CN114364105A - 10 KM-level intelligent lighting control system for emergency rescue of ultra-long highway tunnel - Google Patents

Abstract

The invention discloses a 10 KM-level intelligent lighting control system for emergency rescue of an ultra-long highway tunnel, which comprises an emergency lighting system for providing emergency rescue lighting for vehicles when a traffic accident occurs in the tunnel, a collecting device for collecting tunnel image information, an identification module for identifying abnormal information, an event judgment module for judging accident information, a section division module for dividing tunnel section information, an instruction generation module for generating a corresponding control instruction and a control module for controlling the emergency lighting system to provide emergency lighting.

Description

A smart lighting control system for emergency rescue of 10KM super-long highway tunnels

technical field

The invention relates to the technical field of tunnel lighting, in particular to an emergency rescue intelligent lighting control system for a 10KM super-long highway tunnel.

Background technique

Extra-long tunnels have the characteristics of long lines, limited space, special environment, difficulty in rescue after an accident, and difficulty in evacuation. The risk of emergencies such as vehicle breakdowns, traffic accidents and even fires has greatly increased, making accident losses and accident severity much higher than in general road sections.

Most of the existing tunnel lighting can not adjust the light, the operation mode is relatively fixed, the lighting brightness mode of each section is equivalent, the working mode has potential safety hazards, and the processing form of emergencies is single. In addition, the existing emergency rescue lighting in the tunnel lacks effective guidance for drivers, rescuers, and supervisors, resulting in long evacuation times, inability to effectively control accidents, and difficulties in rescue, thus increasing the time for traffic control and affecting traffic. efficiency.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to provide a smart lighting control system for emergency rescue of 10KM-level super-extra-long highway tunnels, so as to solve the problem that the evacuation time of crowds after a traffic accident in the tunnel is too long, the accident cannot be effectively controlled, and the rescue operation in the prior art is solved. difficult question.

In order to achieve the above purpose, the present invention provides a 10KM-level super-extra-long highway tunnel emergency rescue intelligent lighting control system, including:

Emergency lighting system, used to provide emergency rescue lighting for vehicles in the event of a traffic accident in the tunnel;

a collection device for collecting real-time images of each road section in the tunnel, and splicing the real-time images to form complete image information in the tunnel;

an identification module, configured to identify the driving state of the vehicle in the tunnel according to the image information, and when the driving state of the vehicle is abnormal, extract the driving track of the abnormal vehicle and/or the on-site traffic condition to form abnormal information;

an event determination module, configured to match the abnormal information with a preset traffic accident, and when the matching is successful, generate accident information corresponding to the traffic accident, wherein the accident information includes the type of the traffic accident and the accident location;

a section division module, configured to judge the impact degree of the traffic accident according to the accident information, and divide the tunnel into accident sections, affected sections and non-influenced sections based on the impact degree to form tunnel section information;

an instruction generation module, configured to respectively generate corresponding emergency instructions, evacuation instructions and hold instructions in the accident section, the affected section and the non-influenced section according to the accident information and the tunnel section information; and

The control module is configured to control the emergency lighting system of the accident section based on the confirmation request and according to the emergency instruction to provide emergency lighting according to the set emergency lighting rules, and to control the emergency lighting system of the affected section according to the set emergency lighting according to the evacuation instruction. The evacuation lighting rule provides evacuation lighting and controls the emergency lighting system of the non-affected section to maintain the current lighting state according to the maintain instruction.

Further, it also includes an evacuation path generation module;

The evacuation path generation module is used to calculate the shortest distance between the non-traffic accident vehicles in the tunnel and the tunnel entrance and exit according to the tunnel section information, the real-time position of each vehicle in the tunnel, and the designed entrance and exit positions of the tunnel and the positions of each cross passage entrance. The evacuation path forms the corresponding evacuation path information;

The instruction generation module is further configured to generate the evacuation instruction in the affected section according to the evacuation path information;

The control module is further configured to control the emergency lighting system along the evacuation path to provide evacuation lighting according to the set evacuation lighting rules based on the confirmation request and according to the evacuation instruction.

