Dynamic risk assessment and syndrome early warning method and device for tunnel exit section

Abstract

The invention relates to a dynamic risk assessment and syndrome early warning method and device for a tunnel exit section, and belongs to the technical field of tunnel safety. According to the method, vehicle behavior data are collected, a rear-end collision risk assessment model of an exit section is constructed, the dynamic risk level is output, and the early warning area and the warning mode of the lamp strip are dynamically adjusted according to the dynamic risk level and the vehicle driving track, so that accompanying early warning of vehicle driving lamp accompanying is realized. According to the method, the potential operation risk of the accident high-frequency road section can be measured and controlled, the operation risk of the tunnel exit road section is effectively reduced, the risk active prevention and control level of the tunnel exit road section is improved, and the driving safety level of the tunnel exit road section is improved.

Description

Technical Field

[0001] This invention belongs to the field of tunnel safety technology, specifically relating to a dynamic risk assessment and accompanying early warning method and equipment for tunnel exit sections. Background Technology

[0002] Tunnels are key nodes in road traffic networks. The tunnel exit section, as the transition interface between enclosed space and open road area, has a significant impact on driving behavior due to the dynamic changes in light environment, driving load, and vehicle operating status, making it a high-incidence area for traffic accidents. Among these accidents, rear-end collisions account for the highest proportion.

[0003] To reduce traffic accident rates and casualties, numerous active and passive technologies for improving road conditions and traffic safety management methods have been applied to tunnel sections. These include active safety measures such as installing traffic warning signs at tunnel entrances, visual modifications to tunnel entrances and exits, lighting design, and the application of special road markings. However, driving conditions in tunnel sections are affected by many factors, and drivers' inability to effectively and comprehensively perceive driving conditions, traffic conditions, and driving risks is a significant factor contributing to the high incidence of traffic accidents. Existing static control measures cannot monitor traffic information at tunnel exits in real time, and drivers cannot react correctly to the operational risks at tunnel exits, resulting in poor early warning and control effectiveness.

[0004] Therefore, in order to proactively guide vehicles at tunnel exit sections and achieve intelligent management of traffic risks that are measurable, controllable, and collaborative, there is an urgent need to propose an economical, scientific, and reasonable method for risk detection and targeted early warning. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies and provide a method and device for dynamic risk assessment and accompanying early warning of tunnel exit sections.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A dynamic risk assessment and accompanying early warning method for tunnel exit sections includes the following steps: S1: Determine the early warning deployment range of the tunnel exit section. The tunnel exit section is the section 0 to L m upstream of the tunnel exit portal in the opposite direction of traffic direction. The exit portal is the boundary between the inside and outside of the tunnel. Where L ≤ 500 m. S2: Install integrated radar-visual equipment at the tunnel exit section to collect traffic behavior data at the tunnel exit section in real time; the traffic behavior data includes vehicle speed, vehicle spacing and vehicle type; S3: Based on real-time traffic behavior data, the real-time rear-end collision risk value of each vehicle passing through the exit section is calculated through a dynamic rear-end collision risk assessment model. S4: Based on the rear-end collision risk value, the dynamic risk level is divided into four levels: no warning, low risk, medium risk and high risk. The dynamic risk level is mapped to the preset smart light warning strategy. S5: Based on dynamic risk level and vehicle location data, control the intelligent warning lights to execute corresponding intelligent light warning strategies to achieve accompanying warnings with the vehicle lights.

[0007] Furthermore, preferably, in step S1: the L is determined based on the distribution of historical accident locations and / or the tunnel length, specifically: provided that L does not exceed 500 m, the L is taken as the minimum length covering the accident concentration area upstream of the exit portal; when historical accident data is insufficient or there is no obvious concentration area of ​​accident locations, the L is taken as a preset default length, and the preset default length is 200 m.

[0008] Furthermore, preferably, vehicle type, vehicle speed, and vehicle spacing data are collected using a radar-guided integrated camera, with the detection range covering the tunnel exit section.

[0009] Furthermore, preferably, in step S2: there are multiple integrated radar-visual units, which are evenly distributed along the tunnel axis at the top of the tunnel interior, and the spacing between adjacent integrated radar-visual units is no greater than 1 / 2 of the maximum effective detection distance of the integrated radar-visual units, so as to ensure the continuity of vehicle detection and tracking in the exit section.

