A Highway Traffic Diversion Alert Broadcasting System and Method Based on AI Recognition

Abstract

This application discloses a highway traffic diversion alarm broadcasting system and method based on AI recognition, relating to the field of highway diversion technology. This application analyzes the load of each vehicle and the driving experience, historical highway violations, and historical highway accidents of each driver to divert vehicles, diverting high-risk vehicles to safer highways, thereby ensuring vehicle safety. Different safe following distances and speeds are set for different highways to avoid traffic accidents caused by excessively close following distances or excessive speeds. Furthermore, the system monitors all vehicles traveling on each highway, broadcasting warnings when accidents or violations are detected, which enhances deterrence against drivers and prevents repeated violations from leading to traffic accidents.

Description

Technical Field

[0001] This application relates to the field of highway traffic diversion technology, specifically to a highway traffic diversion alarm broadcasting system and method based on AI recognition. Background Technology

[0002] With the continuous development of society and economy, the number of private cars is increasing, and traffic accidents are also emerging one after another. Using reasonable methods to divert vehicles on highways is conducive to improving travel efficiency and avoiding risks, so that highways can play their maximum role under the premise of safety. Therefore, this application proposes a highway traffic diversion alarm broadcasting system and method based on AI recognition.

[0003] Existing technology, such as the invention application patent with announcement number CN114333362B, discloses a highway traffic flow detection device and diversion management method. The highway traffic flow detection device includes multiple vehicle detection stations arranged along the highway; a designated location after the starting point of each road segment is designated as a detection node, and a vehicle detection station is set up at each detection node; the vehicle detection station includes a gantry bracket and a camera, with both the camera and the vehicle speed detector pointing vertically downwards, and are used to count the number of vehicles in each lane of the road segment and collect the speed of vehicles passing through the detection node, respectively. The number of vehicles is counted by the camera, and the gantry bracket does not need to damage the road surface during construction; it has three traffic flow management methods, namely diversion method, flow restriction method, and speed restriction method, which can match the vehicle speed and traffic flow control according to actual needs, thereby ensuring the balance of highway transportation pressure and reducing the accident rate.

[0004] Existing technology, such as the invention application patent with announcement number CN109064754B, discloses a method for diverting traffic at highway entrances and coordinating traffic flow control. This method includes the following steps: based on highway holiday traffic demand distribution information, it obtains OD (Original Demand) information for highway entrances and exits at different times of the day through multi-source data analysis technology. Then, it calculates the traffic demand within adjacent road segments at each highway entrance and exit at different times. Finally, it adjusts and controls the flow rate at each entrance using a core control strategy, thereby avoiding traffic congestion on sections with insufficient capacity, preventing reduced capacity and traffic accident risks caused by congestion, and ultimately achieving smooth traffic flow on all highway segments to maximize the efficiency of the highway mainline. This technology has a good effect on congestion prevention and flow control in the event of large-scale traffic congestion on highways during holidays, and has a wide range of applications and good prospects in my country's intelligent traffic management and control of highways.

[0005] The above-mentioned solutions have the following technical problems: 1. Current technology mainly analyzes the number, speed and OD information of vehicles on highways to divert traffic. However, current technology does not take into account the impact of vehicle load and driver's driving skills on vehicle operation. When vehicles are overloaded or drivers have poor driving skills, traffic accidents may occur, which may lead to traffic congestion on highways.

[0006] 2. Current technology does not monitor or broadcast information about the driving process of each vehicle on the highway. When a vehicle violates regulations or an accident occurs, it should be broadcast to alert not only the violating vehicle but also other vehicles nearby to take timely action. The current technology's neglect of this aspect results in a lack of completeness and comprehensiveness in the diversion of traffic on highways. Summary of the Invention

[0007] The purpose of this application is to provide a highway traffic diversion warning broadcast system and method based on AI recognition, which solves the problems existing in the background technology.

[0008] To solve the above-mentioned technical problems, this application adopts the following technical solution: In the first aspect, this application provides a highway traffic diversion alarm broadcast system based on AI recognition, including: a diversion module including an entrance diversion unit and a road center diversion unit.

