A traffic management method and system for avoiding continuous road non-obstruction congestion

Abstract

The application provides a traffic management method and system for avoiding continuous road non-obstacle congestion. The method is as follows: a road side unit receives driving information and vehicle ID sent by a vehicle OBU in a jurisdiction range, and reports the vehicle ID to a cloud control platform; the cloud control platform issues a dangerous driving behavior history distribution model of a vehicle in the jurisdiction range to each road side unit in a road section; the road side unit obtains road environment information in the jurisdiction range, and maps the road environment information, traffic flow and driving information to the dangerous driving behavior history distribution model of the vehicle to obtain a dangerous behavior probability score of each vehicle; the road side unit determines a speed adjustment instruction of a following vehicle according to the dangerous behavior probability score of the vehicle; and each vehicle OBU adjusts a speed adjustment prompt or adjusts a driving speed according to the speed adjustment instruction. The method can effectively judge and warn before a continuous road non-obstacle congestion event occurs, accurately and effectively guide automatic driving, improve traffic safety, and reduce congestion.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a traffic management method and system for avoiding continuous road congestion without barriers. Background Technology

[0002] Vehicles have become an essential tool for daily travel. On roads with limited traffic capacity, the number of vehicles is increasing, and congestion is becoming more and more serious. In real-world scenarios, the following congestion situations exist: When vehicles enter a continuous road, different drivers exhibit significant differences in their driving behavior. Without encountering obstacles, if the driver of the vehicle in front suddenly changes lanes or brakes suddenly due to low driving skills, lack of experience, or emotional factors, the following vehicles are forced to slow down or come to a stop. Due to the chain reaction, this behavior becomes more and more serious as subsequent vehicles join the road, eventually leading to road congestion. In severe cases, the congestion can last for several hours. This type of congestion in real-world scenarios is called continuous road non-obstacle congestion.

[0003] Existing technologies have seen some research on improving intersection congestion, including vehicle-centric intelligence led by automakers and congestion mitigation measures based on roadside monitoring and management, primarily relying on intelligent infrastructure. However, these approaches suffer from limitations such as the limited range of single-vehicle perception and the simplistic nature of roadside monitoring and management, failing to effectively prevent and handle continuous non-obstacle-prone road congestion events. While vehicle-road cooperative technologies have made significant improvements in vehicle-road communication, the lack of robust backend algorithms hinders their ability to prevent and respond to incidents in advance. They are unable to effectively assess and warn of continuous non-obstacle-prone road congestion events, thus failing to provide accurate and effective guidance for autonomous driving. Summary of the Invention

[0004] The present invention aims to at least solve the technical problems existing in the prior art, and provide a traffic management method and system for avoiding non-barrier congestion on continuous roads.

[0005] To achieve the above-mentioned objectives of the present invention, according to a first aspect of the present invention, the present invention provides a traffic management method for avoiding continuous non-barrier congestion on roads, which executes in each control cycle as follows: a roadside unit receives driving information and vehicle IDs sent by vehicle OBUs within its jurisdiction, calculates traffic flow based on the received vehicle IDs, and reports the vehicle IDs to a cloud control platform; the cloud control platform sends a historical distribution model of dangerous driving behaviors of vehicles within the jurisdiction of each roadside unit to each roadside unit within the road segment, the dangerous driving behaviors including sudden braking and / or sudden lane changes; the roadside unit obtains road environment information perceived by roadside sensing devices within its jurisdiction; the roadside unit maps the road environment information, traffic flow, and driving information of each vehicle within its jurisdiction to the historical distribution model of dangerous driving behaviors of the vehicle to obtain the probability of dangerous behavior of the vehicle; the roadside unit determines the speed adjustment command of the vehicle following the vehicle based on the probability of dangerous behavior of each vehicle within its jurisdiction, and sends the speed adjustment command to the corresponding vehicle OBU; each vehicle OBU provides a speed adjustment prompt or adjusts its driving speed according to the received speed adjustment command.

[0006] To achieve the above-mentioned objectives of the present invention, according to a second aspect of the present invention, the present invention provides a system based on the traffic management method for avoiding continuous road non-barrier congestion provided in the first aspect of the present invention, comprising a cloud control platform, roadside devices and roadside sensing devices installed on the road, and vehicle OBUs on vehicles in the road segment; each roadside device manages the roadside sensing devices and vehicle OBUs in a region on the road segment, and the roadside devices are respectively connected to the roadside sensing devices and vehicle OBUs within their jurisdiction, and also connected and communicate with the cloud control platform.

