Method for identifying key vehicles based on roadside group trajectories and intervention through navigation

Abstract

This invention relates to a method for identifying key vehicles based on roadside group trajectories and intervening through navigation, comprising the following steps: S1, acquiring real-time traffic information and vehicle motion parameters of the road segment through radar trajectory data; S2, setting a threshold for the interaction and influence distance between vehicles to determine the impact of surrounding vehicles; S3, obtaining different vehicle group types according to the vehicle group identification model process; S4, completing the classification and division of vehicles across the entire road area, verifying the vehicle composition of each vehicle group, and performing vehicle group fusion; S5, constructing an evaluation index system, comparing the deviation of vehicle motion parameters within the vehicle group from the overall motion parameters, and identifying key vehicles based on preset thresholds; S6, verifying the navigation binding status of key vehicles and monitoring and correcting their behavior. This invention can identify different vehicle group types in traffic flow through roadside trajectories, identify key vehicles in the vehicle group that affect traffic safety or efficiency, and manage them through real-time navigation alerts, effectively improving driving safety or efficiency and reducing traffic accident rates and traffic delays.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and more specifically, to a method for identifying key vehicles based on roadside group trajectories and intervening through navigation. Background Technology

[0002] With the acceleration of urbanization and the continuous growth of motor vehicle ownership, road traffic flow is becoming increasingly dense, and group driving scenarios are becoming more and more common, which in turn increases traffic risks. During group driving, some vehicles' undesirable behaviors, such as continuous lane changes, slow lane occupation, and sudden acceleration and deceleration, can not only easily lead to local traffic congestion, but may also induce rear-end collisions, scrapes, and other traffic accidents, posing a serious threat to road safety.

[0003] However, existing traffic management technologies largely rely on single-point vehicle capture or driver feedback, leading to limited monitoring range, delayed response, and insufficient targeted vehicle control. For example, traditional electronic traffic enforcement systems can only record violations at fixed locations, failing to identify vehicles that significantly impact overall traffic safety from a group perspective. Furthermore, even when dangerous behavior is detected, it's difficult to quickly and accurately deliver alerts to the corresponding drivers, resulting in untimely risk intervention. In addition, while existing navigation systems offer route planning and basic traffic alerts, they lack deep integration with roadside traffic sensing systems, hindering the identification of critical risk vehicles and targeted alerts based on real-time vehicle trajectory data. Therefore, accurately identifying key vehicles from complex vehicle trajectories and efficiently and specifically delivering risk alerts to them has become an urgent need to improve road safety. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a method for identifying key vehicles based on roadside group trajectory and intervening through navigation. This method can identify vehicle groups by collecting vehicle group trajectory data in real time through roadside sensing devices, accurately identify vehicles that have a key impact on traffic flow safety, and then push risk behavior reminders to target vehicles through the navigation system, thereby achieving proactive intervention and precise control of traffic risks.

[0005] The technical solution adopted by this invention to solve its technical problem is: to construct a method for identifying key vehicles based on roadside group trajectories and intervening through navigation, including the following steps: S1. Obtain real-time traffic information and vehicle motion parameters of road sections through radar trajectory data; S2. Define the vehicle group area, divide the scenario according to the lane and vehicle type, set the distance threshold for the interaction between vehicles, and judge the impact of surrounding vehicles. S3. Through threshold verification, different vehicle group types are obtained according to the vehicle group identification model process; S4. Complete the classification and division of vehicles in the entire road area, verify the vehicle composition of each vehicle group and merge the vehicle groups. S5. Construct an evaluation index system, compare the deviation of vehicle motion parameters within the vehicle group with the overall motion parameters, and identify key vehicles by combining preset thresholds. S6. Verify the navigation binding status of key vehicles and monitor and correct their behavior through navigation orientation reminders and feedback.

[0006] According to the above scheme, the specific process of obtaining real-time traffic information of the road segment and the motion parameters of each vehicle in step S1 is as follows: S101. A combination of millimeter-wave radar and video capture equipment capable of recognizing license plates is used to obtain real-time traffic information and vehicle motion parameters of road sections. The equipment covers three lanes in one direction and the data acquisition frequency is not less than 10Hz. S102. Real-time acquisition of the movement trajectories of all vehicles within the monitoring area, designating any single vehicle used for observation and analysis in a three-lane driving environment as the target vehicle, and obtaining the vehicle's local coordinates. ), driving speed acceleration The vehicle's direction of travel and license plate information are used to generate raw motion trajectory data. S103. Calculate the longitudinal distance between the target vehicle and the vehicle in the same lane directly in front, based on local coordinates. The straight-line distance from vehicles in the adjacent lanes on the left and right along the direction of travel ( ), calculate the headway between the target vehicle and the vehicle directly in front based on the obtained longitudinal distance. The formula for calculating the headway is as follows:

[0007] According to the above scheme, in step S2, the specific process of setting the distance threshold for interaction between vehicles and judging the influence of surrounding vehicles is as follows: S201. Determine the distance threshold system for inter-vehicle interaction effects: The vertical influence threshold is or headway <4s, the combined lateral and longitudinal impact thresholds for the left and right lanes are respectively ,

[0008] Negative distance values ​​indicate that the vehicle is in the area behind the target vehicle, while positive values ​​indicate that it is in the area in front of the target vehicle. The longitudinal interaction impact threshold for large vehicles is set to be twice that for small vehicles. The interaction impact distance threshold system for large vehicles is as follows: , , ; S202. When the target vehicle is in the left or right lane, focus on monitoring vehicles in the adjacent inner lane and the lane directly in front; when it is in the middle lane, simultaneously monitor vehicles in the adjacent lanes on both sides and the lane directly in front; when there are multiple interfering vehicles on the same side, retain the closest vehicle within the range of influence as the core interfering vehicle according to the principle of "minimum distance priority".