[0009] The highway entrance diversion unit is used to collect vehicle information and driving information of each vehicle through AI recognition technology, and then analyzes and obtains the safety behavior characteristic value of each vehicle. At the same time, it obtains the road information of each highway from the data center, and then analyzes and obtains the road behavior characteristic value of each highway. Based on the safety behavior characteristic value of each vehicle and the road behavior characteristic value of each highway, the unit diverts each vehicle and sets different vehicle distance thresholds and speed thresholds according to the road characteristic values ​​of each highway.

[0010] The road-center diversion unit is used to monitor all vehicles traveling on each highway and to divert vehicles when an accident is detected.

[0011] Broadcast alarm module: used to broadcast alerts when accidents or violations are detected in vehicles.

[0012] In its second aspect, this application provides a method for broadcasting traffic diversion alerts on highways based on AI recognition, comprising: Step 1, highway entrance diversion: collecting vehicle information and driving information of each vehicle through AI recognition technology, and then analyzing the safety behavior characteristic values ​​of each vehicle; simultaneously obtaining road information of each highway from the data center, and then analyzing the road behavior characteristic values ​​of each highway; thereby diverting each vehicle according to the safety behavior characteristic values ​​of each vehicle and the road behavior characteristic values ​​of each highway, and setting different vehicle distance thresholds and speed thresholds according to the road characteristic values ​​of each highway.

[0013] Step 2, Traffic Diversion: Monitor all vehicles traveling on each highway and divert traffic when an accident is detected.

[0014] Step 3: Broadcast Alarm: When an accident or violation is detected in any vehicle, a broadcast will be made to alert the driver.

[0015] The beneficial effects of this application are as follows: 1. This application provides a highway traffic diversion alarm broadcasting system and method based on AI recognition. By analyzing the load of each vehicle and the driving experience, historical highway violations, and historical highway accidents of each driver, the system diverts vehicles, diverting high-risk vehicles to safer highways, thereby ensuring vehicle driving safety. At the same time, different safe following distances and safe speeds are set for different highways to avoid traffic accidents caused by excessively close following distances or excessive speeds. Furthermore, the system monitors all vehicles traveling on each highway, and when an accident or violation is detected, it broadcasts the warning, which helps to increase the deterrent effect on drivers and prevent repeated violations from causing traffic accidents.

[0016] 2. This application analyzes the load of each vehicle, the driving experience of each driver, the number of highway accidents, and the number of highway violations. Based on this analysis, vehicles are diverted to different highways, and different safe following distances and safe speeds are set. This allows high-risk vehicles to be diverted to smoother and easier-to-drive highways, while setting lower speeds and larger following distances for high-risk vehicles. This helps avoid traffic accidents caused by inexperienced drivers, improves the efficiency and safety of highway traffic, and also helps reduce the traffic accident rate.

[0017] 3. This application monitors all vehicles traveling on highways. When a traffic accident or violation occurs, it broadcasts the incident and diverts the vehicles accordingly. This can quickly alleviate congestion, reduce traffic jams, and help ensure the efficiency of highway traffic. Notifying violating vehicles also helps to increase the deterrent effect on drivers and avoid traffic accidents caused by repeated violations. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the system structure connection of this application.

[0020] Figure 2 This is a flowchart illustrating the steps involved in implementing the method described in this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] Reference Figure 1 As shown, in a first aspect, this application provides a highway traffic diversion warning broadcast system based on AI recognition, including the following modules: diversion module: including an entrance diversion unit and a roadside diversion unit.

[0023] The highway entrance diversion unit is used to collect vehicle information and driving information of each vehicle through AI recognition technology, and then analyzes and obtains the safety behavior characteristic value of each vehicle. At the same time, it obtains the road information of each highway from the data center, and then analyzes and obtains the road behavior characteristic value of each highway. Based on the safety behavior characteristic value of each vehicle and the road behavior characteristic value of each highway, the unit diverts each vehicle and sets different vehicle distance thresholds and speed thresholds according to the road characteristic values ​​of each highway.

[0024] In a specific example, the process of collecting vehicle information and driver information using AI recognition technology is as follows: A sensor weighing system and AI camera equipment are installed at the toll station entrance. The sensor weighing system obtains the overall weight of each vehicle, and the AI ​​camera equipment captures images of each vehicle and its cab. The images of each vehicle are then used to identify its model, and its tare weight and standard load capacity are obtained based on the model. Simultaneously, facial recognition is performed on each driver based on the images of their cab, and the facial images of each driver are matched with the facial images of each driver stored in the data center. Once a match is successful, the driver's driving information is obtained.