[0010] Furthermore, preferably, the specific method for conducting an operational risk assessment of accidents at the tunnel exit section in step S3 is as follows: S3.1: Conduct risk assessment based on vehicle driving status data of monitored exit road sections. When the distance between two vehicles is less than 100m, identify the types of adjacent vehicles and classify the following vehicle combinations into four situations: small car following small car (SS), small car following large car (SB), large car following large car (BB), and large car following small car (BS). S3.2: Based on the four scenarios described above, input the vehicle type and subsequent vehicle speed. Speed ​​of the vehicle in front Distance between vehicles The rear-end collision risk value Y of the following vehicle is calculated by selecting the corresponding rear-end collision risk model, wherein the rear-end collision risk model includes: in, , , , These represent the risk values ​​for rear-end collisions when a small car follows another small car, a small car follows another large car, a large car follows another large car, and a large car follows another small car, respectively.

[0011] Further, preferably, in step S4, a first threshold value T1, a second threshold value T2, and a third threshold value T3 are set, and the dynamic risk level is divided according to the rear-end collision risk value and the corresponding warning level is triggered. Specifically: When Y < T1, it is determined as a non-warning state and no warning is required; When T1 ≤ Y < T2, it is determined as a low risk and a level I warning is triggered; When T2 ≤ Y < T3, it is determined as a medium risk and a level II warning is triggered When Y ≥ T3, it is determined as a high risk and a level III warning is triggered.

[0012] The first threshold value T1 = 1.986, the second threshold value T2 = 2.665, and the third threshold value T3 = 3.654.

[0013] Further, preferably, in step S4, the preset intelligent lamp warning strategy adopts a following-type lamp warning. Specifically: when in the non-warning state, the intelligent warning lamp is constantly on in yellow; when in level I, level II, and level III warnings, the intelligent warning lamp flashes in red, and the flashing frequencies are 30 times / min, 60 times / min, and 120 times / min respectively; The intelligent warning lamp device is installed on the inner side wall of the tunnel, arranged along the driving direction, and is arranged in a bilateral linkage manner; the installation height of the intelligent warning lamp device from the road surface is 1.2 m, and the longitudinal arrangement spacing is 25 m.

[0014] Further, preferably, the specific method of step S5 is: S5.1: Obtain the real-time position of the following vehicle through the radar-vision integrated machine. When it is detected that the rear-end collision risk value of the vehicle is greater than or equal to T1, activate 2 groups of lights downstream of the vehicle to form a light strip; S5.2: The light strip will be dynamically adjusted with the movement of the warning vehicle to ensure that the intelligent warning lamp always follows in front of the warning vehicle. The migration speed U of the light strip L = max( , 60 km / h); S5.3: When the detected risk value drops below T1, the intelligent warning lamp is adjusted to the non-warning state.

[0015] Further, preferably, in step S5, when there are multiple warning vehicles and the corresponding dynamic risk levels conflict, the intelligent lamp warning strategy corresponding to the higher dynamic risk level is adopted for control.

[0016] The present invention also provides a dynamic risk assessment and accompanying warning device for the tunnel exit section, adopting the dynamic risk assessment and accompanying warning method for the tunnel exit section, including a radar-vision integrated machine, a warning controller, and an intelligent warning lamp. The warning controller is respectively connected to the radar-vision integrated machine and the intelligent warning lamp; The radar-visual integrated machine is used to collect traffic behavior data at the tunnel exit section; the traffic behavior data includes vehicle speed, vehicle spacing, and vehicle type. The warning controller is used to calculate the real-time rear-end collision risk value of each vehicle passing through the exit section based on real-time traffic behavior data and a dynamic rear-end collision risk assessment model; based on the rear-end collision risk value, the corresponding dynamic risk level is obtained, and the corresponding intelligent light warning strategy is obtained. Then, based on the dynamic risk level and the location data of the following vehicle, the intelligent warning lights are controlled to execute the corresponding intelligent light warning strategy, so as to realize the accompanying warning of the vehicle's traffic lights.

[0017] In this invention, the accident concentration zone can be defined as: a continuous sub-segment in the upstream direction of the exit portal where the accident point density or accident frequency is significantly higher than that of other segments (e.g., a segment higher than the average level or exceeding a preset threshold).