Further, the evacuation path generation module includes:

a position acquisition sub-module, used for acquiring the real-time position of each vehicle in the tunnel, the position of the entrance and exit of the tunnel and the position of each transverse passage according to the real-time image information;

The node configuration sub-module is used to configure the entrance and exit of the tunnel and each transverse passage as link nodes according to the construction design drawings of the tunnel, and to set the traffic direction of each link node and to assign each link node according to the tunnel entrance to the transverse passage. The directions to the tunnel exit are numbered in sequence;

The path link sub-module is used to link the real-time position of each vehicle with each link node with the direction of travel according to the real-time position of the non-traffic accident vehicle in the tunnel, the position of the entrance and exit of the tunnel and the position of each cross passage, and generate a path link. a number of possible paths; and

The path screening sub-module is used to screen out the feasible paths with the shortest distance between each vehicle and the tunnel entrance and exit from the several feasible paths to form an evacuation path, and then form the evacuation path information of each vehicle.

Further, the collection device includes:

Cameras evenly arranged along the length of the tunnel are used to collect real-time images of each road section in the tunnel, wherein the critical areas of the effective monitoring ranges of two adjacent cameras are connected to each other or overlap; and

The image stitching sub-module is used to stitch the real-time images of each road section in the tunnel to form complete and continuous image information in the tunnel.

Further, the identification module includes:

an image preprocessing sub-module, used for extracting the moving vehicle in the image information by using the background difference method, and identifying the outline of the moving vehicle to form a tracking target;

The target tracking sub-module is used to divide the image information by frame to obtain the video frame sequence of the tracking target, and then track each moving vehicle with the same tracking target as the unit of each frame image of the video frame sequence, and combine with Kalman The filtering algorithm predicts the position of the next frame of the tracking target;

The abnormal information identification model is used to take the video frame sequence as input, identify the motion speed and motion direction of the moving vehicle in the video frame sequence, and change the threshold between the motion speed and motion direction of the moving vehicle and preset parameters When there is no match, it is determined that the moving vehicle is in an abnormal driving state, and a vehicle driving abnormal signal is output;

An abnormal information extraction sub-module, configured to extract the driving track and/or the on-site traffic condition of the moving vehicle with abnormal driving state in the video frame sequence according to the vehicle driving abnormal signal to form the abnormal information; and

The model training sub-module is used to iteratively train the abnormal information recognition model with the acquired historical traffic accident data as the training set, until the set loss function tends to be stable or the maximum number of iterations is reached, and the training is completed.

Further, the segment dividing module includes:

The sub-module of tunnel segment is used to divide the tunnel into several sub-tunnel segments according to the position of the cross passage according to the construction design drawings of the tunnel;

The statistics and calculation sub-module is used to extract the image of each sub-tunnel segment from the image information, and count the proportion of all vehicles in each image in the image, and calculate the cross-sectional traffic occupancy rate of each sub-tunnel segment;

The comparison sub-module is used to compare the traffic occupancy rate of the section with the set first threshold and the second threshold to obtain the influence degree of each sub-tunnel section when a traffic accident occurs, and the first threshold is less than the first threshold. two thresholds; and

A section determination sub-module, configured to determine the section category of each sub-tunnel section as an accident section, an affected section and a non-influenced section according to the comparison result of the comparison sub-module to form the tunnel section information .

Further, the statistics and calculation sub-module adopts the following formula to calculate the cross-sectional traffic occupancy rate in each sub-tunnel segment:

Where: O _i is the cross-sectional traffic occupancy rate of the _ith sub-tunnel section, Li is the total area occupied by all vehicles in the image of the ith sub-tunnel section, D _i is the total area corresponding to the image of the ith sub-tunnel section, i =1,2,...,n, where n is the number of sub-tunnel segments.

Further, the emergency lighting system includes emergency lighting lamps uniformly arranged on both side walls of the tunnel along the length of the tunnel, and the emergency lighting lamps have a variety of lighting colors and flashing frequencies that can be combined to form corresponding lighting rules; the Emergency lighting rules are set in one-to-one correspondence with the traffic accident types.

Further, it also includes a traffic management server and an alarm module;

The traffic management server is used to realize data interaction with the control module, store and display the accident information, and can generate an emergency trigger signal based on the confirmation request of the management personnel for the accident information or the automatic confirmation request of the system delay. ;

The instruction generation module is further configured to generate an alarm instruction according to the emergency trigger signal;

The alarm module is used for executing corresponding alarm according to the alarm instruction.