[0025] The vehicle information for each vehicle includes the vehicle's overall weight, tare weight, and standard load capacity; the driving information for each vehicle includes driving experience, number of historical highway accidents, and number of historical highway violations.

[0026] It should be noted that when a driver's facial image fails to match any of the facial images of drivers stored in the data center, the driver's driving experience is set to the minimum value, while the number of historical highway accidents and the number of historical highway violations are set to the maximum value.

[0027] It should be noted that highway accidents include collisions and scrapes, while highway violations include speeding, not wearing seat belts, and improper following distance.

[0028] In a specific example, the analysis yields the safety behavior characteristic values ​​of each vehicle. The specific analysis process is as follows: Based on the vehicle information of each vehicle, the overall weight, tare weight, and standard load of each vehicle are obtained, and then the vehicle safety assessment coefficient aw of each vehicle is obtained through analysis. i , where i is the vehicle number and i is a positive integer;

[0029] Based on the driving information of each vehicle, the driving experience, number of historical highway accidents, and number of historical highway violations of each driver are obtained, and then the driving safety assessment coefficient bw of each vehicle is obtained through analysis. i .

[0030] Substitute the vehicle safety assessment coefficient and driving safety assessment coefficient of each vehicle into the vehicle behavior model, and then, according to the vehicle behavior model expression: Output the safety behavior feature value α of the i-th vehicle. i , where w' and w'' are the upper and lower limits of the set vehicle safety behavior evaluation coefficient, respectively.

[0031] It should be noted that the upper and lower limits of the vehicle safety behavior assessment coefficient are set by the relevant staff. For example, the safety behavior assessment coefficients of each vehicle passing through the toll station within any time period can be distributed normally. The value at one-third of the normal distribution is recorded as the lower limit of the vehicle safety behavior assessment coefficient, and the value at two-thirds of the normal distribution is recorded as the upper limit of the vehicle safety behavior assessment coefficient. No specific restrictions are imposed here.

[0032] In a specific example, the analysis yields the vehicle safety assessment system and driving safety assessment coefficient for each vehicle. The specific analysis process is as follows:

[0033] Let Q be the overall weight, tare weight, and standard load capacity of each vehicle. i T i P i According to the calculation formula: The vehicle safety assessment coefficient aw for the i-th vehicle is calculated. i , where χ represents the correction factor for the set vehicle load.

[0034] It should be noted that the correction factor for the vehicle load is determined by the sensor weighing system used, and can be obtained by consulting the manufacturer's manual of the sensor weighing system.

[0035] After normalizing the driving experience, number of historical highway accidents, and number of historical highway violations of the drivers corresponding to each vehicle, they are respectively denoted as S. i K i G i According to the calculation formula: The driving safety assessment coefficient bw for the i-th vehicle is calculated. i , where ω1 and ω2 represent the weighting factors corresponding to the number of high-speed accidents and the number of high-speed violations of the driver, respectively, and θ1 and θ2 represent the weighting factors corresponding to the driver's driving experience and the driving skill evaluation coefficient, respectively.

[0036] It should be noted that the weighting factors corresponding to the number of high-speed accidents and the number of high-speed violations of the driver are obtained through the Analytic Hierarchy Process (AHP). The weighting factors corresponding to the number of high-speed accidents and the number of high-speed violations of the driver are obtained through steps such as constructing a hierarchical structure, constructing a judgment matrix, consistency testing, and weight calculation. The AHP is existing technology and will not be described in detail here.

[0037] The analysis methods for the weighting factors corresponding to the driver's driving experience and driving skill evaluation coefficient are the same as those for the weighting factors corresponding to the number of high-speed accidents and the number of high-speed violations.

[0038] In a specific example, the road information of each expressway is obtained from the data center, and then the road behavior characteristic values ​​of each expressway are analyzed. The specific analysis process is as follows: the road information of each expressway includes the number of historical accidents, the number of curves, and the angle of each curve.