[0007] The above technical solution: In each control cycle, the roadside unit uploads the vehicle IDs within its jurisdiction to the cloud control platform in real time. The cloud control platform queries the corresponding historical distribution model of dangerous driving behavior from the pre-built model library and sends it to the roadside unit. Based on the historical distribution model of dangerous driving behavior and the current driving information and road environment information of each vehicle, the roadside unit obtains the probability of dangerous behavior of each vehicle within its jurisdiction. The higher the probability of dangerous behavior, the greater the probability of the vehicle performing dangerous driving behaviors such as sudden braking and sudden lane changes. After obtaining the probability of dangerous behavior of each vehicle, the corresponding speed adjustment command is issued to the following vehicles according to the probability of dangerous behavior of each vehicle. By adjusting the speed of the following vehicles, continuous non-obstacle congestion events on the road and the traffic safety accidents caused by such events are avoided. This application can effectively judge and warn before continuous non-obstacle congestion events on the road occur, provide accurate and effective guidance for autonomous driving, improve traffic safety, reduce congestion, and the calculation process of this method is distributed between the roadside unit and the cloud control platform, which improves the response rate. Attached Figure Description

[0008] Figure 1This is a flowchart illustrating a preferred embodiment of the traffic management method for avoiding non-barrier congestion on continuous roads according to the present invention.

[0009] Figure 2 This is a schematic diagram illustrating the calculation of the joint distribution probability of driving time and traffic flow in a preferred embodiment of the present invention;

[0010] Figure 3 This is a schematic diagram of the system layout in a preferred embodiment of the present invention. Detailed Implementation

[0011] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0012] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0013] In the description of this invention, unless otherwise specified and limited, it should be noted that the terms "installation", "connection" and "linking" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, or internal connections between two components. They can be direct connections or indirect connections through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0014] This invention discloses a traffic management method for avoiding non-barrier congestion on continuous roads, such as... Figure 1 As shown, this method is based on Figure 3 The system implementation shown describes a method that executes control according to a control cycle. The control cycle is preferably, but not limited to, any value within the range of 0.5 to 5 seconds. Within each control cycle, the following is executed:

[0015] Step S1: The roadside unit (RSU) receives driving information and vehicle IDs sent by the vehicle OBUs within its jurisdiction, calculates traffic flow based on the received vehicle IDs, and reports the vehicle IDs to the cloud control platform.

[0016] In this embodiment, such as Figure 3As shown, a roadside unit (RSU) is installed on the roadside. It can communicate with the vehicle's onboard unit (OBU) and also with a remote cloud control platform, which is preferably, but not limited to, a cloud server. This cloud control platform is responsible for the information management of the entire road segment. Multiple roadside units are distributed sequentially and at intervals within the road segment. Each roadside unit has a jurisdiction area on the road segment, and all vehicles within its jurisdiction are equipped with one. The roadside unit communicates with the OBUs of the vehicles within its jurisdiction.

[0017] In this implementation, the vehicle OBUs within the jurisdiction periodically upload driving information and vehicle IDs. Driving information includes vehicle speed and the duration of the current trip. If the time interval between the start time of the current trip and the end time of the previous trip is greater than or equal to a preset duration, the start time of that trip is considered the start time of a new trip. The duration of the current trip represents the time interval between the current time and the start time of this trip. The preset duration is preferably, but not limited to, 0.5 hours or more. The roadside unit counts the number of vehicle IDs received within its jurisdiction per unit time to obtain traffic flow.

[0018] Step S2: The cloud control platform sends a historical distribution model of dangerous driving behaviors of vehicles within the jurisdiction of each roadside unit to each roadside unit in the road segment. Dangerous driving behaviors include sudden braking and / or sudden lane changes.

[0019] In this embodiment, preferably, to improve management efficiency, the cloud control platform internally constructs a model library, which contains multiple historical distribution models of dangerous driving behaviors that correspond one-to-one with vehicle IDs. The cloud control platform queries the model library for the corresponding pre-built historical distribution model of dangerous driving behaviors based on the vehicle ID, and then distributes the historical distribution model of dangerous driving behaviors corresponding to each vehicle ID to the roadside unit to which that vehicle ID belongs.