[0018] In this invention, the rear-end collision risk value Y is a risk metric used to characterize the degree of rear-end collision risk. A higher value indicates a higher risk, and the value can be calibrated and updated based on actual data. The rear-end collision risk model is an empirical model calibrated from collected vehicle following data. Preferably, under the constraint of the statistical distribution of measured following data, following condition samples are generated using the Monte Carlo method, and regression fitting is used to determine the model coefficients for different vehicle combinations, thereby achieving rapid real-time calculation of the rear-end collision risk value Y. The model coefficients can be recalibrated and updated based on data from different tunnel scenarios.

[0019] In this invention, the threshold can be updated based on the traffic composition and sample data of different tunnels.

[0020] In this invention, the integrated radar-visual device is used to collect traffic behavior data in real time within the tunnel exit section. The traffic behavior data includes at least vehicle speed, vehicle spacing, and vehicle type. When multiple integrated radar-visual devices are set up, the data collected by each integrated radar-visual device is transmitted to the early warning controller to achieve multi-device data fusion and sharing.

[0021] In this invention, the early warning controller processes the traffic behavior data: it calculates the real-time rear-end collision risk value of the vehicle within the exit section based on a dynamic rear-end collision risk assessment model, determines the corresponding dynamic risk level based on the rear-end collision risk value, and maps the dynamic risk level to an intelligent light warning strategy; furthermore, the early warning controller, combined with the positioning data of the following vehicle, controls the intelligent warning lights to output light prompts according to the intelligent light warning strategy, forming a light strip that moves with the vehicle, realizing a accompanying warning with the vehicle's lights. The early warning controller can be installed in a roadside cabinet near the tunnel entrance, in a tunnel auxiliary equipment room, or in a monitoring room, and communicates with the radar-visual integrated machine and the intelligent warning lights via wired or wireless means.

[0022] This invention constructs a real-time rear-end collision risk assessment model by rationally deploying integrated radar and vision devices, intelligent warning lights, and other equipment. It provides dynamic information alerts for high-risk vehicles at tunnel exits, thereby reducing driving risks at tunnel exits and improving driving safety.

[0023] Compared with the prior art, the beneficial effects of this invention are as follows: (1) This invention calculates and dynamically classifies the risk of rear-end collision based on the real-time data of vehicle speed, vehicle spacing and vehicle type collected by the radar-visual integrated machine, so as to realize the real-time perception and timely warning of the operational risk of the exit section; (2) The present invention adopts an accompanying light strip warning. The warning light is dynamically adjusted with the position of the target vehicle and remains in front of the vehicle. Compared with fixed position or static prompting methods, it has the characteristics of intuitive prompting, strong directionality and little interference. (3) The parameters of the present invention are highly configurable. The exit section length L, risk threshold and light strip control strategy can be set or updated according to the tunnel traffic characteristics and accident distribution, which is convenient for engineering deployment and promotion. The effect of the present invention can be quantitatively evaluated by indicators such as the number of accidents / conflict events, the number of sudden deceleration events, the advance warning amount, and the false alarm rate / missed alarm rate. Attached Figure Description

[0024] Figure 1 This is a flowchart of the dynamic risk assessment and accompanying early warning method for the tunnel exit section of the present invention. Figure 2 This is a schematic diagram showing the deployment locations of the integrated radar-visual unit and the intelligent early warning light.

[0025] Figure 3 This is a flowchart illustrating the accompanying early warning method of the present invention. Detailed Implementation

[0026] The present invention will now be described in further detail with reference to the embodiments.

[0027] Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be construed as limiting the scope of the invention. Where specific techniques or conditions are not specified in the embodiments, they are performed in accordance with the techniques or conditions described in the literature in the field or according to the product instructions. Materials or equipment whose manufacturers are not specified are all conventional products that can be obtained by purchase.