[0009] According to the above scheme, in step S3, different vehicle group types are obtained based on the vehicle group recognition model process. The specific process is as follows: S301. Based on roadside group trajectories and preset thresholds, five basic types of vehicle groups are obtained through a vehicle group recognition model process. S302. Determine the lane type of the target vehicle. Based on the lane position attributes of the target vehicle, collect the longitudinal distance between the target vehicle and the vehicle in front and the lateral distance between the vehicles in the left and right lanes to complete the threshold verification. S303. To distinguish between the different vehicle group types corresponding to the left and right lanes in the outer lanes, different correction values ​​are assigned to avoid confusion of vehicle group types caused by the same combination of basic parameters within the outer lanes, leading to the introduction of lane differentiation correction coefficients. The calculation formula is as follows:

[0010] in, Lane position indicators: 1 indicates the outer lane, including the left and right lanes; 2 indicates the middle lane. Let be the lateral position coordinates of the target vehicle within the lane. The lateral coordinates of the road's transverse centerline; S304. By integrating threshold tests and parameters, a comprehensive mapping formula is constructed to fully cover various typical scenarios of target vehicles in the left lane, right lane, and middle lane, achieving initial vehicle group type identification. The calculation formula is as follows:

[0011] in, This serves as the initial vehicle group type identifier; Warning sign indicating an impact on vehicles ahead. This is a sign indicating the impact on vehicles in the left lane. This is an indication of the impact on vehicles in the right lane; S305. After obtaining the initial vehicle group, the core dimensions are divided according to the direction of interference from the vehicle in front and the lateral interference. The influence of the lane is weakened, the positional difference redundancy under the same interference is eliminated, the key features are retained and the complexity is reduced. After being adapted to the actual road conditions, five basic vehicle group types are formed.

[0012] According to the above scheme, in step S305, the five basic vehicle group types are: a vehicle group driving freely without interference, a vehicle group only affected by the vehicle in front, a vehicle group affected by interference on one side of the lane, a vehicle group affected by a combination of interference from the vehicle in front and one side of the lane, and a vehicle group affected by a combination of interference from all directions.

[0013] According to the above scheme, in step S4, the vehicle group classification of all vehicles in the road area is completed, the vehicle composition of each group is verified, and the vehicle groups are merged. The specific process is as follows: S401: Based on the five vehicle group types, classify all vehicles traveling on the road one by one, complete the vehicle group division of the entire area, verify the vehicle elements of different sub-vehicle groups, and merge them into the same vehicle group when two adjacent vehicle groups have the same vehicles. S402: The vehicle group integration process specifically involves, when the vehicle group... car group The two groups of vehicles overlap, that is... This merges subgroups of shared vehicles into a single complete vehicle group. .

[0014] According to the above scheme, the specific process of constructing the evaluation index system in step S5 is as follows: S501. To achieve quantitative judgment of indicators and efficient identification of key vehicles, a key vehicle evaluation indicator system is constructed to identify key vehicles with adverse behaviors. This system includes basic behavioral indicators and safe distance indicators. Meeting at least one of these indicators is sufficient to determine whether a vehicle is a key vehicle. The comprehensive judgment formula for key vehicles is as follows:

[0015] Where K represents the key vehicle identification result, 1 indicates that it is determined to be a key vehicle, and 0 indicates that it is not a key vehicle; The result is a comprehensive assessment of basic behavioral indicators. 1 indicates that at least one basic behavioral indicator is met, and 0 indicates that no basic behavioral indicator is met. The result of the safety distance indicator is 1, which indicates that the safety distance indicator is met, and 0 indicates that the safety distance indicator is not met. S502, Basic Behavioral Indicators include: The absolute value of the deviation between the driving speed and the average speed of the group of vehicles exceeds 30%.

[0016] in, For the first The vehicle's real-time speed; The average speed of the vehicle group; The absolute value of acceleration exceeds 2 m / s² for 3 consecutive seconds.

[0017] in, For the first The car at any time Instantaneous acceleration; The start time of the time window; the time window is... That is, a continuous period of 3 seconds; Changing lanes more than twice in one minute

[0018] in, For the first The car in the time window Number of lane changes within; This is the start time of the time window; Driving at a speed below the minimum speed limit in the overtaking lane of a highway or urban expressway.