[0039] After normalizing the historical accident counts and curve angles of each expressway, they are denoted as F. j and N jl Where j is the number of each expressway, j is a positive integer, and l is the number of each curve, l is a positive integer. The normalized historical accident counts, number of curves, and angles of each curve for each expressway are substituted into the highway behavior model, and expressed by the highway behavior model expression: Output the road behavior feature value β of the j-th highway. j Where L is the total number of curves, ρ1 and ρ2 represent the weighting factors corresponding to the number of historical accidents and the curve angle of the highway, respectively, F' and N' are the standard values ​​of the number of accidents and the standard values ​​of the curve angle of the highway, respectively. The minimum number of accidents for each highway is recorded as the standard value of the number of accidents, and the minimum curve angle for each highway is recorded as the standard value of the curve angle. e' and e'' represent the upper limit and lower limit of the road safety assessment coefficient of the highway, respectively.

[0040] It should be noted that the weighting factors for the number of historical accidents on highways and the weighting factors for the curve angles are set in the same way as the weighting factors for the number of highway accidents and the number of highway violations committed by drivers; the upper and lower limits of the road safety assessment coefficient for highways are set in the same way as the upper and lower limits of the vehicle safety behavior assessment coefficient, so they will not be repeated here.

[0041] In a specific example, the process of diverting vehicles based on their safety behavior characteristic values ​​and the road behavior characteristic values ​​of each highway is as follows: vehicles with a safety behavior characteristic value of 11 are diverted to highways with a road behavior characteristic value of 00; vehicles with a safety behavior characteristic value of 10 are diverted to highways with a road behavior characteristic value of 10; and vehicles with a safety behavior characteristic value of 00 are diverted to highways with a road behavior characteristic value of 11, thereby achieving the diversion of vehicles.

[0042] Meanwhile, vehicles with safety behavior characteristic values ​​of 11, 10, and 00 are respectively classified as Class I vehicles, Class II vehicles, and Class III vehicles; vehicles with road behavior characteristic values ​​of 11, 10, and 00 are respectively classified as Class I expressways, Class II expressways, and Class III expressways.

[0043] In a specific example, the process of setting different vehicle distance thresholds and vehicle speed thresholds based on the road characteristic values ​​of each expressway is as follows: For each type of expressway, a type I vehicle distance, a type I minimum vehicle speed, and a type I maximum vehicle speed are set to control each type of vehicle.

[0044] For each Class II expressway, Class II vehicle spacing, minimum Class II vehicle speed, and maximum Class II vehicle speed are set to control each Class II vehicle.

[0045] For each of the three types of safe roads, the three types of vehicle distance, the three types of minimum vehicle speed, and the three types of maximum vehicle speed are set to control the vehicles of each of the three types.

[0046] Among them, the distance between vehicles in category 3 is less than that between vehicles in category 2 and vehicles in category 1; the maximum speed of vehicles in category 3 is greater than that of vehicles in category 2 and greater than that of vehicles in category 1; and the minimum speed of vehicles in category 3 is greater than that of vehicles in category 2 and greater than that of vehicles in category 3.

[0047] It should be noted that the various vehicle distances, maximum and minimum vehicle speeds are all set by relevant departments based on the lane conditions of each highway. For example, when all highways are two-lane, the following can be set: Class I vehicle distance is 60 meters, Class II vehicle distance is 50 meters, Class III vehicle distance is 30 meters, the maximum and minimum speeds for Class I vehicles are 120 km / h and 90 km / h, the maximum and minimum speeds for Class II vehicles are 100 km / h and 80 km / h, and the maximum and minimum speeds for Class III vehicles are 90 km / h and 80 km / h, etc.

[0048] The road-center diversion unit is used to monitor all vehicles traveling on each highway and to divert vehicles when an accident is detected.

[0049] In a specific example, when an accident is detected, the process of diverting vehicles is as follows: the location of the accident is recorded as the target accident point, the nearest diversion intersection to the target accident point is obtained and recorded as the target diversion intersection, the accident images and videos of the current highway are uploaded to the cloud data terminal, the accident level is output through the accident recognition model, and the number of vehicles to be diverted is determined according to each accident level.

[0050] It should be noted that the accident recognition model is constructed by using an AI model to identify videos and images of traffic accidents, thereby determining the severity of the traffic accident. The construction of the AI ​​model to identify videos and images of traffic accidents is existing technology, so it will not be described in detail here.