Further, the control module includes:

The upper computer is used to display and store accident information;

a controller host, configured to send control instructions to each accident section, affected section and non-affected section respectively according to the emergency trigger signal and corresponding emergency instructions, evacuation instructions and hold instructions; and

A number of section control extensions are used to control the emergency lighting system in its section according to the control instructions sent by the controller host to provide emergency lighting according to the set emergency lighting rules, provide evacuation lighting according to the set evacuation lighting rules, or maintain Current lighting state.

In this scheme, by setting the acquisition device, the identification module and the event determination module, based on the collected image information combined with the preset traffic accident, a double determination of whether a traffic accident has occurred in the tunnel can be realized, so as to improve the determination accuracy of the traffic accident; The division module can divide the tunnel into accident sections, affected sections and non-affected sections according to the impact of the traffic accident on each section of the tunnel after a traffic accident occurs, and control the emergency response in the corresponding section according to the different types of sections. The lighting system provides lighting for the tunnel according to the set emergency lighting rules, evacuation lighting rules or maintaining the original lighting state, so that the vehicles and drivers in the tunnel can be rescued and evacuated at the fastest speed, and the resulting traffic congestion can be reduced. , to minimize accident losses.

Description of drawings

FIG. 1 is a structural block diagram of a smart lighting control system for emergency rescue of a 10KM-level super extra-long highway tunnel according to the present invention.

FIG. 2 is a structural block diagram of the acquisition module in FIG. 1 .

FIG. 3 is a structural block diagram of the identification module in FIG. 1 .

FIG. 4 is a structural block diagram of the segment dividing module in FIG. 1 .

FIG. 5 is a control block diagram of the evacuation path generation module in FIG. 1 .

FIG. 6 is a control block diagram of the control module in FIG. 1 .

FIG. 7 is a structural block diagram of a smart lighting control system for emergency rescue of a 10KM-level super-extra-long highway tunnel according to another embodiment of the present invention.

Detailed ways

The following is further described in detail by specific embodiments:

Example

Although the emergency rescue intelligent lighting control system in this embodiment is described by taking a 10KM level road tunnel as an example, it is not limited to the emergency rescue lighting for 10KM level road tunnels, but can also be used for other longer or shorter level roads. The emergency rescue lighting in the tunnel is used to guide and evacuate the vehicles and drivers in the tunnel in time when a traffic accident occurs in the tunnel, thereby minimizing the impact of the accident.

As shown in FIG. 1 , a control block diagram of a smart lighting control system for emergency rescue of a 10KM-level super-extra-long highway tunnel according to this embodiment. This embodiment includes an emergency lighting system 1 , a collection device 2 , an identification module 3 , an event determination module 4 , a section division module 5 , an evacuation path generation module 6 , an instruction generation module 7 and a control module 8 . in:

The emergency lighting system 1 can provide emergency rescue lighting for vehicles when a traffic accident occurs in the tunnel; it specifically includes emergency lighting and evacuation lighting when a traffic accident occurs. The emergency lighting system 1 includes emergency lighting lamps uniformly arranged on both side walls of the tunnel along the length of the tunnel, and each emergency lighting lamp has a variety of lighting colors and flickering frequencies, so as to combine the lighting and flickering frequencies of different colors. A variety of lighting rules are formed to realize emergency lighting and evacuation lighting in the event of a traffic accident in the tunnel.

The collection device 2 can collect real-time images of each road section in the tunnel, and splices the collected real-time images through image splicing technology to form complete image information in the entire tunnel.

Specifically, as shown in FIG. 2 , the collection device 2 includes a plurality of cameras 21 and an image splicing sub-module 22 uniformly arranged on the side walls of the tunnel along the length direction of the tunnel. The cameras 21 can realize real-time monitoring of the corresponding road section tunnel Image collection, and the critical areas of the effective monitoring range of two adjacent cameras 21 are connected to each other or have overlapping parts, so that the collected real-time images of the tunnel ignore blind spots in the area and reach the maximum value, so as to increase the accuracy of traffic accident determination. The image stitching sub-module 22 can be connected with each camera 21 through a video distributor to receive the real-time images of the corresponding road sections in the tunnel collected by each camera 21, and stitch each real-time image through an image stitching algorithm to form a complete and continuous tunnel in the tunnel. The seamless panoramic image is obtained, and the image information in the tunnel is obtained and transmitted to the identification module 3.