[0020] Step S3: The roadside unit acquires road environment information sensed by roadside sensing devices within its jurisdiction.

[0021] In this embodiment, road environment information includes weather type and road type. The roadside sensing device includes a weather detection module, a processor, and a communication module. The weather detection module includes a rain and snow sensor for detecting rain and snow. The processor obtains the weather type based on the output information of the rain and snow sensor, which includes rainy / snowy days and non-rainy / snowy days. The processor uploads the obtained weather type and pre-stored road type to its corresponding roadside unit via the communication module. Road types include urban roads and expressways.

[0022] In step S4, the roadside unit maps the road environment information and driving information of each vehicle within its jurisdiction to the vehicle's dangerous driving behavior history distribution model to obtain the probability of the vehicle's dangerous behavior.

[0023] In this embodiment, dangerous driving behaviors such as sudden braking and abrupt lane changes are related to the driver's driving habits. When drivers misjudge road conditions ahead, encounter complex traffic environments, lack experience, or experience tension, anxiety, or other mental and emotional states, they may habitually engage in dangerous driving behaviors. Therefore, a historical distribution model of dangerous driving behaviors constructed using vehicle historical driving data can effectively reflect drivers' dangerous driving habits and identify high-risk vehicles on road sections.

[0024] In this embodiment, preferably, the historical distribution model of dangerous driving behavior includes weather probability distribution, road type probability distribution, vehicle speed probability distribution, and a joint probability distribution of driving time and traffic volume. Weather, road type, vehicle speed, current driving time, and traffic volume can all trigger dangerous driving habits in drivers. For example, in rainy or snowy weather, drivers may habitually trigger dangerous driving behaviors due to anxiety or lack of experience, as their vision is obstructed and they cannot accurately judge the road conditions ahead. When the road type is an urban highway, the high population and vehicle density in the city may cause tension for inexperienced drivers, leading to dangerous driving behaviors. Furthermore, in heavy traffic, inexperienced drivers may also experience tension, potentially triggering dangerous driving behaviors. When the current driving time is excessively long, the driver may become fatigued, which can also trigger dangerous driving behaviors. Excessive traffic volume and long driving time both positively contribute to triggering dangerous driving behaviors, while low traffic volume and short driving time reduce the likelihood of triggering dangerous driving behaviors. Therefore, establishing a joint probability distribution of driving time and traffic volume can more accurately estimate the probability of dangerous driving behaviors occurring.

[0025] In this embodiment, more preferably, step S4 includes:

[0026] Step S41: Obtain the probability of dangerous driving behavior corresponding to the current weather type based on the weather probability distribution, and denot it as the first probability.

[0027] Step S42: Obtain the probability of dangerous driving behavior corresponding to the current road type based on the road type probability distribution, denoted as the second probability.

[0028] Step S43: Based on the vehicle speed probability distribution, obtain the probability of dangerous driving behavior occurring in the speed range where the current vehicle speed of each vehicle is located, denoted as the third probability.

[0029] Step S44: Based on the joint probability distribution of driving time and traffic flow, obtain the probability of dangerous driving behavior corresponding to the current driving time and traffic flow of each vehicle in this trip, which is denoted as the fourth probability.

[0030] Step S45: Combine the first probability, second probability, third probability and fourth probability of each vehicle to determine the probability of dangerous behavior of each vehicle.

[0031] Specifically, the probability of dangerous behavior for each vehicle can be obtained by averaging the first probability, the second probability, the third probability, and the fourth probability.

[0032] In step S5, the roadside unit determines the speed adjustment command for the vehicle following each vehicle based on the probability of dangerous behavior of each vehicle within its jurisdiction, and sends the speed adjustment command to the corresponding OBU of the following vehicle. By adjusting the speed of the following vehicle, safety is improved, and non-obstacle congestion caused by sudden braking or lane changes by the vehicle in front is avoided.

[0033] Preferably, step S5 includes:

[0034] If the probability of dangerous behavior of the vehicle is less than a preset probability threshold, preferably but not limited to 0.7 to 0.8, then the speed adjustment command of the vehicle behind the vehicle is to maintain the current speed.

[0035] If the probability of dangerous behavior of the vehicle is greater than or equal to the probability threshold, and the current distance between the vehicle and the following vehicle is greater than or equal to the braking distance of the following vehicle, then the speed adjustment command of the following vehicle is to maintain the current speed.

