithm is used to perform data smoothing processing on the vehicle motion state matrix to establish a trajectory prediction model, wherein the improved Kalman filter algorithm introduces an adaptive noise covariance matrix.

6. The single-lane meeting safety analysis method based on the Internet of Vehicles as claimed in claim 5, characterized in that: The specific formula of the trajectory prediction model is as follows: Among them, P(t) is the predicted position value at time t, α i is the adaptive weight coefficient, β is the speed sensitivity parameter, v i is the current speed of the i-th vehicle, μ is the speed reference value, n is the number of sampling points, is the position x of the i-th vehicle i The rate of change is the instantaneous velocity, λ is the time attenuation factor, τ is the integral variable, and R(t) is the residual correction term.

7. The single-lane meeting safety analysis method based on the Internet of Vehicles according to claim 5 or 6, characterized in that: The method for obtaining the original data is: Scan vehicles traveling in a single lane through the Internet of Vehicles road test board to collect vehicle data; Based on the vehicle data, the position coordinates of the vehicle traveling in the single lane are obtained according to the principle of laser ranging, and the travel speed of the vehicle traveling in the single lane is calculated according to the change of the position coordinates at adjacent moments; Determining the driving direction of the single-lane vehicle by comparing the change trends of the position coordinates at adjacent moments; The position coordinates, the driving speed and the driving direction are taken as raw data by the Internet of Vehicles road test board, and the raw data are transmitted to a roadside edge computing unit through the Internet of Vehicles communication protocol, wherein the roadside edge computing unit receives the raw data transmitted by several Internet of Vehicles road test boards.

8. A single-lane vehicle-to-vehicle networking-based safety analysis system, based on the single-lane vehicle-to-vehicle networking-based safety analysis method according to any one of claims 1 to 7, characterized in that: include, A collection module, used to collect raw data of vehicles traveling in a single lane through a vehicle networking road test board, and transmit the raw data to a roadside edge computing unit; A model building module, used to use the roadside edge computing unit to divide the raw data into a forward driving vehicle data group and a reverse driving vehicle data group according to the driving direction, and to establish a trajectory prediction model according to an improved Kalman filter algorithm; A prediction module, based on the trajectory prediction model, predicts the driving trajectories and longitudinal spacing values ​​of two vehicles traveling in opposite directions, and calculates a minimum safe passing distance threshold of the two vehicles; A judgment module, configured to judge the longitudinal spacing value and the minimum safe meeting distance threshold through the roadside edge calculation unit, and when the longitudinal spacing value is less than the minimum safe meeting distance threshold, calculate the meeting risk coefficient based on the driving speed, and send the meeting risk coefficient to the Internet of Vehicles road test board; The oncoming vehicle risk warning module is used to receive the oncoming vehicle risk coefficient through the Internet of Vehicles road test board and convert it into an acoustic and optical warning signal, and use the roadside edge computing unit to send the oncoming vehicle risk warning information to the vehicle.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the single-lane meeting safety analysis method based on the Internet of Vehicles according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the single-lane meeting safety analysis method based on the Internet of Vehicles according to any one of claims 1 to 7 are implemented.

Images