The recognition module 3 receives the image information of the image stitching sub-module 22, and identifies and judges the driving state of the vehicle in the tunnel according to the image information. The driving state includes a normal driving state and an abnormal driving state. When the driving state of the vehicle is abnormal, the abnormal information is formed by extracting the driving track of the abnormal vehicle and/or the on-site traffic condition.

As shown in FIG. 3 , the recognition module 3 includes a model training sub-module 31 , an image preprocessing sub-module 32 , a target tracking sub-module 33 , an abnormal information recognition model 34 and an abnormal information extraction sub-module 35 . In this embodiment, the abnormal information recognition model 34 is implemented based on a neural network, and the abnormal information recognition model 34 can be trained by the model training sub-module 31, so that after the training is completed, the abnormal information recognition model 34 can be based on the image The video frame sequence obtained after processing by the preprocessing sub-module 32 and the target tracking sub-module 33 identifies the parameters of the moving vehicle in the image information and judges whether its driving state is abnormal. in:

The model training sub-module 31 takes the historical traffic accident data obtained in the traffic control department or through crawling as a training set, and inputs the training set into the abnormal information recognition model 34, and the abnormal information recognition model 34 is based on the abnormal information recognition model 34. (that is, the neural network) forward propagation, extract the corresponding parameters, and input the obtained parameters into the abnormal information identification model 34 for iterative training until the set loss function tends to be stable or reaches the maximum number of iterations After the training is completed, a trained abnormal information recognition model 34 is obtained, and the abnormal information recognition model 34 is then used for the recognition of the parameters of the moving vehicle in the image information and the judgment of the abnormal driving state.

Specifically, the image preprocessing sub-module 32 receives the image information, uses the background difference method to detect the moving vehicle in the image information based on the video image processing technology, and detects the moving vehicle in the image information. For medium moving vehicles, through adaptive threshold segmentation, morphological denoising, shadow removal and vehicle outline identification, only moving vehicles are included in the foreground image, and the moving vehicles are used to form tracking targets.

The target tracking sub-module 33 receives the image information processed by the image preprocessing sub-module 32, and divides the image information into frames to obtain a video frame sequence of the tracking target. In this embodiment, the target tracking sub-module 33 uses the feature matching method to track the same target in each frame; in specific implementation, a camshift tracker can be established for each detected moving vehicle (ie, the same tracking target). , in order to achieve the tracking of multiple moving vehicles (that is, multiple tracking targets), and combine the Kalman filtering algorithm to predict the position of the next frame of the tracking target, which can solve the problem of partial occlusion of vehicles and obtain the motion in the video frame sequence. Complete vehicle trajectory and on-site traffic status images.

The abnormal information recognition model 34 takes the video frame sequence processed by the target tracking sub-module 33 as input, and identifies the motion speed and motion direction of the moving vehicle in the video frame sequence based on the forward propagation of the neural network. When the motion speed and motion direction of the moving vehicle do not match the preset parameter change threshold, it is determined that the moving vehicle is in an abnormal driving state, and a vehicle driving abnormal signal is output. In this embodiment, the preset parameter change threshold may be set according to rules such as speed limit and traffic limit in the tunnel.

The abnormal information extraction sub-module 35 receives the abnormal signal output by the abnormal information identification model 34, and extracts the driving track and/or the on-site traffic condition of the moving vehicle with abnormal driving state in the video frame sequence to form the abnormal information, so as to form the abnormal information. It is used for automatic determination of subsequent traffic accidents.

The event determination module 4 receives the abnormal information extracted by the abnormal information extraction sub-module 35, matches the abnormal information with a preset traffic accident, and when the matching is successful, generates accident information corresponding to the traffic accident, wherein, The accident information includes the type of traffic accident and the location of the accident. In this embodiment, the types of traffic accidents in the tunnel include at least Type I, II, III, and IV traffic accidents, the Type I traffic accidents mainly include fire, etc., and the Type II traffic accidents mainly include vehicle rear-end collision, two Vehicle collisions, etc., the Class III traffic accidents mainly include vehicles hitting the inner wall of the tunnel, etc., and the Class IV traffic accidents mainly include vehicle rollovers and the like. In specific implementation, it can be implemented based on machine learning algorithms, that is, a neural network model is trained by extracting the characteristic values of the historical data of the above four types of traffic accidents as training samples, and when abnormal information is received, the abnormal information is extracted. The characteristic value of , and the characteristic value is cyclically compared with the set threshold to realize the determination of the type of traffic accident and the location of the accident.