[0036] If the probability of a vehicle's dangerous behavior is greater than or equal to a probability threshold, and the current distance between the vehicle and the following vehicle is less than the braking distance of the following vehicle, then the speed adjustment command for the following vehicle is to adjust its speed to be less than the speed of the vehicle in front. In step S6, each vehicle's OBU provides a speed adjustment prompt or adjusts its driving speed according to the speed adjustment command. This can be done via voice prompts to the driver to slow down, or automatically through the vehicle's electronic control system.

[0037] In a preferred embodiment of the present invention, the process by which the cloud control platform pre-constructs a historical distribution model of dangerous driving behavior for each vehicle is as follows:

[0038] Step A: Collect historical driving data of the vehicle corresponding to the vehicle ID, identify dangerous driving behaviors from the historical driving data, and extract the weather type, road type, vehicle speed, traffic flow, and driving time corresponding to each dangerous driving behavior.

[0039] Historical vehicle driving data includes vehicle driving records uploaded by the vehicle's On-Board Unit (OBU) across the entire road network and weather and road type data uploaded by roadside sensing devices corresponding to the vehicle's travel segment. Vehicle driving records include braking data and the braking time for each braking action; if the braking time is less than the standard braking time, it is considered an emergency braking event. Vehicle driving records also include steering wheel angle; when the steering wheel angle exceeds a set threshold, it is considered a sharp turn, i.e., a sudden lane change. The number of emergency braking and sharp turns is counted, and the corresponding weather type, road type, vehicle speed, traffic flow, and travel time for each emergency braking or sharp turn are obtained.

[0040] Step B involves statistically analyzing the frequency of dangerous driving behaviors for the vehicle under different weather conditions. Based on these frequencies, the probability of dangerous driving behaviors occurring under different weather conditions is obtained, resulting in a weather probability distribution. The frequencies of dangerous driving behaviors corresponding to all weather types are linearly mapped to the interval between 0 and 1 to obtain the probability of dangerous driving behaviors occurring under different weather conditions. The sum of the probabilities of dangerous driving behaviors under all weather types is 1. For example, if the frequency of dangerous driving behaviors for a vehicle is 2 on non-rainy / snowy days and 8 on rainy / snowy days, then the probability of dangerous driving behaviors on non-rainy / snowy days is 0.2, and the probability of dangerous driving behaviors on rainy / snowy days is 0.8. The weather probability distribution is represented as follows:

[0041]

[0042] Among them, f no-rs f represents the frequency of dangerous driving behavior by vehicles in non-snowy or rainy weather. rs This represents the frequency of dangerous driving behaviors of the vehicle in rainy or snowy weather, with F indicating the total number of dangerous driving behaviors of the vehicle.

[0043] Step C involves statistically analyzing the frequency of dangerous driving behaviors on different road types. Based on these frequencies, the probability of dangerous driving behaviors on different road types is obtained, resulting in a road type probability distribution. The frequencies of dangerous driving behaviors corresponding to all road types are linearly mapped to the interval between 0 and 1 to obtain the probability of dangerous driving behaviors on different road types. The sum of the probabilities of dangerous driving behaviors on all road types is 1. For example, if the frequency of dangerous driving behaviors by a certain vehicle on urban roads is 7 and on highways is 3, then the probability of dangerous driving behaviors on urban roads is 0.7, and the probability of dangerous driving behaviors on highways is 0.3. The road type probability distribution is expressed as:

[0044]

[0045] Among them, f H f represents the frequency of dangerous driving behaviors by vehicles on highways. LThis refers to the frequency of dangerous driving behaviors on urban roads.

[0046] Step D: Count the frequency of dangerous driving behaviors in different speed ranges of the vehicle, obtain the probability of dangerous driving behaviors in different speed ranges based on the frequency, and obtain the speed probability distribution.

[0047] The vehicle speed range obtained from the vehicle driving record is [v a ,v b The vehicle speed range is divided into multiple consecutive speed intervals. The frequency of dangerous driving behaviors in different speed intervals is counted. The frequency of dangerous driving behaviors in all speed intervals is linearly mapped to the interval [0,1] to obtain the probability of dangerous driving behavior in each speed interval. The sum of the probabilities of dangerous driving behaviors in all speed intervals is 1. Suppose there are K speed intervals, and suppose the vehicle speed v falls into the k-th speed interval, then the vehicle speed probability distribution is expressed as:

[0048]

[0049] Among them, f k Let v represent the frequency of dangerous driving behavior that occurs in the k-th speed range where v falls, k∈[1,K].