[0051] The number of vehicles to be diverted is determined based on the accident level. The specific number of vehicles to be diverted is set by the relevant staff. For example, when the accident level is Level 1, 40% of the vehicles on the highway will be diverted to other highways. When the accident level is Level 2, 30% of the vehicles on the highway will be diverted to other highways. This can help avoid traffic congestion on highways caused by traffic accidents.

[0052] Next, obtain the real-time number of vehicles on the highway associated with the target diversion intersection, and then determine whether the highway associated with the target diversion intersection is saturated. If the vehicles are saturated, then the second closest diversion intersection to the target accident point is recorded as the target diversion intersection, and it is determined whether the highway associated with the target diversion intersection is saturated, until the diversion is completed.

[0053] It should be noted that the specific process for determining whether the highway associated with the target diversion intersection is saturated is as follows: The real-time number of vehicles on the highway associated with the target diversion intersection is compared with a set vehicle capacity threshold. If the real-time number of vehicles on the highway associated with the target diversion intersection is less than 80% of the set vehicle capacity threshold, it is determined that the vehicle is not saturated; otherwise, it is determined that the vehicle is saturated. The vehicle capacity threshold is set by relevant departments. For example, the "Highway Capacity Manual" stipulates that the vehicle capacity threshold for a single lane in one direction is 1,800 vehicles per hour.

[0054] Broadcast alarm module: used to broadcast alerts when accidents or violations are detected in vehicles.

[0055] In a specific example, when an accident or violation is detected in any vehicle, a broadcast notification is given. The specific process is as follows: when an accident occurs, the broadcast devices within a preset distance of the accident are identified, and the broadcast is sent to each vehicle to remind them to slow down.

[0056] When a vehicle violates a traffic rule, the license plate number of the violating vehicle is obtained, and a notification is broadcast to the violating vehicle through the broadcasting equipment located closest to the violating vehicle.

[0057] Reference Figure 2 As shown, in its second aspect, this application provides a highway traffic diversion alarm broadcasting system and method based on AI recognition, including the following steps: Step 1, highway entrance diversion: vehicle information and driving information of each vehicle are collected through AI recognition technology, and then the safety behavior characteristic values ​​of each vehicle are analyzed to obtain them. At the same time, road information of each highway is obtained from the data center, and then the road behavior characteristic values ​​of each highway are analyzed to obtain them. Based on the safety behavior characteristic values ​​of each vehicle and the road behavior characteristic values ​​of each highway, vehicles are diverted, and different vehicle distance thresholds and speed thresholds are set according to the road characteristic values ​​of each highway.

[0058] Step 2, Traffic Diversion: Monitor all vehicles traveling on each highway and divert traffic when an accident is detected.

[0059] Step 3: Broadcast Alarm: When an accident or violation is detected in any vehicle, a broadcast will be made to alert the driver.

[0060] This application provides an AI-based method for broadcasting traffic diversion alerts on highways. By analyzing the load of each vehicle, the driving experience of each driver, the number of historical highway violations, and the number of historical highway accidents, the method diverts vehicles, directing high-risk vehicles to safer highway routes to ensure driving safety. Different safe following distances and speeds are set for different highways to prevent traffic accidents caused by excessively close following distances or excessive speeds. Furthermore, the method monitors all vehicles traveling on each highway, broadcasting alerts when accidents or violations are detected, thereby increasing deterrence on drivers and preventing repeated violations from leading to traffic accidents.

[0061] The above content is merely an example and illustration of the concept of this application. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the inventive concept or exceed the scope defined in this application, they should all fall within the protection scope of this application.