[0028] Example 1 like Figure 1 As shown, a dynamic risk assessment and accompanying early warning method for tunnel exit sections includes the following steps: S1: Determine the early warning deployment range of the tunnel exit section. The tunnel exit section is the section 0 to L m upstream of the tunnel exit portal in the opposite direction of traffic direction. The exit portal is the boundary between the inside and outside of the tunnel. Where L ≤ 500 m. S2: Install integrated radar-visual equipment at the tunnel exit section to collect traffic behavior data at the tunnel exit section in real time; the traffic behavior data includes vehicle speed, vehicle spacing and vehicle type; S3: Based on real-time traffic behavior data, the real-time rear-end collision risk value of each vehicle passing through the exit section is calculated through a dynamic rear-end collision risk assessment model. S4: Based on the rear-end collision risk value, the dynamic risk level is divided into four levels: no warning, low risk, medium risk and high risk. The dynamic risk level is mapped to the preset smart light warning strategy. S5: Based on dynamic risk level and vehicle location data, control the intelligent warning lights to execute corresponding intelligent light warning strategies to achieve accompanying warnings with the vehicle lights.

[0029] Example 2 like Figure 1 As shown, a dynamic risk assessment and accompanying early warning method for tunnel exit sections includes the following steps: S1: Determine the early warning deployment range of the tunnel exit section. The tunnel exit section is the section 0 to L m upstream of the tunnel exit portal in the opposite direction of traffic direction. The exit portal is the boundary between the inside and outside of the tunnel. Where L ≤ 500 m. S2: Install integrated radar-visual equipment at the tunnel exit section to collect traffic behavior data at the tunnel exit section in real time; the traffic behavior data includes vehicle speed, vehicle spacing and vehicle type; S3: Based on real-time traffic behavior data, the real-time rear-end collision risk value of each vehicle passing through the exit section is calculated through a dynamic rear-end collision risk assessment model. S4: Based on the rear-end collision risk value, the dynamic risk level is divided into four levels: no warning, low risk, medium risk and high risk. The dynamic risk level is mapped to the preset smart light warning strategy. S5: Based on dynamic risk level and vehicle location data, control the intelligent warning lights to execute corresponding intelligent light warning strategies to achieve accompanying warnings with the vehicle lights.

[0030] In step S1: L is determined based on the distribution of historical accident locations and / or the tunnel length. Specifically, under the premise that L does not exceed 500 m, L is taken as the minimum length covering the accident concentration area upstream of the exit portal; when historical accident data is insufficient or there is no obvious concentration area of ​​accident locations, L is taken as a preset default length, which is 200 m.

[0031] Collect data on vehicle type, vehicle speed, and vehicle spacing through a combined radar and camera system, with the detection range covering the tunnel exit section.

[0032] In step S2: There are multiple combined radar and camera systems, evenly arranged along the axis of the tunnel at the top inside the tunnel, and the installation spacing between adjacent combined radar and camera systems is not greater than 1 / 2 of the maximum effective detection distance of the combined radar and camera system.

[0033] In step S3, the specific method for conducting an operational risk assessment of accidents in the tunnel exit section is as follows: S3.1: Conduct a risk assessment based on the vehicle driving state data monitored in the exit section. When the distance between two vehicles is less than 100 m, identify the types of adjacent vehicles, and divide the following vehicle combinations into four situations: small vehicle following small vehicle (SS), small vehicle following large vehicle (SB), large vehicle following large vehicle (BB), and large vehicle following small vehicle (BS); S3.2: According to the four situations, by inputting the vehicle type, the speed of the following vehicle , the speed of the leading vehicle , and the vehicle distance between the leading and following vehicles , select the corresponding rear-end collision risk model to calculate the rear-end collision risk value Y of the following vehicle. The rear-end collision risk models include: Among them, , , , respectively represent the rear-end collision risk values for small vehicle following small vehicle, small vehicle following large vehicle, large vehicle following large vehicle, and large vehicle following small vehicle.

[0034] In step S4, set the first threshold T1, the second threshold T2, and the third threshold T3, and divide the dynamic risk levels and trigger the corresponding warning levels according to the rear-end collision risk value. Specifically: When Y < T1, it is determined as a non-warning state, and no warning is required; When T1 ≤ Y < T2, it is determined as a low risk, and a level I warning is triggered; When T2 ≤ Y < T3, it is determined as a medium risk, and a level II warning is triggered When Y ≥ T3, it is determined as a high risk, and a level III warning is triggered.

[0035] The first threshold T1 = 1.986, the second threshold T2 = 2.665, and the third threshold T3 = 3.654.