[0050] In this embodiment, and more preferably, to establish a more accurate joint probability distribution model of driving time and traffic flow, the model construction process includes:

[0051] Step E1: Collect the driving time-traffic flow rate pairs for each occurrence of a dangerous driving behavior in the vehicle's historical driving data, obtaining a set of driving time-traffic flow rate pairs. The set S of driving time-traffic flow rate pairs is represented as:

[0052] S={(t1,q1),(t2,q2),…,(t F ,q F )};

[0053] Let i denote the index of dangerous driving behavior, 1≤i≤F, t i Let q represent the driving duration corresponding to the i-th dangerous driving incident. i This represents the traffic flow corresponding to the i-th dangerous driving behavior.

[0054] Step E2: Perform clustering on the set of driving time - traffic flow values ​​to obtain cluster centers, denoted as (t0, q0).

[0055] Step E3, as follows Figure 2As shown, a two-dimensional plane is established with driving time as the x-axis and traffic flow as the y-axis. On this two-dimensional plane, each driving time-traffic flow value pair corresponds to a value point. In the two-dimensional plane, a preset circle is first drawn with the cluster center as the center. This preset circle is required to enclose at least 30% of the value points. Then, multiple annexes with a width equal to the radius of the preset circle are drawn sequentially. The width of the annexes represents the difference between the outer radius and the inner radius of the annexes. The number of value points in the set of driving time-traffic flow value pairs (value points) that fall into the circle and each annexes is counted. Based on the number of value points falling into each annexes, the probability of dangerous driving behavior corresponding to the circle and each annexes is obtained, thus obtaining the joint probability distribution of driving time-traffic flow.

[0056] In this embodiment, the circles and rings represent different value ranges. The range division method described above can fully reflect the correlation between driving time and traffic flow, improve the accuracy of value range division, and thus improve the accuracy of the obtained probability of dangerous driving behavior.

[0057] In this embodiment, the probability of dangerous driving behavior corresponding to the circle and each ring is obtained based on the number of numerical points that fall into the circle. Specifically, the number of numerical points falling into the circle and all rings can be linearly mapped to the interval [0,1] to obtain the probability of dangerous driving behavior corresponding to the circle and each ring. The sum of the probabilities of dangerous driving behavior corresponding to the circle and all rings is 1.

[0058] In this embodiment, there are m annexes, and a total of m+1 regions on the two-dimensional plane, including the circles. The circle represents the 0th region, the first annulus from the inside out represents the 1st region, and so on. Let the point pair (t...) i ,q i If a given region belongs to the j-th region, then the joint probability distribution of travel time and traffic flow can be expressed as:

[0059]

[0060] Among them, f j Represents a pair of numerical points (t) i ,q i The number of historical value pairs that fall into the j-th region, j∈[0,m].

[0061] In a preferred embodiment, the risk coefficient of each vehicle is obtained by calculating the weighted sum of the first, second, third, and fourth probabilities. Let the risk coefficient of a vehicle be denoted as W, then: W = λ1P1 + λ2P2 + λ3P3 + λ4P4, where λ1 represents the weight of the first probability of the vehicle, ranging from 0 to 1; λ2 represents the weight of the second probability of the vehicle, ranging from 0 to 1; λ3 represents the weight of the third probability of the vehicle, ranging from 0 to 1; λ4 represents the weight of the fourth probability of the vehicle, ranging from 0 to 1; λ1 + λ2 + λ3 + λ4 = 1. λ1 and λ2 can be set according to the vehicle's age. If the vehicle is older than 2 years, λ1 and λ2 are set to 0.05 to 0.1; if the vehicle is younger than 2 years, λ1 and λ2 are set to 0.2. λ3 can be selected from a value within the range of 0.2 to 0.3. This results in a higher and more accurate value for λ4.

[0062] In a preferred embodiment, within each control cycle, the following is also performed:

[0063] The roadside equipment monitors whether dangerous driving behavior events occur within its jurisdiction. If such events occur, the roadside equipment issues a deceleration command to the vehicle OBU within its jurisdiction and reports the event to the cloud control platform. The cloud control platform then adjusts the speed of vehicles within the subsequent jurisdiction of the roadside equipment.