[0019]

[0020] in, This indicates the lane type the vehicle is currently traveling in; here, it is limited to the "overtaking lane". For the first The speed of the vehicle in the overtaking lane; The minimum speed limit for overtaking lanes on highways or urban expressways; S503, Safety Distance Indicators The safe following distance is defined as: the distance between the vehicle in front and behind is less than the safe following distance, where the safe following distance = driving speed × reaction time + braking distance. The formula for quantifying the safe following distance is as follows:

[0021] in, The theoretical safe following distance to the target vehicle. The real-time speed of the target vehicle. The average reaction time of the driver. This refers to the vehicle's braking acceleration.

[0022] According to the above scheme, the specific process of identifying key vehicles in step S503 is as follows: The first step is to calculate the theoretical safe following distance of the target vehicle using the safe following distance quantification formula. The actual distance between the target vehicle and the vehicles in front and behind is collected by roadside sensing devices, and the actual distance is compared with the theoretical safe distance to determine the safe distance index. The second step involves quantifying each of the four basic behavioral sub-indicators, including: ① Determination of driving speed deviation: calculating the target vehicle speed. The following criteria are used to determine the success of a vehicle's performance: ① Determination of the absolute value of the deviation from the average speed of the vehicle group. If it exceeds 30%, the performance is considered satisfactory. ② Continuous acceleration anomaly determination: Real-time acceleration data of the target vehicle is collected over 3 consecutive seconds. If the absolute value of acceleration exceeds 2 m / s² at any moment, the performance is considered satisfactory. ③ Frequent lane change determination: The number of consecutive lane changes by the target vehicle within 1 minute is counted. If it exceeds 2 times, the performance is considered satisfactory. ④ Low speed determination in the overtaking lane: It is determined whether the target vehicle's current lane is the overtaking lane of a highway / urban expressway. If it is in the overtaking lane and its speed is lower than the minimum speed limit of the corresponding road, the performance is considered satisfactory. If any one of the four sub-indicators is satisfied, the overall result of the basic behavioral indicators is deemed satisfactory and a value is assigned. =1, if none of the four conditions are met, then assign a value. =0; The third step is the final determination of the key vehicles: This involves using the results from the second step... And the first step Substitute the key vehicle comprehensive judgment formula and output the result through logical operation.

[0023] According to the above scheme, in step S6, the navigation binding status of the key vehicle is verified and its behavior is monitored and corrected. The specific process is as follows: S601. Extract the license plate of the key vehicle and upload it to the traffic management or 114 directory assistance via a dedicated interface to obtain the registered driver's phone number. Match this phone number with the phone number bound to the user on the mainstream navigation platform to verify the navigation binding status. S602. Directional reminders via navigation platform adopt a combination of "voice broadcast + pop-up prompts", and the content includes descriptions of risky behaviors, traffic rules for the road section and safety suggestions; S603 If the risky behavior is not corrected within 10 seconds after the reminder, the enhanced reminder mechanism will be activated, and then an enhanced reminder will be executed every 5 seconds, and the vehicle information will be synchronized to the roadside guidance screen and the traffic management department.

[0024] The present invention also provides a system for identifying key vehicles based on roadside group trajectories and intervening through navigation, comprising: The information acquisition module is used to acquire real-time traffic information and vehicle motion parameters of road segments through radar trajectory data; The vehicle impact assessment module is used to identify the vehicle group area, divide the scenario according to the lane and vehicle type, set the interaction impact distance threshold between vehicles, and assess the impact of surrounding vehicles. The vehicle group type acquisition module obtains different vehicle group types through threshold verification and according to the vehicle group recognition model process. The vehicle group classification module is used to classify and classify vehicles across the entire road area, verify the vehicle composition of each vehicle group, and perform vehicle group fusion. The key vehicle identification module is used to construct an evaluation index system, compare the deviation of vehicle motion parameters within the vehicle group with the overall motion parameters, and identify key vehicles in combination with preset thresholds. The critical vehicle navigation binding status module is used to verify the navigation binding status of critical vehicles and monitor and correct their behavior through navigation direction reminders and feedback.

[0025] The method of identifying key vehicles based on roadside group trajectories and intervening through navigation, as described in this invention, has the following beneficial effects: 1. This invention takes the perspective of a vehicle group as its core, collects data with the help of roadside sensing devices, and realizes vehicle group identification through model algorithms. Based on quantitative indicators, it identifies key vehicles that affect the safety or efficiency of vehicle group operation. Then, through navigation and directional reminders and hierarchical feedback, a control closed loop is formed. This not only breaks through the limitations of traditional single-point monitoring, but also improves the accuracy of identification and the timeliness of intervention. It is adaptable to different lane scenarios and can improve road traffic safety and efficiency at the same time. 2. This invention, through specific technical measures, overcomes the limitations of traditional monitoring. Focusing on the perspective of vehicle groups, it utilizes roadside sensing devices to collect data and employs model algorithms to achieve vehicle group identification, effectively meeting the core needs of road safety management and traffic efficiency optimization. It can identify key vehicles causing interference based on quantitative indicators, and then form a closed-loop management system through navigation-oriented alerts and tiered feedback. This not only overcomes the limitations of traditional single-point monitoring but also improves identification accuracy and intervention timeliness, adapting to different lane scenarios and simultaneously enhancing road traffic safety and efficiency. Attached Figure Description