[0036] In step S4, the preset intelligent light warning strategy adopts a following light warning, specifically: when there is no warning, the intelligent warning light is always yellow; when there are Level I, Level II and Level III warnings, the intelligent warning light flashes red, with flashing frequencies of 30 times / min, 60 times / min and 120 times / min respectively. The intelligent early warning light device is installed on the inner wall of the tunnel, along the direction of traffic, and is configured to be deployed on both sides in a coordinated manner; the installation height of the intelligent early warning light device from the road surface is 1.2m, and the longitudinal spacing is 25m.

[0037] The specific method for step S5 is as follows: S5.1: The real-time position of the following vehicle is obtained through the integrated radar vision device. When the rear-end collision risk value of the vehicle is detected to be greater than or equal to T1, the two sets of lights downstream of the vehicle are activated to form a light strip. S5.2: The light strip will dynamically adjust as the warning vehicle moves, ensuring that the intelligent warning light always follows in front of the warning vehicle, with a light strip migration speed U. L =max( (60km / h); S5.3: When the monitored risk value drops below T1, the intelligent warning light switches to a non-warning state; for example... Figure 3 As shown.

[0038] In step S5, when there are multiple warning vehicles and their corresponding dynamic risk levels conflict, the intelligent light warning strategy corresponding to the higher dynamic risk level is used for control.

[0039] A dynamic risk assessment and accompanying early warning device for tunnel exit sections, employing the aforementioned dynamic risk assessment and accompanying early warning method for tunnel exit sections, includes a radar-visual integrated machine, an early warning controller, and an intelligent early warning light, wherein the early warning controller is connected to the radar-visual integrated machine and the intelligent early warning light respectively. The radar-visual integrated machine is used to collect traffic behavior data at the tunnel exit section; the traffic behavior data includes vehicle speed, vehicle spacing, and vehicle type. The warning controller is used to calculate the real-time rear-end collision risk value of each vehicle passing through the exit section based on real-time traffic behavior data and a dynamic rear-end collision risk assessment model; based on the rear-end collision risk value, the corresponding dynamic risk level is obtained, and the corresponding intelligent light warning strategy is obtained. Then, based on the dynamic risk level and the location data of the following vehicle, the intelligent warning lights are controlled to execute the corresponding intelligent light warning strategy, so as to realize the accompanying warning of the vehicle's traffic lights.

[0040] Application Examples 1. Application Scenarios and Deployment Scope The dynamic risk assessment and accompanying early warning system of this invention is installed at the exit direction of a two-lane, one-way tunnel on a highway. The tunnel exit portal is the boundary section between the inside and outside of the tunnel. This embodiment only provides early warning for the exit section inside the tunnel.

[0041] According to step S1, the early warning deployment range of the tunnel exit section is determined as follows: starting from the exit portal, the exit section extends upstream in the opposite direction of traffic from 0 to L meters. When historical accident location data is insufficient or there is no obvious concentration area, the default length L = 200 meters is taken (and L ≤ 500 meters is satisfied). Therefore, the early warning deployment range in this embodiment is the section within the tunnel 0 to 200 meters upstream of the exit portal.

[0042] 2. Equipment Composition and Installation Method The early warning device in this embodiment includes a radar-visual integrated machine, an early warning controller, and an intelligent early warning light (light strip unit). Figure 2 This is a schematic diagram of the deployment of the integrated radar-visual machine and intelligent early warning lights.

[0043] like Figure 2 As shown, the radar-visual integrated machine is a data acquisition device for collecting vehicle driving status. Figure 2 This is a schematic diagram of the deployment of the integrated radar-visual unit and intelligent warning lights. The integrated radar-visual unit should be installed at the center of the tunnel ceiling. During installation, the unit needs to be adjusted, and the equipment needs to be tilted downwards to ensure coverage of the lane area. The integrated radar-visual unit used in this example has a maximum radar range of 200m and is installed at a height of 5m at the center of the tunnel ceiling. The elevation angle β = arctan(installation height / range) = arctan(5 / 200) ≈ 1.43° The deployment spacing of the radar-visual integrated unit is half the maximum effective range of the radar. Therefore, in this example, the deployment spacing is 100m. The data collected by each radar-visual integrated unit is transmitted to the early warning controller to achieve multi-point data fusion and sharing.