[0064] In this implementation, the presence of emergency braking can be determined by identifying whether the emergency brake lights of the vehicle in front are on in images captured by roadside sensing devices or following vehicles. Alternatively, the roadside equipment can use vehicle driving data uploaded by each vehicle's OBU to determine whether there is emergency braking (rapid decrease in speed) or sharp turning (steering wheel angle greater than a certain threshold). If such situations exist, it is considered that the vehicle has engaged in dangerous driving behavior. In this case, the roadside equipment issues a deceleration command to all vehicle OBUs within its jurisdiction, especially vehicles following the vehicle where the incident occurred, adopting a rapid deceleration principle. To avoid long traffic jams, the cloud control platform also sends deceleration commands to vehicles within the jurisdiction of subsequent roadside equipment.

[0065] This invention also discloses a traffic management system based on the above-described traffic management method for avoiding non-barrier congestion on continuous roads, such as... Figure 3 As shown, it includes a cloud control platform, roadside equipment and roadside sensing equipment installed on the road, and vehicle OBUs on the road segment; each roadside equipment manages the roadside sensing equipment and vehicle OBUs within a road segment area, and the roadside equipment is connected and communicates with the roadside sensing equipment and vehicle OBUs within its jurisdiction, as well as the cloud control platform.

[0066] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0067] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A traffic management method for avoiding non-barrier congestion on continuous roads, characterized in that, Execute within each control cycle: The roadside unit receives driving information and vehicle IDs sent by the vehicle OBUs within its jurisdiction, calculates traffic flow based on the received vehicle IDs, and reports the vehicle IDs to the cloud control platform. The cloud control platform sends a historical distribution model of dangerous driving behaviors of vehicles within the jurisdiction of each roadside unit to each roadside unit in the road section. Dangerous driving behaviors include sudden braking and / or sudden lane changes. The roadside unit acquires road environment information sensed by roadside sensing devices within its jurisdiction; The roadside unit maps road environment information, traffic flow and driving information of each vehicle within its jurisdiction to the vehicle’s dangerous driving behavior history distribution model to obtain the probability of the vehicle’s dangerous behavior. The roadside unit determines the speed adjustment command of the vehicle following the vehicle based on the probability of dangerous behavior of each vehicle within its jurisdiction, and sends the speed adjustment command to the corresponding vehicle's OBU. Each vehicle's OBU provides speed adjustment prompts or adjusts the driving speed according to the speed adjustment command.

2. The traffic management method for avoiding non-barrier congestion on continuous roads as described in claim 1, characterized in that, The cloud control platform distributes historical distribution models of dangerous driving behaviors of vehicles within the jurisdiction of each roadside unit to each roadside unit within the road segment, including: The cloud control platform queries the model library based on the vehicle ID to find the pre-built historical distribution model of dangerous driving behavior corresponding to the vehicle ID, and then distributes the historical distribution model of dangerous driving behavior corresponding to each vehicle ID to the roadside unit to which the vehicle ID belongs.

3. The traffic management method for avoiding continuous road congestion as described in claim 1 or 2, characterized in that, The road environment information includes weather and road type; The driving information includes the current vehicle speed and the current driving duration. The historical distribution model of dangerous driving behavior includes weather probability distribution, road type probability distribution, vehicle speed probability distribution, and joint probability distribution of driving time and traffic flow.

4. The traffic management method for avoiding non-barrier congestion on continuous roads as described in claim 3, characterized in that, The process by which the cloud control platform pre-builds a historical distribution model of dangerous driving behaviors for each vehicle is as follows: Collect historical driving data of vehicles, identify dangerous driving behaviors from the historical driving data, and extract the weather type, road type, vehicle speed, traffic flow and driving time corresponding to each dangerous driving behavior; The frequency of dangerous driving behaviors of the vehicle under different weather conditions is statistically analyzed, and the probability of dangerous driving behaviors under different weather conditions is obtained based on the frequency, thus obtaining the weather probability distribution; The frequency of dangerous driving behavior of the vehicle under different road types is statistically analyzed, and the probability of dangerous driving behavior under different road types is obtained based on the frequency, thus obtaining the probability distribution of road types; The frequency of dangerous driving behaviors in different speed ranges of the vehicle is statistically analyzed, and the probability of dangerous driving behaviors in different speed ranges is obtained based on the frequency, thus obtaining the speed probability distribution.