Claims

1. An expressway traffic diversion warning broadcasting system based on AI recognition, characterized in that, Comprise: Shunt module: including entrance shunt unit and road shunt unit; The highway entrance shunt unit is used for collecting vehicle information of each vehicle and driving information of each vehicle by AI recognition technology, and then analyzing to obtain safety behavior characteristic value of each vehicle, and obtaining road information of each highway from the data center, and then analyzing to obtain road behavior characteristic value of each highway, so as to shunt each vehicle according to safety behavior characteristic value of each vehicle and road behavior characteristic value of each highway, and set different vehicle distance threshold and vehicle speed threshold according to road characteristic value of each highway; The specific process of collecting vehicle information of each vehicle and driving information of each vehicle by AI recognition technology is as follows: Install sensor weighing system and AI camera equipment at the entrance of toll station, obtain the overall weight of each vehicle through the sensor weighing system, and capture the image of each vehicle and cab through the AI camera equipment, and then identify the model of each vehicle through the image of each vehicle, obtain the self weight and standard load of each vehicle according to the model of each vehicle, and at the same time, identify the face of each driver according to the image of each cab, and match the face image of each driver with the face image of each driver stored in the data center, and obtain the driving information of each driver after successful matching; The vehicle information of each vehicle includes the overall weight, self weight and standard load of the vehicle; the driving information of each vehicle includes the driving age, the number of historical highway accidents and the number of historical highway violations; The specific analysis process of analyzing to obtain safety behavior characteristic value of each vehicle is as follows: The overall weight, self-weight and standard load of each vehicle are obtained based on the vehicle information of each vehicle, and a vehicle safety evaluation coefficient aw of each vehicle is obtained by analysis i , i is the number of each vehicle, and i is a positive integer. Based on the driving information of each vehicle, the driving age, the number of historical high-speed accidents and the number of historical high-speed violations of each driver are obtained, and then the driving safety evaluation coefficient bw of each vehicle is analyzed i ; The vehicle safety evaluation coefficient and the driving safety evaluation coefficient of each vehicle are substituted into the vehicle behavior model, and the safety behavior characteristic value α of the i th vehicle is output according to the expression of the vehicle behavior model: The safety behavior characteristic value α of the i th vehicle is output. i where w' and w" are respectively the upper limit value and the lower limit value of the safety behavior evaluation coefficient of the vehicle. The specific analysis process of obtaining road information of each highway from the data center and then analyzing to obtain road behavior characteristic value of each highway is as follows: The road information of each highway includes the number of historical accidents, the number of curves and the angle of each curve of the highway; The historical accident occurrence number of each expressway and the angle of each curve are normalized and recorded as F j and N jl , wherein j is the number of each expressway, j is a positive integer, l is the number of each curve, and l is a positive integer. The historical accident occurrence number, the number of curves, and the angle of each curve of the normalized expressway are substituted into the road behavior model, and the road behavior model expression is: The road behavior characteristic value β j of the jth expressway is output, wherein L is the total number of curves, ρ1 and ρ2 respectively represent the weight factor corresponding to the historical accident number of the expressway and the weight factor corresponding to the angle of the curve, F' and N' are respectively the accident number standard value and the curve angle standard value of the expressway, wherein the minimum accident number of each expressway is recorded as the accident number standard value, and the minimum curve angle of each expressway is recorded as the curve angle standard value, e' and e'' respectively represent the upper limit value and the lower limit value of the road safety evaluation coefficient of the expressway. The specific process of shunting each vehicle according to safety behavior characteristic value of each vehicle and road behavior characteristic value of each highway is as follows: Shunt each vehicle with safety behavior characteristic value of 11 to each highway with road behavior characteristic value of 00, shunt each vehicle with safety behavior characteristic value of 10 to each highway with road behavior characteristic value of 10, and shunt each vehicle with safety behavior characteristic value of 00 to each highway with road behavior characteristic value of 11, thereby realizing the shunting of each vehicle; At the same time, each vehicle with safety behavior characteristic value of 11, 10 and 00 is recorded as each first type vehicle, each second type vehicle and each third type vehicle respectively; each vehicle with road behavior characteristic value of 11, 10 and 00 is recorded as each first type highway, each second type highway and each third type highway respectively; The specific process of setting different vehicle distance threshold and vehicle speed threshold according to road characteristic value of each highway is as follows: For each first type highway, set a first type vehicle distance, a first type vehicle speed minimum value and a first type vehicle speed maximum value to control each first type vehicle; For each second type highway, set a second type vehicle distance, a second type vehicle speed minimum value and a second type vehicle speed maximum value to control each second type vehicle; For each third type highway, set a third type vehicle distance, a third type vehicle speed minimum value and a third type vehicle speed maximum value to control each third type vehicle. For each of the three types of safe roads, the set of three types of vehicle distance, the minimum value of three types of vehicle speed and the maximum value of three types of vehicle speed control each of the three types of vehicles; Among them, the three-class vehicle distance is less than the two-class vehicle distance, which is less than the one-class vehicle distance; the maximum value of three-class vehicle speed is greater than the maximum value of two-class vehicle speed, which is greater than the maximum value of one-class vehicle speed; the minimum value of three-class vehicle speed is greater than the minimum value of two-class vehicle speed, which is greater than the minimum value of three-class vehicle speed; The road splitting unit is used to monitor each vehicle driving on each expressway, and when an accident is detected, each vehicle is accident diverted; The specific process of diverting each vehicle when an accident is detected is as follows: Record the accident occurrence point as the target accident occurrence point, and then obtain the closest diversion intersection to the target accident occurrence point, record it as the target diversion intersection, upload the accident image and video of the current expressway to the cloud data terminal, output the accident grade through the accident identification model, and then determine the number of diverted vehicles according to each accident grade; Then get the real-time vehicle quantity of the expressway associated with the target diversion intersection, and then judge whether the vehicle of the expressway associated with the target diversion intersection is saturated, if the vehicle is saturated, record the second closest diversion intersection to the target accident occurrence point as the target diversion intersection, and judge whether the vehicle of the expressway associated with the target diversion intersection is saturated until the diversion is completed. Broadcast alarm module: used for broadcasting and prompting when monitoring each vehicle accident or violation.