[0026] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart of the method for identifying key vehicles based on roadside group trajectories and intervening through navigation, as per the present invention; Figure 2 This is a schematic diagram of the structure of the first embodiment of the method for identifying key vehicles based on roadside group trajectories and intervening through navigation according to the present invention; Figure 3 This is a schematic diagram of the five vehicle group types of the present invention. Detailed Implementation

[0027] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0028] like Figure 1-3 As shown, the method of identifying key vehicles based on roadside group trajectories and intervening through navigation according to the present invention includes the following steps: S1. In a three-lane driving environment, any single vehicle used for observation and analysis is designated as the target vehicle. Real-time traffic information and motion parameters of each vehicle are obtained through radar trajectory data. The specific process is as follows: S101, the roadside sensing equipment is a combination of millimeter-wave radar and high-definition video acquisition equipment that can recognize license plates. There is no limit to the spacing between the radar and the high-definition camera. The spacing should be adjusted according to the actual situation. Each set of equipment covers three lanes in one direction, and the data acquisition frequency is not less than 10Hz. S102. Real-time acquisition of the movement trajectories of all vehicles within the monitoring area. In a three-lane driving environment, any single vehicle used for observation and analysis will be designated as the target vehicle. The equipment must cover seven key targets surrounding the target vehicle. Local coordinates of each vehicle will be obtained. ), driving speed ( ), acceleration ( ), driving direction and license plate information, to form raw motion trajectory data; S103. Calculate the longitudinal distance between the target vehicle and the vehicle in the same lane directly in front based on the local coordinates. ), and the straight-line distance from vehicles in the adjacent lanes on the left and right along the direction of travel ( Then, based on the obtained longitudinal distance, calculate the headway between the target vehicle and the vehicle directly in front. The formula for calculating the headway is as follows:

[0029] S2. Define the vehicle group area, divide the scene according to the lane, and set the distance threshold for the interaction between vehicles to determine the impact of surrounding vehicles. The specific process is as follows: S201, The specific system for the distance threshold for vehicle-to-vehicle interaction is as follows: ① The vertical influence threshold is or headway <4s, ②The combined lateral and longitudinal influence thresholds for the left and right lanes are respectively ,

[0030] Negative distance values ​​indicate that the vehicle is behind the target vehicle, while positive values ​​indicate that it is in front of the target vehicle. Based on the JTG B05-2015 requirements for parking sight distance verification of medium and large vehicles and industry test data, the longitudinal interaction impact threshold for large vehicles is set to be twice that of small vehicles. Therefore, the specific threshold system for interaction impact distance between large vehicles is as follows: , , It includes a 10% safety redundancy.

[0031] S202. Implement differentiated interference vehicle monitoring strategies for different lane scenarios: When the target vehicle is in the left or right lane, focus on monitoring vehicles in the adjacent inner lane and the lane directly in front; when it is in the middle lane, simultaneously monitor vehicles in the adjacent lanes on both sides and the lane directly in front; when there are multiple interference vehicles on the same side, retain the closest vehicle within the range of influence as the core interference vehicle according to the principle of "minimum distance priority".

[0032] S3. Through threshold verification, different vehicle group types are obtained according to the vehicle group recognition model process. The specific process is as follows: S301, based on roadside group trajectories and preset thresholds, the vehicle group recognition model process can ultimately obtain five basic vehicle group types.

[0033] S302. Determine the lane type of the target vehicle. Based on the lane position attribute of the target vehicle (outer lane or middle lane), collect the longitudinal distance between the target vehicle and the vehicle in front and the lateral distance between vehicles in the left and right lanes to complete the threshold verification. S303. To accurately distinguish between the different vehicle group types corresponding to the left and right lanes in the outer lanes, different correction values ​​are assigned to avoid confusion of vehicle group types caused by the same combination of basic parameters within the outer lanes, ensuring the integrity and uniqueness of the mapping of the sixteen vehicle group types. This leads to the introduction of lane differentiation correction coefficients. The calculation formula is as follows:

[0034] in, For lane position identification, 1 indicates the outer lane, including the left and right lanes; 2 indicates the middle lane, used to clarify the basic lane attributes of the target vehicle; The lateral position coordinates of the target vehicle in the lane are used to accurately locate the target vehicle's specific position in the lateral direction of the road. It serves as the lateral coordinate of the road's transverse centerline, acting as the dividing line between the left and right lanes in the outer lanes; S304. Taking a car as an example, by integrating the above threshold checks and parameters, a unified multi-dimensional comprehensive mapping formula is constructed to fully cover various typical scenarios of target vehicles driving in the left lane, right lane (outer lane), and middle lane without constraints, with single-direction constraints and multi-direction joint constraints. Following the progressive order of lane determination, then longitudinal distance verification, and then lateral distance verification, and combining the verification results of each step, sixteen initial vehicle group types are identified: four for the left lane, four for the right lane, and eight for the middle lane. The calculation formula is as follows:

[0035] in, : Initial vehicle group type identifier, with a value of 1-16, uniquely corresponding to 16 simplified vehicle group types, covering all typical vehicle group states of the target vehicle under different lanes and different combinations of surrounding vehicle influences; This is a vehicle-ahead impact indicator; 1 indicates that the distance between the vehicle ahead and the target vehicle meets the requirement. This imposes a driving constraint on the target vehicle; 0 indicates the distance between the vehicle ahead and the target vehicle. Unconstrained effect; This is a vehicle impact indicator for the left lane; 1 indicates that the distance between the vehicle in the left lane and the target vehicle meets the requirement. This has a driving constraint effect on the target vehicle; 0 indicates that the distance between the vehicle in the left lane and the target vehicle exceeds this range, and there is no constraint effect. This is a vehicle impact indicator for the right lane; 1 indicates that the distance between the vehicle in the right lane and the target vehicle meets the requirement. This has a driving constraint effect on the target vehicle; 0 indicates that the distance between the vehicle in the right lane and the target vehicle exceeds this range, and there is no constraint effect. S305. After obtaining sixteen basic vehicle group types, the core dimensions are further divided according to whether there is interference from vehicles in front or in the direction of lateral interference. The influence of the target vehicle's lane is weakened, and minor positional differences and redundancy under similar interference are eliminated. While retaining key features, the complexity of recognition and calculation is reduced. The vehicles are then merged to adapt to actual road conditions, resulting in five vehicle group types: Uninterrupted free-moving vehicle groups, i.e., vehicles without interference from vehicles in front, behind, left, or right, satisfying... , The group of vehicles is only affected by interference from the vehicle in front, and only the vehicle in front meets the requirements. No interference from left or right; interference from a single lane affects only the left or right lane. The vehicle in front and the combined interference from vehicles in one lane on one side of the road, the vehicles in front meet the requirements. Furthermore, there is interference in a single lane, either to the left or right; or there is omnidirectional compound interference, meaning there is interference in the lanes in front and on both sides, or interference from multiple directions is superimposed.

[0036] S4. Complete the classification and division of vehicles across the entire road area, verify the vehicle composition of each vehicle group, and merge the vehicle groups. The specific process is as follows: S401. Based on the five basic vehicle group types, classify all vehicles traveling on the road one by one to complete the vehicle group division of the entire area. Then, verify the vehicle elements of different sub-vehicle groups. When two adjacent vehicle groups have the same vehicles, merge them into the same vehicle group.

[0037] S402, The vehicle group integration process specifically involves, when the vehicle group... car group The two groups of vehicles overlap. Then these subgroups of shared vehicles will be merged into a single complete group. .

[0038] S5. Construct an evaluation index system, compare the deviations of vehicle motion parameters within the vehicle group with the overall motion parameters, and identify key vehicles based on preset thresholds. The specific process is as follows: S501. To achieve quantitative judgment of indicators and efficient identification of key vehicles, a key vehicle evaluation indicator system is constructed to identify key vehicles with adverse behaviors. This indicator system includes basic behavioral indicators and safe distance indicators. Meeting at least one of these indicators is sufficient to determine whether a vehicle is a key vehicle. The comprehensive judgment formula for key vehicles is as follows:

[0039] Where K represents the key vehicle identification result, 1 indicates that it is determined to be a key vehicle, and 0 indicates that it is not a key vehicle; The result is a comprehensive assessment of basic behavioral indicators. 1 indicates that at least one basic behavioral indicator is met, and 0 indicates that no basic behavioral indicator is met. The result of the safety distance indicator is 1, which indicates that the safety distance indicator is met, and 0 indicates that the safety distance indicator is not met. S502, the aforementioned basic behavioral indicators include: The absolute value of the deviation between the driving speed and the average speed of the group of vehicles exceeds 30%.

[0040]

[0041] in, For the first The vehicle's real-time speed; The average speed of the vehicle group; the absolute value of acceleration exceeds 2 m / s² for 3 consecutive seconds;

[0042] in, For the first The car at any time Instantaneous acceleration; The start time of the time window; the time window is... That is, a continuous period of 3 seconds.

[0043] Changing lanes more than twice in one minute.

[0044]

[0045] in, For the first The car in the time window Number of lane changes within 1 minute; This represents the start time of the time window.

[0046] Driving at a speed below the minimum speed limit in the overtaking lane of a highway or urban expressway.

[0047]

[0048] in, This indicates the lane type the vehicle is currently traveling in; here, it is limited to the "overtaking lane". For the first The speed of the vehicle in the overtaking lane; The minimum speed limit for overtaking lanes on highways or urban expressways.