[0044] like Figure 2 As shown, the intelligent early warning light device at the tunnel exit is installed symmetrically on the left and right side walls in the direction of travel. Two lights on each cross section form a group, controlled in a linked manner to achieve synchronous activation and deactivation. The installation height from the road surface is 1.2m, the longitudinal spacing is 25m, and it is connected to the early warning controller for linked control and display on the vehicle's accompanying light strip.

[0045] The early warning controller can be installed in a roadside cabinet near the tunnel entrance, in the tunnel's auxiliary equipment room, or in the monitoring room, and can communicate with the integrated radar-visual unit and intelligent early warning lights via wired or wireless means.

[0046] 3. Identification of car-following risk and calculation of rear-end collision risk value like Figure 3As shown, according to step S2, the radar-vision integrated machine collects real-time traffic behavior data of vehicles in the exit section, including at least vehicle speed, vehicle spacing, and vehicle type, and outputs vehicle trajectories and positioning information.

[0047] According to step S3, a running risk assessment is performed on the vehicles in the exit section, specifically: S3.1 (Following identification and situation classification): When the distance between two adjacent vehicles is detected to be less than 100 m, the warning controller determines that the vehicle pair is in an effective following relationship, identifies the types of adjacent vehicles, and divides the following vehicle combinations into four situations: SS (small vehicle following small vehicle), SB (small vehicle following large vehicle), BB (large vehicle following large vehicle), and BS (large vehicle following small vehicle).

[0048] S3.2 (Risk value calculation): For the four situations, the warning controller inputs vehicle type, the speed of the following vehicle , the speed of the leading vehicle , and the distance between the leading and following vehicles , selects the rear-end collision risk model corresponding to the situation, and calculates the rear-end collision risk value Y of the following vehicle in real time. The Y is a risk metric indicating the degree of rear-end collision risk, and the larger the value, the higher the rear-end collision risk. The rear-end collision risk models for the four situations are as follows: To improve the model's feasibility and reproducibility, in this embodiment, the calibration process of "measured data - simulation calibration - regression fitting" is adopted for establishing the rear-end collision risk model: First, collect vehicle following operation data to obtain , , , and sample combinations such as vehicle types; secondly, generate representative following working conditions through the Monte Carlo method under the constraint of the statistical distribution of the measured samples, and assign risk level labels to the simulation working conditions according to the existing risk calculation and grading reference framework; finally, perform linear regression fitting for the four vehicle type combinations respectively to obtain four rear-end collision risk models. The model coefficients can be recalibrated and updated according to the traffic composition and sample data of different tunnels.

[0049] 4. Dynamic risk level and warning level triggering According to step S4, set the first threshold , the second threshold , and the third threshold , and divide the dynamic risk level according to the rear-end collision risk value Y and trigger the corresponding warning level: When Y < T1, it is determined as a non-warning state and no warning is required; When T1 ≤ Y < T2, it is determined as a low risk and a level I warning is triggered; When T2 ≤ Y < T3, it is determined as a medium risk and a level II warning is triggered When Y≥T3, it is judged as high risk and a Level III warning is triggered.

[0050] In this embodiment, take , , .

[0051] 5. Accompanying LED strip early warning strategy and LED strip migration According to step S5, the dynamic risk level is divided into three levels: low risk, medium risk and high risk, based on the rear-end collision risk value. The dynamic risk level is then mapped to the preset smart light warning strategy. The warning formats for different warning levels are shown in Table 1.

[0052] Table 1 When a rear-end collision risk value is detected for a following vehicle When the warning threshold of Level I or above is reached, the warning controller activates two sets of intelligent warning lights downstream of the vehicle to form a light strip. The light strip will dynamically adjust as the warning vehicle moves, ensuring that the intelligent warning lights always follow in front of the warning vehicle. The light strip migration speed U L =max( (60km / h); when Y falls back to In the following situations, the warning for the vehicle will be deactivated and the relevant light strip will be restored to the non-warning state. When multiple vehicles are under warning and the light strip control conflicts, the warning controller will prioritize the warning strategy corresponding to the higher dynamic risk level.