2. The AI recognition-based highway traffic diversion warning broadcast system of claim 1, wherein The vehicle safety evaluation coefficient and the driving safety evaluation coefficient of each vehicle are obtained by analysis, and the specific analysis process is as follows: Let the overall weight, self-weight and standard load of each vehicle be denoted as Q i , T i and P i , respectively, and the vehicle safety evaluation coefficient aw of the i-th vehicle be calculated according to the calculation formula: i where χ represents a correction factor for the set vehicle load. The driving age, the number of historical high-speed accidents and the number of historical high-speed violations of the driving personnel corresponding to each vehicle are normalized and recorded as S i , K i , G i , and the driving safety evaluation coefficient bw i of the i-th vehicle is calculated according to the calculation formula: bw i =S i *θ1+(K i *ω1+G i *ω2)*θ2, wherein ω1 and ω2 respectively represent the weight factor corresponding to the number of high-speed accidents of the driving personnel and the weight factor corresponding to the number of high-speed violations, and θ1 and θ2 respectively represent the weight factor corresponding to the driving age of the driving personnel and the weight factor corresponding to the driving skill evaluation coefficient.

3. The AI recognition-based highway traffic diversion warning broadcast system of claim 2, wherein The specific process of broadcasting and prompting when monitoring each vehicle accident or violation is as follows: When an accident occurs, obtain each broadcast device within a preset distance from the accident, broadcast to each driving vehicle through each broadcast device, and then remind each vehicle to slow down; When the vehicle is in violation, the license plate number of the vehicle is obtained, and the broadcast device closest to the violation vehicle is used to broadcast and prompt the violation vehicle.