5. The traffic management method for avoiding continuous road congestion as described in claim 4, characterized in that, The process of constructing the joint probability distribution of driving time and traffic flow includes: Collect the driving time-traffic flow rate pairs for each occurrence of dangerous driving behavior in the vehicle's historical driving data, and obtain a set of driving time-traffic flow rate pairs. Clustering is performed on the set of driving time-traffic volume values ​​to obtain cluster centers; A two-dimensional plane is established with driving time as the horizontal axis and traffic flow as the vertical axis. In the two-dimensional plane, a preset circle is first drawn with the cluster center as the center. Then, multiple annexes with a width equal to the radius of the circle are drawn in sequence. The number of logarithms of the driving time-traffic flow value pairs falling into the circle and each annexe is counted. The probability of dangerous driving behavior corresponding to the circle and each annexe is obtained based on the number of logarithms of the values ​​falling into the circle, and then the joint probability distribution of driving time-traffic flow is obtained.

6. The traffic management method for avoiding non-barrier congestion on continuous roads as described in claim 5, characterized in that, The roadside unit maps road environment information, traffic flow, and driving information of each vehicle within its jurisdiction to a historical distribution model of dangerous driving behavior for that vehicle to obtain the probability of dangerous behavior for each vehicle, including: The probability of dangerous driving behavior occurring corresponding to the current weather type is obtained based on the weather probability distribution and is denoted as the first probability. The probability of dangerous driving behavior occurring for the current road type is obtained based on the road type probability distribution, and is denoted as the second probability. The probability of dangerous driving behavior occurring in the speed range where the current vehicle speed is located is obtained based on the vehicle speed probability distribution, and is denoted as the third probability. The probability of dangerous driving behavior corresponding to the current driving time and the current traffic flow of each vehicle is obtained based on the joint probability distribution of driving time and traffic flow, and is denoted as the fourth probability. The probability of dangerous behavior for each vehicle is determined by combining the first, second, third, and fourth probabilities.

7. The traffic management method for avoiding non-barrier congestion on continuous roads as described in claim 6, characterized in that, The determination of the probability of dangerous behavior for each vehicle by combining the first, second, third, and fourth probabilities of each vehicle includes: Calculate the weighted sum of the first, second, third, and fourth probabilities for each vehicle to obtain the probability of dangerous behavior for each vehicle.

8. The traffic management method for avoiding non-barrier congestion on continuous roads as described in claim 7, characterized in that, The roadside unit determines the speed control command for the vehicle behind it based on the probability of dangerous behavior of each vehicle within its jurisdiction, including: If the probability of dangerous behavior of the vehicle is less than the preset probability threshold, the speed adjustment command of the vehicle behind the vehicle is to maintain the current speed. If the probability of dangerous behavior of the vehicle is greater than or equal to the probability threshold, and the current distance between the vehicle and the following vehicle is greater than or equal to the braking distance of the following vehicle, then the speed adjustment command of the following vehicle is to maintain the current speed. If the probability of dangerous behavior of the vehicle is greater than or equal to the probability threshold, and the current distance between the vehicle and the following vehicle is less than the braking distance of the following vehicle, then the speed adjustment command of the following vehicle is to adjust its speed to be less than the speed of the vehicle in front.

9. The traffic management method for avoiding continuous road congestion as described in claim 1, 2, 4, 5, 6, 7, or 8, characterized in that, Within each control cycle, the following is also executed: The roadside equipment monitors whether dangerous driving behavior events occur within its jurisdiction. If such events occur, the roadside equipment issues a deceleration command to the vehicle OBU within its jurisdiction and reports the event to the cloud control platform. The cloud control platform then adjusts the speed of vehicles within the subsequent jurisdiction of the roadside equipment.

10. A system based on the traffic management method for avoiding continuous road unimpeded congestion as described in any one of claims 1-9, characterized in that, This includes cloud control platforms, roadside equipment and roadside sensing devices installed on the road, and vehicle OBUs on vehicles in the road segment; Each roadside device manages the roadside sensing devices and vehicle OBUs within a certain area of ​​the road segment. The roadside devices are connected to the roadside sensing devices and vehicle OBUs within their jurisdiction, and also communicate with the cloud control platform.

Images