The section division module 5 receives the accident information, and calculates the cross-sectional traffic occupancy rate of each section in the tunnel according to the accident information, thereby judging the degree of influence of the traffic accident on each road section in the tunnel, based on the degree of influence Dividing the tunnel into accident sections, affected sections and non-affected sections forms tunnel section information.

As shown in FIG. 4 , the segment dividing module 5 includes a tunnel segment sub-module 51 , a statistics and calculation sub-module 52 , a comparison sub-module 53 and a segment determination sub-module 54 . in:

The tunnel segment sub-module 51 divides the tunnel into several sub-tunnel segments according to the positions of the transverse passages according to the construction design drawings of the tunnel obtained from the relevant departments. In other optional embodiments, the tunnel may also be divided into several sub-tunnel segments according to the installation positions of the cameras 21 or emergency lighting lamps or even the installation positions of the ventilation openings in the tunnel.

The statistics and calculation sub-module 52 extracts the image of each sub-tunnel segment from the image information according to the segmentation of the tunnel by the tunnel segmentation sub-module 51, and detects the cross-section traffic occupancy rate measurement method based on video image processing. And count the number of vehicles in each image, approximate the area of each vehicle in the image as the total number of all pixels in each sub-tunnel The traffic share of the section is calculated. Specifically, the total area Li of all vehicles in the image of the corresponding sub-tunnel segment in the image of the sub-tunnel segment i is obtained by summing the average standard, and the total area occupied by all vehicles in the image of the sub-tunnel segment _i is compared to The total area corresponding to the image of the i-th sub-tunnel segment is calculated to obtain the cross-sectional traffic occupancy O _i of each sub-tunnel segment:

Where: O _i is the cross-sectional traffic occupancy rate of the _ith sub-tunnel section, Li is the total area occupied by all vehicles in the image of the ith sub-tunnel section, D _i is the total area corresponding to the image of the ith sub-tunnel section, i =1,2,...,n, where n is the number of sub-tunnel segments.

The comparison sub-module 53 receives the cross-section traffic occupancy rate of each sub-tunnel segment, and compares the cross-section traffic occupancy rate with the set first and second thresholds to obtain the traffic occupancy rate of each sub-tunnel segment when a traffic accident occurs. Influence degree, in this embodiment, the first threshold value is smaller than the second threshold value, and the first threshold value and the second threshold value can be determined according to the number of lanes of the tunnel and the width of the lanes.

The section determination sub-module 54 determines the section type of each sub-tunnel section as accident section, affected section and non-influenced section according to the comparison result of the comparison sub-module 53 to form the tunnel section information. In this embodiment, when the traffic occupancy rate of the section is less than a first threshold, the sub-tunnel segment is determined to be a non-influenced section; when the traffic occupancy rate of the section is greater than or equal to the first threshold and less than a second threshold , the sub-tunnel section is determined to be an affected section; when the cross-section traffic occupancy rate is greater than or equal to the second threshold, the sub-tunnel section is determined to be an accident section. When determining the affected section, according to the accident section and the current driving direction of the tunnel, the road sections behind the accident section to the tunnel entrance can be determined as the affected section, and the corresponding accident section The road section ahead until the tunnel exit is determined as a non-influenced section.

In other optional embodiments, the section dividing module 5 may also determine the degree of influence of the traffic accident on each road section in the tunnel by calculating the average speed of the vehicle in each sub-tunnel section per unit time according to the accident information. Specifically, the statistics and calculation sub-module 52 can use the following formula to calculate the average speed of the vehicle per unit time in each sub-tunnel section

分别为运动车辆在t时刻和t-1时刻相对于x轴的质心位置；

分别为运动车辆在t时刻和t-1时刻相对于y轴的质心位置；m为第i子隧道段经过的车辆总数。in:

are the position of the center of mass of the moving vehicle relative to the x-axis at time t and time t-1, respectively;

are the position of the center of mass of the moving vehicle relative to the y-axis at time t and time t-1, respectively; m is the total number of vehicles passing through the i-th sub-tunnel segment.