[0049] S503, the aforementioned safety distance indicator The safe following distance is defined as: the distance between the vehicle in front and behind is less than the safe following distance, where the safe following distance = driving speed × reaction time + braking distance. The formula for quantifying the safe following distance is as follows:

[0050] in The theoretical safe following distance to the target vehicle, in meters, is the minimum safe distance to ensure that the vehicle does not collide with the vehicle in front or behind during emergency braking. The real-time speed of the target vehicle, in m / s, needs to be accurately collected by roadside sensing devices; The average driver reaction time is measured in seconds. The preset value range is 0.5-1.5 seconds, taking into account the characteristics of the road scenario. The upper limit can be 1.2-1.5 seconds for highway scenarios and 0.8-1.2 seconds for urban expressways. This is the vehicle braking acceleration, measured in m / s², and is a positive value. The preset range for this value is 5-8 m / s², based on the typical braking performance of vehicles. It can be finely adjusted depending on the vehicle type, such as small cars or large cars.

[0051] Based on the above formulas and indicator system, the complete logic of key vehicle identification can be broken down into three steps, as follows: The first step is to determine and obtain safe distance indicators. First, the theoretical safe following distance of the target vehicle is calculated using the safe following distance quantification formula. Next, roadside sensing devices collect the actual distances between the target vehicle and the vehicles in front and behind. These actual distances are compared to the theoretical safe following distance. If the actual distance is less than the theoretical safe following distance, the safe distance between vehicles is too small, posing a driving risk. If the safe distance requirement is met, a value is assigned. =1, otherwise assign a value. =0.

[0052] The second step involves obtaining comprehensive assessments of basic behavioral indicators. The four basic behavioral sub-indicators are quantified one by one, including: ① Determination of driving speed deviation: Calculation of the target vehicle speed. The following four sub-indicators are considered satisfied: ① **Determination of absolute deviation from the average speed of the vehicle group:** If the percentage exceeds 30%, the sub-indicator is satisfied; ② **Continuous acceleration anomaly judgment:** Collect real-time acceleration data of the target vehicle over 3 consecutive seconds. If the absolute value of acceleration exceeds 2 m / s² at any moment, the sub-indicator is satisfied; ③ **Frequent lane change judgment:** Count the number of consecutive lane changes by the target vehicle within 1 minute. If it exceeds 2 times, the sub-indicator is satisfied; ④ **Low speed judgment in overtaking lane:** Identify whether the target vehicle's current lane is the overtaking lane of a highway / urban expressway. If it is in the overtaking lane and its speed is lower than the minimum speed limit of the corresponding road, the sub-indicator is satisfied; If any one of the four sub-indicators is satisfied, the overall result of the basic behavioral indicators is considered satisfied and a value is assigned. =1, if none of the four conditions are met, then assign a value. =0.

[0053] The third step is the final determination of the key vehicles: This involves using the results from the second step... And the first step Substitute the key vehicle comprehensive judgment formula, output the result through logical operation. If K=1, the vehicle is judged as a key vehicle. If K=0, it is judged as a non-key vehicle. Finally, the identification of key vehicles with interference is completed.

[0054] S6. Verify the navigation binding status of key vehicles and monitor and correct their behavior through navigation orientation reminders and feedback. The specific process is as follows: S601. Extract the license plate of the key vehicle and upload it to the traffic management or 114 directory assistance through a dedicated interface to obtain the registered driver's phone number. Compare and match this phone number with the personal phone number bound to the user during registration on mainstream navigation platforms such as Gaode Map and Baidu Map to verify the navigation binding status. S602. The system uses a "voice broadcast + pop-up prompt" approach to provide targeted reminders via navigation or telephone. The reminders include specific risky behaviors of the vehicle, such as speed deviation, sudden acceleration or deceleration, frequent lane changes, and low-speed lane occupation, as well as the traffic rules for the road segment, such as speed limits and lane usage regulations. Finally, it provides specific and feasible safe driving suggestions. S603 If the risky behavior is not corrected within 10 seconds of the warning, the system will activate an enhanced warning mechanism, issuing a warning every 5 seconds to continuously urge the driver to rectify the situation. Simultaneously, the system will upload the vehicle's license plate number, current location, and the risky behavior to the roadside guidance screens on both sides of the road, allowing nearby drivers to be aware and avoid the area. This information will also be simultaneously fed back to the traffic management department, providing data support for their subsequent monitoring and handling.

[0055] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for identifying key vehicles based on roadside group trajectories and intervening via navigation, characterized in that, Includes the following steps: S1. Obtain real-time traffic information and vehicle motion parameters of road sections through radar trajectory data; S2. Define the vehicle group area, divide the scenario according to the lane and vehicle type, set the distance threshold for the interaction between vehicles, and judge the impact of surrounding vehicles. S3. Through threshold verification, different vehicle group types are obtained according to the vehicle group identification model process; S4. Complete the classification and division of vehicles in the entire road area, verify the vehicle composition of each vehicle group and merge the vehicle groups. S5. Construct an evaluation index system, compare the deviation of vehicle motion parameters within the vehicle group with the overall motion parameters, and identify key vehicles by combining preset thresholds. S6. Verify the navigation binding status of key vehicles and monitor and correct their behavior through navigation orientation reminders and feedback.

2. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 1, characterized in that, In step S1, the specific process of obtaining real-time traffic information of the road segment and motion parameters of each vehicle is as follows: S101. A combination of millimeter-wave radar and video capture equipment capable of recognizing license plates is used to obtain real-time traffic information and vehicle motion parameters of road sections. The equipment covers three lanes in one direction and the data acquisition frequency is not less than 10Hz. S102. Real-time acquisition of the movement trajectories of all vehicles within the monitoring area, designating any single vehicle used for observation and analysis in a three-lane driving environment as the target vehicle, and obtaining the vehicle's local coordinates. ), driving speed acceleration The vehicle's direction of travel and license plate information are used to generate raw motion trajectory data. S103. Calculate the longitudinal distance between the target vehicle and the vehicle in the same lane directly in front, based on local coordinates. The straight-line distance from vehicles in the adjacent lanes on the left and right along the direction of travel ( ), calculate the headway between the target vehicle and the vehicle directly in front based on the obtained longitudinal distance. The formula for calculating the headway is as follows: 。 3. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 1, characterized in that, In step S2, the specific process of setting a threshold for the interaction and influence distance between vehicles and determining the influence of surrounding vehicles is as follows: S201. Determine the distance threshold system for inter-vehicle interaction effects: The vertical influence threshold is or headway <4s, the combined lateral and longitudinal impact thresholds for the left and right lanes are respectively , Negative distance values ​​indicate that the vehicle is in the area behind the target vehicle, while positive values ​​indicate that it is in the area in front of the target vehicle. The longitudinal interaction impact threshold for large vehicles is set to be twice that for small vehicles. The interaction impact distance threshold system for large vehicles is as follows: 、 、 ； S202. When the target vehicle is in the left or right lane, focus on monitoring vehicles in the adjacent inner lane and the lane directly in front; when it is in the middle lane, monitor vehicles in the adjacent lanes on both sides and the lane directly in front simultaneously; when there are multiple interfering vehicles on the same side, retain the closest vehicle within the range of influence as the core interfering vehicle according to the principle of "minimum distance priority".

4. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 1, characterized in that, In step S3, different vehicle group types are obtained according to the vehicle group recognition model process. The specific process is as follows: S301. Based on roadside group trajectories and preset thresholds, five basic types of vehicle groups are obtained through a vehicle group recognition model process. S302. Determine the lane type of the target vehicle. Based on the lane position attributes of the target vehicle, collect the longitudinal distance between the target vehicle and the vehicle in front and the lateral distance between the vehicles in the left and right lanes to complete the threshold verification. S303. To distinguish between the different vehicle group types corresponding to the left and right lanes in the outer lanes, different correction values ​​are assigned to avoid confusion of vehicle group types caused by the same combination of basic parameters within the outer lanes, leading to the introduction of lane differentiation correction coefficients. The calculation formula is as follows: in, Lane position indicators: 1 indicates the outer lane, including the left and right lanes; 2 indicates the middle lane. Let be the lateral position coordinates of the target vehicle within the lane. The lateral coordinates of the road's transverse centerline; S304. By integrating threshold tests and parameters, a comprehensive mapping formula is constructed to fully cover various typical scenarios of target vehicles in the left lane, right lane, and middle lane, achieving initial vehicle group type identification. The calculation formula is as follows: in, This serves as the initial vehicle group type identifier; Warning sign indicating an impact on vehicles ahead. This is a sign indicating the impact on vehicles in the left lane. This is an indication of the impact on vehicles in the right lane; S305. After obtaining the initial vehicle group, the core dimensions are divided according to the direction of interference from the vehicle in front and the lateral interference. The influence of the lane is weakened, the positional difference redundancy under the same interference is eliminated, the key features are retained and the complexity is reduced. After being adapted to the actual road conditions, five basic vehicle group types are formed.

5. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 4, characterized in that, In step S305, the five basic vehicle group types are: a vehicle group driving freely without interference, a vehicle group only affected by the vehicle in front, a vehicle group affected by interference on one side of the lane, a vehicle group affected by a combination of interference from the vehicle in front and one side of the lane, and a vehicle group affected by a combination of interference from all directions.

6. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 1, characterized in that, In step S4, the vehicle group classification of all vehicles in the road area is completed, the vehicle composition of each group is verified, and the vehicle groups are merged. The specific process is as follows: S401: Based on the five vehicle group types, classify all vehicles traveling on the road one by one, complete the vehicle group division of the entire area, verify the vehicle elements of different sub-vehicle groups, and merge them into the same vehicle group when two adjacent vehicle groups have the same vehicles. S402: The vehicle group integration process specifically involves, when the vehicle group... car group The two groups of vehicles overlap, that is... This merges subgroups of shared vehicles into a single complete vehicle group. .

7. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 1, characterized in that, In step S5, the specific process of constructing the evaluation index system is as follows: S501. To achieve quantitative judgment of indicators and efficient identification of key vehicles, a key vehicle evaluation indicator system is constructed to identify key vehicles with adverse behaviors. This system includes basic behavioral indicators and safe distance indicators. Meeting at least one of these indicators is sufficient to determine whether a vehicle is a key vehicle. The comprehensive judgment formula for key vehicles is as follows: Where K represents the key vehicle identification result, 1 indicates that it is determined to be a key vehicle, and 0 indicates that it is not a key vehicle; The result is a comprehensive assessment of basic behavioral indicators. 1 indicates that at least one basic behavioral indicator is met, and 0 indicates that no basic behavioral indicator is met. The result of the safety distance indicator is 1, which indicates that the safety distance indicator is met, and 0 indicates that the safety distance indicator is not met. S502, Basic Behavioral Indicators include: The absolute value of the deviation between the driving speed and the average speed of the group of vehicles exceeds 30%. in, For the first The vehicle's real-time speed; The average speed of the vehicle group; The absolute value of acceleration exceeds 2 m / s² for 3 consecutive seconds. in, For the first The car at any time Instantaneous acceleration; The start time of the time window; the time window is... That is, a continuous period of 3 seconds; Changing lanes more than twice in one minute in, For the first The car in the time window Number of lane changes within; This is the start time of the time window; Driving below the minimum speed limit in the overtaking lane of a highway or urban expressway in, This indicates the lane type the vehicle is currently traveling in; here, it is limited to the "overtaking lane". For the first The speed of the vehicle in the overtaking lane; The minimum speed limit for overtaking lanes on highways or urban expressways; S503, Safety Distance Indicators The safe following distance is defined as: the distance between the vehicle in front and behind is less than the safe following distance, where the safe following distance = driving speed × reaction time + braking distance. The formula for quantifying the safe following distance is as follows: in, The theoretical safe following distance to the target vehicle. The real-time speed of the target vehicle. The average reaction time of the driver. This refers to the vehicle's braking acceleration.

8. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 7, characterized in that, In step S503, the specific process for identifying the key vehicle is as follows: The first step is to calculate the theoretical safe following distance of the target vehicle using the safe following distance quantification formula. ; The actual distance between the target vehicle and the vehicles in front and behind is collected by roadside sensing devices. The actual distance is compared with the theoretical safe distance to determine the safe distance index. The second step involves quantifying each of the four basic behavioral sub-indicators, including: ① Determination of driving speed deviation: calculating the target vehicle speed. The following criteria are used to determine the success of a vehicle's performance: ① Determination of the absolute value of the deviation from the average speed of the vehicle group. If it exceeds 30%, the performance is considered satisfactory. ② Continuous acceleration anomaly determination: Real-time acceleration data of the target vehicle is collected over 3 consecutive seconds. If the absolute value of acceleration exceeds 2 m / s² at any moment, the performance is considered satisfactory. ③ Frequent lane change determination: The number of consecutive lane changes by the target vehicle within 1 minute is counted. If it exceeds 2 times, the performance is considered satisfactory. ④ Low speed determination in the overtaking lane: It is determined whether the target vehicle's current lane is the overtaking lane of a highway / urban expressway. If it is in the overtaking lane and its speed is lower than the minimum speed limit of the corresponding road, the performance is considered satisfactory. If any one of the four sub-indicators is satisfied, the overall result of the basic behavioral indicators is deemed satisfactory and a value is assigned. =1, if none of the four conditions are met, then assign a value. =0; The third step is the final determination of the key vehicles: This involves using the results from the second step... And the first step Substitute the key vehicle comprehensive judgment formula and output the result through logical operation.

9. The method for identifying key vehicles based on roadside group trajectories and intervening via navigation according to claim 1, characterized in that, In step S6, the navigation binding status of key vehicles is verified and their behavior is monitored and corrected. The specific process is as follows: S601. Extract the license plate of the key vehicle and upload it to the traffic management or 114 directory assistance via a dedicated interface to obtain the registered driver's phone number. Match this phone number with the phone number bound to the user on the mainstream navigation platform to verify the navigation binding status. S602. Directional reminders via navigation platform adopt a combination of "voice broadcast + pop-up prompts", and the content includes descriptions of risky behaviors, traffic rules for the road section and safety suggestions; S603 If the risky behavior is not corrected within 10 seconds after the reminder, the enhanced reminder mechanism will be activated, and then an enhanced reminder will be executed every 5 seconds, and the vehicle information will be synchronized to the roadside guidance screen and the traffic management department.

10. A system for identifying key vehicles based on roadside group trajectories and intervening through navigation, characterized in that, include: The information acquisition module is used to acquire real-time traffic information and vehicle motion parameters of road segments through radar trajectory data; The vehicle impact assessment module is used to identify the vehicle group area, divide the scenario according to the lane and vehicle type, set the interaction impact distance threshold between vehicles, and assess the impact of surrounding vehicles. The vehicle group type acquisition module obtains different vehicle group types through threshold verification and according to the vehicle group recognition model process. The vehicle group classification module is used to classify and classify vehicles across the entire road area, verify the vehicle composition of each vehicle group, and perform vehicle group fusion. The key vehicle identification module is used to construct an evaluation index system, compare the deviation of vehicle motion parameters within the vehicle group with the overall motion parameters, and identify key vehicles in combination with preset thresholds. The critical vehicle navigation binding status module is used to verify the navigation binding status of critical vehicles and monitor and correct their behavior through navigation direction reminders and feedback.

Images