4. An AI recognition-based highway traffic diversion warning broadcasting method performed by the AI recognition-based highway traffic diversion warning broadcasting system according to any one of claims 1 to 3, characterized by, Including: Step one, expressway entrance diversion: collect vehicle information and driving information of each vehicle through AI recognition technology, and then analyze to obtain safety behavior characteristic value of each vehicle, and obtain road information of each expressway from data center, and then analyze to obtain road behavior characteristic value of each expressway, so as to divert each vehicle according to safety behavior characteristic value of each vehicle and road behavior characteristic value of each expressway, and set different vehicle distance threshold and vehicle speed threshold according to road characteristic value of each expressway; The specific process of collecting vehicle information and driving information of each vehicle through AI recognition technology is as follows: The sensor weighing system and the AI camera equipment are installed at the entrance of the toll station, the overall weight of each vehicle is obtained through the sensor weighing system, the AI camera equipment is used to capture the image of each vehicle and the cab of each vehicle, the model of each vehicle is identified through the image of each vehicle, the self-weight and the standard load of each vehicle are obtained according to the model of each vehicle, the face of each driver is recognized according to the image of each cab, and the face image of each driver is matched with the face image of each driver stored in the data center, and the driving information of each driver is obtained after the matching is successful; The vehicle information of each vehicle includes the overall weight, the self-weight and the standard load of the vehicle, and the driving information of each vehicle includes the driving age, the number of historical high-speed accidents and the number of historical high-speed violations; The safety behavior characteristic value of each vehicle is obtained through the analysis, and the specific analysis process is as follows: The overall weight, self-weight and standard load of each vehicle are obtained based on the vehicle information of each vehicle, and a vehicle safety evaluation coefficient aw of each vehicle is obtained by analysis i , i is the number of each vehicle, and i is a positive integer. Based on the driving information of each vehicle, the driving age, the number of historical high-speed accidents and the number of historical high-speed violations of each driver are obtained, and then the driving safety evaluation coefficient bw of each vehicle is analyzed i ; The vehicle safety evaluation coefficient and the driving safety evaluation coefficient of each vehicle are substituted into the vehicle behavior model, and the vehicle behavior model expression is as follows: The safety behavior characteristic value α of the i-th vehicle is output i where w' and w" are the upper limit value and the lower limit value of the safety behavior evaluation coefficient of the vehicle, respectively. The road information of each expressway is obtained from the data center, and the road behavior characteristic value of each expressway is obtained through analysis, and the specific analysis process is as follows: The road information of each expressway includes the number of historical accidents, the number of curves and the angle of each curve of the expressway; The historical accident occurrence number of each expressway and the angle of each curve are normalized and recorded as F j and N jl , wherein j is the number of each expressway, j is a positive integer, l is the number of each curve, and l is a positive integer. The historical accident occurrence number, the number of curves, and the angle of each curve of the normalized expressway are substituted into the road behavior model, and the road behavior model expression is: The road behavior characteristic value β j of the jth expressway is output, wherein L is the total number of curves, ρ1 and ρ2 respectively represent the weight factor corresponding to the historical accident number of the expressway and the weight factor corresponding to the angle of the curve, F' and N' are respectively the accident number standard value and the curve angle standard value of the expressway, wherein the minimum accident number of each expressway is recorded as the accident number standard value, and the minimum curve angle of each expressway is recorded as the curve angle standard value, e' and e'' respectively represent the upper limit value and the lower limit value of the road safety evaluation coefficient of the expressway. The specific process is as follows: The vehicles with a safety behavior characteristic value of 11 are divided into the vehicles with a road behavior characteristic value of 00, the vehicles with a safety behavior characteristic value of 10 are divided into the vehicles with a road behavior characteristic value of 10, and the vehicles with a safety behavior characteristic value of 00 are divided into the vehicles with a road behavior characteristic value of 11, thereby realizing the division of the vehicles; Meanwhile, the vehicles with safety behavior characteristic values of 11, 10 and 00 are respectively recorded as each first type of vehicle, each second type of vehicle and each third type of vehicle; the vehicles with road behavior characteristic values of 11, 10 and 00 are respectively recorded as each first type of expressway, each second type of expressway and each third type of expressway; The specific process is as follows: For each first type of expressway, a first type of distance, a first type of speed minimum value and a first type of speed maximum value are set to control each first type of vehicle; For each second type of expressway, a second type of distance, a second type of speed minimum value and a second type of speed maximum value are set to control each second type of vehicle; For each third type of safety road, a third type of distance, a third type of speed minimum value and a third type of speed maximum value are set to control each third type of vehicle; The third type of distance is smaller than the second type of distance, which is smaller than the first type of distance; the third type of speed maximum value is greater than the second type of speed maximum value, which is greater than the first type of speed maximum value; the third type of speed minimum value is greater than the second type of speed minimum value, which is greater than the first type of speed minimum value; Step 2, mid-road division: each vehicle driving on each expressway is monitored, and each vehicle is divided when an accident is monitored; The specific process is as follows: The accident occurrence point is recorded as a target accident occurrence point, and then the closest shunting intersection to the target accident occurrence point is obtained and recorded as a target shunting intersection; the accident image and video of the current expressway are uploaded to a cloud data terminal, an accident grade is output through an accident identification model, and then the number of shunted vehicles is determined according to each accident grade; The real-time vehicle number of the expressway associated with the target shunting intersection is obtained, and then it is judged whether the vehicle of the expressway associated with the target shunting intersection is saturated; if the vehicle is saturated, the second closest shunting intersection to the target accident occurrence point is recorded as the target shunting intersection, and it is judged whether the vehicle of the expressway associated with the target shunting intersection is saturated until the shunting is completed. Step three, broadcast alarm: when monitoring that each vehicle has an accident or a violation, the broadcast is played to prompt.

Images