与交通事故影响等级的映射关系确定对应子隧道段内的交通事故的影响程度，所述区段确定子模块54根据对应影响程度确定每一子隧道段的区段类别。Correspondingly, the comparison sub-module 53 calculates the average speed of the vehicle per unit time in each sub-tunnel section.

The mapping relationship with the traffic accident impact level determines the impact degree of the traffic accident in the corresponding sub-tunnel segment, and the segment determination sub-module 54 determines the segment type of each sub-tunnel segment according to the corresponding impact degree.

The evacuation route generation module 6 receives the tunnel section information, determines the accident section and the affected section, and obtains the real-time positions of the vehicles in the accident section and the affected section in the tunnel and the entrances and exits of the tunnel design based on the image information. According to the position and the position of each cross passage entrance, the corresponding evacuation path information is calculated to form the evacuation path with the shortest distance between the non-traffic accident vehicles in the accident section and the affected section in the tunnel and the tunnel entrance and exit.

As shown in FIG. 5 , the evacuation route generation module 6 includes a location acquisition submodule 61 , a node configuration submodule 62 , a route link submodule 63 and a route screening submodule 64 . in:

The position obtaining sub-module 61 can obtain the real-time position of each vehicle in the tunnel, the position of the entrance and exit of the tunnel and the position of each cross passage according to the real-time image information, so as to be used for subsequent calculation of feasible paths and determination of evacuation paths.

The node configuration sub-module 62 configures the tunnel's entrance and exit and each transverse passageway as link nodes according to the construction design drawings of the tunnel obtained from the relevant departments, and sets the traffic direction of each link node, and each link node is a link node. One-way traffic, and then number each link node in turn according to the direction from the tunnel entrance to the cross channel entrance and then to the tunnel exit to form a link node sequence with a traffic direction.

The path link sub-module 63 receives the real-time position of the non-traffic accident vehicle in the tunnel, the tunnel entrance and exit positions and the positions of each cross passage, and performs path linking between the real-time position of each vehicle and each link node with the direction of travel (when When a traffic accident occurs, each link node corresponding to the accident section fails, and should be removed during path linking), and several feasible paths can be generated for each vehicle to leave the tunnel.

The path screening sub-module 64 receives all feasible paths, and selects the feasible path with the shortest distance between each vehicle and the tunnel entrance and exit from the several feasible paths as the evacuation path of the vehicle, thereby forming the evacuation path information of each vehicle.

The instruction generation module 7 can generate an emergency instruction corresponding to controlling the emergency lighting in the section in the accident section and generate an emergency instruction corresponding to controlling the emergency lighting in the section in the non-affected section according to the accident information and the tunnel section information. Keeping instructions for lighting lamps; at the same time, according to the accident information, tunnel section information and evacuation path information, an evacuation instruction corresponding to controlling the emergency lighting in the section is generated in the affected section.

The control module 8 can control the emergency lighting system 1 of the accident section based on the confirmation request and according to the emergency instruction to provide emergency lighting according to the set emergency lighting rules, and control the affected section (that is, the generated evacuation) according to the evacuation instruction. The emergency lighting system 1 along the route) provides evacuation lighting according to the set evacuation lighting rules, and controls the emergency lighting system 1 of the non-affected section to maintain the current lighting state according to the maintaining instruction. In this embodiment, the confirmation request can be sent by a traffic management server 9 to the control module 8. In the specific implementation, the traffic management server 9 can realize data interaction with the control module 8 through wireless means, and store and display the data. The accident information is based on the confirmation request made by the management personnel to determine whether the accident information actually occurred or the confirmation request of the system for automatic delay confirmation of the accident information after a preset time countdown ends, and based on the confirmation request of the accident information. The confirmation request generates an emergency trigger signal and sends it to the control module 8 .