[0053] 6. Examples of specific applications Taking an exit section with a length of L=200m as an example, the exit section is defined as the section extending 0-200m upstream from the exit portal in the opposite direction of traffic flow into the tunnel. Intelligent warning lights are deployed along the traffic flow direction with a longitudinal spacing of 25m. Each light strip consists of two consecutive sets of intelligent warning lights in front of the vehicle being warned, with a corresponding light strip coverage length of 50m. The threshold values ​​are T1=1.986, T2=2.665, and T3=3.654.

[0054] At time t0, the warning controller detects a pair of adjacent vehicles forming a valid following relationship approximately 150m upstream of the exit gate: the following vehicle is a small car, the preceding vehicle is a large car (in case SB), and the distance between the two vehicles is less than 100m. The warning controller collects and inputs the speed V of the following vehicle. B (t0), speed of the vehicle in front V F(t0) and the front and rear vehicle distances (Ds(t0)) are used to calculate the rear-end collision risk value Y(t0). When T2 ≤ Y(t0) < T3, it is determined as medium risk and a level-II warning is triggered; the warning controller immediately controls two consecutive groups of intelligent warning lights in front of the driving direction of the following vehicle to light up to form a light strip, and outputs a warning signal in a red flashing mode, with a flashing frequency of 60 times / min.

[0055] Subsequently, the vehicle continues to travel towards the exit portal until time t1. When the following vehicle travels to about 120 m upstream of the exit portal, the warning controller recalculates and obtains the risk value Y(t1). When Y(t1) ≥ T3, the warning controller upgrades the warning level to level-III warning and adjusts the flashing frequency of the light strip to 120 times / min. At the same time, the warning controller dynamically migrates the light strip according to the real-time position of the vehicle output by the radar-vision integrated machine: when the advancing distance of the warned vehicle reaches or exceeds the layout interval of a warning light (25 m), turn off a group of warning lights at the tail of the light strip and turn on a group of warning lights in front of the light strip, so that the whole light strip moves forward in the driving direction and always remains within a preset range of about 50 m in front of the warned vehicle; the migration speed of the light strip satisfies U L = max( , 60 km / h), where is the real-time driving speed of the warned vehicle, to ensure that the migration of the light strip does not lag behind the vehicle driving.

[0056] At time t2, when the vehicle decelerates and increases the vehicle distance, the risk value decreases and satisfies Y(t2) < T1. The warning controller cancels the warning for this vehicle, restores the relevant warning lights to the non-warning state (constant yellow light), and at the same time stops the migration control of the light strip for this vehicle.

[0057] When there are multiple vehicles triggering warnings and the control intervals of the light strips overlap or conflict, the warning controller preferentially uses the warning strategy corresponding to the higher dynamic risk level to uniformly control the intelligent warning lights in the overlapping section to ensure the consistency and effectiveness of the warning prompts.

[0058] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic risk assessment and accompanying early warning method for tunnel exit sections, characterized in that, It includes the following steps: S1: Determine the early warning layout range of the tunnel exit section. The tunnel exit section is the section 0 to L m in the upstream direction inside the tunnel along the driving direction in the opposite direction of the tunnel exit portal. The exit portal is the boundary between the inside and outside of the tunnel. Here, L ≤ 500 m; S2: Install a radar-vision integrated machine in the tunnel exit section to collect traffic behavior data of the tunnel exit section in real time. The traffic behavior data includes vehicle speed, vehicle distance, and vehicle type; S3: Based on the real-time traffic behavior data, calculate the real-time rear-end collision risk value of each vehicle passing through the exit section through a dynamic rear-end collision risk assessment model; S4: Divide the dynamic risk level according to the rear-end collision risk value, including four levels: non-warning, low risk, medium risk, and high risk, and map the dynamic risk level to a preset intelligent light early warning strategy; S5: Based on the dynamic risk level and following vehicle positioning data, control the intelligent warning light to execute the corresponding intelligent light early warning strategy to achieve the accompanying early warning of the vehicle driving lights.

2. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, In step S1: The L is determined according to the historical accident point distribution and / or tunnel length. Specifically: on the premise that L does not exceed 500 m, L is taken as the minimum length covering the accident concentration area upstream of the exit portal; when the historical accident data is insufficient or there is no obvious concentration area of accident points, L is taken as the preset default length, and the preset default length is 200 m.

3. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, Collect vehicle type, vehicle speed, and vehicle distance data through a radar-vision integrated machine, and the detection range covers the tunnel exit section.

4. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, In step S2: There are multiple radar-vision integrated machines, which are evenly arranged along the axis of the tunnel at the top inside the tunnel, and the layout interval between adjacent radar-vision integrated machines is not greater than 1 / 2 of the maximum effective detection distance of the radar-vision integrated machine.

5. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, In step S3, the specific method for conducting an operation risk assessment on the accidents in the tunnel exit section is as follows: S3.1: Conduct a risk assessment based on the vehicle driving state data monitored in the exit section. When the distance between two vehicles is less than 100 m, identify the types of adjacent vehicles, and divide the following vehicle combinations into four situations: small vehicle following small vehicle, small vehicle following large vehicle, large vehicle following large vehicle, and large vehicle following small vehicle; S3.2: Based on the four scenarios described above, input the vehicle type and subsequent vehicle speed. Speed ​​of the vehicle in front Distance between vehicles The rear-end collision risk value Y of the following vehicle is calculated by selecting the corresponding rear-end collision risk model, wherein the rear-end collision risk model includes: in, , , , These represent the risk values ​​for rear-end collisions when a small car follows another small car, a small car follows another large car, a large car follows another large car, and a large car follows another small car, respectively.

6. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, In step S4, set the first threshold T1, the second threshold T2, and the third threshold T3, and divide the dynamic risk level according to the rear-end collision risk value and trigger the corresponding warning level. Specifically: When Y < T1, it is determined as the non-warning state and no warning is required; When T1 ≤ Y < T2, it is determined as low risk and trigger a level I warning; When T2 ≤ Y < T3, it is determined as medium risk and trigger a level II warning When Y ≥ T3, it is determined as high risk and trigger a level III warning; The first threshold T1 = 1.986, the second threshold T2 = 2.665, and the third threshold T3 = 3.

654.

7. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, In step S4, the preset intelligent light early warning strategy adopts a following type of light early warning. Specifically: when in the non-warning state, the intelligent warning light is yellow and always on; when in level I, II, and III warnings, the intelligent warning light is red and flashing, and the flashing frequencies are 30 times / min, 60 times / min, and 120 times / min respectively; The intelligent early warning light device is installed on the inner side wall of the tunnel and is arranged along the direction of traffic; the intelligent early warning light device is installed at a height of 1.2m above the road surface and the longitudinal spacing is 25m.

8. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, The specific method for step S5 is as follows: S5.1: The real-time position of the following vehicle is obtained through the integrated radar vision device. When the rear-end collision risk value of the vehicle is detected to be greater than or equal to T1, the two sets of lights downstream of the vehicle are activated to form a light strip. S5.2: The light strip will dynamically adjust as the warning vehicle moves, ensuring that the intelligent warning light always follows in front of the warning vehicle, with a light strip migration speed U. L =max( (60km / h); S5.3: When the detected risk value drops below T1, the intelligent warning light is switched to non-warning mode.

9. The dynamic risk assessment and accompanying early warning method for tunnel exit sections according to claim 1, characterized in that, In step S5, when there are multiple warning vehicles and their corresponding dynamic risk levels conflict, the intelligent light warning strategy corresponding to the higher dynamic risk level is used for control.

10. A dynamic risk assessment and accompanying early warning device for tunnel exit sections, employing the dynamic risk assessment and accompanying early warning method for tunnel exit sections as described in any one of claims 1 to 9, characterized in that, It includes a radar-visual integrated machine, an early warning controller, and an intelligent early warning light. The early warning controller is connected to the radar-visual integrated machine and the intelligent early warning light, respectively. The integrated radar-visual system is used to collect traffic behavior data at the tunnel exit section; The traffic behavior data includes vehicle speed, vehicle spacing, and vehicle type. The early warning controller is used to calculate the real-time rear-end collision risk value of each vehicle when it passes through the exit section based on real-time traffic behavior data and a dynamic rear-end collision risk assessment model. Based on the rear-end collision risk value, the corresponding dynamic risk level is obtained, and the corresponding intelligent light warning strategy is derived. Then, based on the dynamic risk level and the location data of the following vehicle, the intelligent warning lights are controlled to execute the corresponding intelligent light warning strategy, so as to realize the accompanying warning of the vehicle's running lights.

Images