As shown in FIG. 6 , the control module 8 includes a host computer 81 , a controller host 82 and a plurality of segment control extensions 83 . In this embodiment, the host computer 81 transmits data with the controller host 82 through the RS485 bus, so as to monitor the entire system, and display and store accident information; the controller host 82 and the segment control extension 83 The CAN bus technology is used to realize data transmission between them. in:

The controller host 82 respectively sends control instructions to each accident section, affected section and non-affected section according to the emergency trigger signal and the corresponding emergency instructions, evacuation instructions and hold instructions, so as to control the corresponding section control. The extension 83 performs corresponding control operations.

The section control extension 83 controls the emergency lighting system 1 in its section according to the control instructions sent by the controller host 82 to provide emergency lighting according to the set emergency lighting rules, provide evacuation lighting according to the set evacuation lighting rules, or Keep the current lighting state.

In this embodiment, the lighting rules (including emergency lighting rules and evacuation lighting rules) are set in a one-to-one correspondence with the traffic accident types. specific:

The lighting rules corresponding to the Class I traffic accident are:

Control the emergency lights on both sides of the accident section to light red to warn vehicles and drivers to stay away from the accident section; control the emergency lights on both sides of the affected section to light green, and flash in sequence according to the evacuation route information and the set evacuation lighting rules. Form a flow of light to guide vehicles and drivers to evacuate according to the evacuation path; and control the emergency lighting around fire equipment boxes and emergency escape passages to flash alternately red and blue to induce drivers and passengers to self-rescue, extinguish fires and escape; the emergency lighting The flashing frequency of the lamp is 3Hz. At the same time, adjust the color of the normal driving lights in the tunnel to yellow or other lights with strong penetration.

The lighting rules corresponding to the Class II traffic accident are:

If the vehicle rear-ends or collides but does not affect the adjacent lane, the emergency lighting of the lane on the side where the accident occurred is controlled to light red, and the emergency lighting of the other lane to light green. The lighting rules flicker in sequence to form a light flow, and guide vehicles and drivers to evacuate according to the evacuation path; if a vehicle rear-end collision or collision affects adjacent lanes, control the emergency lighting in the accident section to light red and affect both sides of the section. The emergency lighting of the vehicle lights up green, and flashes in sequence according to the evacuation path information and the set evacuation lighting rules to form a light flow to guide the vehicle and the occupants to evacuate according to the evacuation path; and control the emergency lighting around the fire equipment box and the emergency escape passage. The light flashes alternately in red and blue light to induce drivers and passengers to self-rescue and extinguish fire and escape; the flashing frequency of the emergency lighting is 2Hz. At the same time, adjust the color of the normal driving lights in the tunnel to yellow or other lights with strong penetration.

The lighting rules corresponding to the Class III traffic accidents are:

Generally, a vehicle collides with a sidewall and only affects the driving of one lane. In the accident zone, the emergency lighting of the lane on the side where the accident occurred is controlled to light red, and the emergency lighting of the other lane to light green. The lighting rules flicker in sequence to form a light flow, and guide vehicles and drivers to evacuate according to the evacuation path. At the same time, adjust the color of the normal driving lights in the tunnel to yellow or other lights with strong penetration.

The lighting rules corresponding to the Class IV traffic accidents are:

For a small vehicle rollover, the driver and passengers can generally rescue themselves and restore the correct posture of the vehicle, so they only need to control the emergency lighting to light green. The emergency lighting of the vehicle lights up green, and flashes in sequence according to the evacuation route information and the set evacuation lighting rules to form a light flow, and guide vehicles and drivers to evacuate according to the evacuation route. At the same time, adjust the color of the normal driving lights in the tunnel to yellow or other lights with strong penetration.

The emergency lighting lamp in this embodiment can be a smart lighting lamp with monitoring, and has an independent address, which can integrate the emergency lighting lamp and the collecting device 2 into one, reducing the procurement cost and construction cost, and this embodiment adopts a centralized power supply to concentrate In the control mode, a battery is set inside the emergency lighting, which can monitor the working status of the emergency lighting in real time, reduce the maintenance and operating costs of the emergency lighting, and achieve the purpose of energy saving and environmental protection.

The intelligent lighting control system for emergency rescue of a 10KM-level super-extra-long highway tunnel in this embodiment is set with various types of traffic accidents as a reference. , the situation will be automatically determined as this type of traffic accident; the identification module 3 and the event determination module 4 will double-judge whether there is a traffic accident in the tunnel, which can improve the accuracy of the judgment of traffic accidents in the tunnel, and then the occurrence of traffic accidents in the tunnel can be improved. In the event of a traffic accident, according to the divided tunnel section information, the evacuation path is automatically calculated, and the emergency lighting system 1 is controlled to guide the vehicles and the drivers and passengers to evacuate in time according to different lighting rules, so that the rescue and evacuation can be carried out at the fastest speed. Reduce the resulting traffic jams and minimize accident losses.

As another embodiment of the present invention, as shown in FIG. 7 , this embodiment further includes an alarm module 10, which is used to issue an alarm sound to remind the rescue team or other rescuers when a traffic accident occurs in the tunnel, and guide the rescuers Get to the scene quickly to improve rescue efficiency. Specifically, the instruction generation module 7 may also generate an alarm instruction according to the emergency trigger signal; the alarm module 10 executes a corresponding warning according to the alarm instruction to prompt rescuers.

The above descriptions are only embodiments of the present invention, and common knowledge such as well-known specific structures and characteristics in the solution are not described too much here. It should be pointed out that for those skilled in the art, some modifications and improvements can be made without departing from the structure of the present invention. These should also be regarded as the protection scope of the present invention, and these will not affect the implementation of the present invention. Effects and applicability of the present invention.

Claims (10)

1. The utility model provides a 10KM level super extra long highway tunnel emergency rescue wisdom lighting control system which characterized in that includes:

the emergency lighting system is used for providing emergency rescue lighting for the vehicle when a traffic accident occurs in the tunnel;

the acquisition device is used for acquiring real-time images of all road sections in the tunnel and splicing the real-time images to form complete image information in the tunnel;

the identification module is used for identifying the running state of the vehicle in the tunnel according to the image information, and extracting the running track of the abnormal vehicle and/or the site traffic condition to form abnormal information when the running state of the vehicle is abnormal;

the event judgment module is used for matching the abnormal information with a preset traffic accident and generating accident information corresponding to the traffic accident when the matching is successful, wherein the accident information comprises a traffic accident type and an accident position;

the section dividing module is used for judging the influence degree of the traffic accident according to the accident information and dividing the tunnel into an accident section, an influence section and a non-influence section based on the influence degree to form tunnel section information;

the instruction generation module is used for respectively generating corresponding emergency instructions, evacuation instructions and holding instructions in the accident zone, the affected zone and the non-affected zone according to the accident information and the tunnel zone information; and

and the control module is used for controlling the emergency lighting system of the accident zone to provide emergency lighting according to a set emergency lighting rule based on the confirmation request and according to the emergency instruction, controlling the emergency lighting system of the affected zone to provide evacuation lighting according to a set evacuation lighting rule according to the evacuation instruction, and controlling the emergency lighting system of the non-affected zone to keep a current lighting state according to the keeping instruction.

2. The intelligent 10 KM-class ultra-long highway tunnel emergency rescue lighting control system according to claim 1, further comprising an evacuation path generation module;

the evacuation path generation module is used for calculating the shortest distance evacuation path between the non-traffic accident vehicle in the tunnel and the tunnel entrance and exit according to the tunnel section information, the real-time position of each vehicle in the tunnel, the designed entrance and exit position of the tunnel and the position of each transverse passage entrance to form corresponding evacuation path information;

the instruction generating module is further used for generating the evacuation instruction in the influence section according to the evacuation path information;

the control module is also used for controlling emergency lighting systems along the evacuation path to provide evacuation lighting according to set evacuation lighting rules based on the confirmation request and according to the evacuation instructions.

3. The intelligent 10 KM-class ultra-long highway tunnel emergency rescue lighting control system according to claim 2, wherein the evacuation path generating module comprises:

the position acquisition submodule is used for acquiring the real-time positions of all vehicles in the tunnel, the positions of the entrance and the exit of the tunnel and the positions of all transverse passage openings according to the real-time image information;

the node configuration submodule is used for configuring the entrance and the exit of the tunnel and each transverse passage opening into link nodes according to a construction design drawing of the tunnel, setting the passing direction of each link node and numbering each link node in sequence according to the direction from the tunnel entrance to the transverse passage opening and then to the tunnel exit;

the path linking submodule is used for performing path linking on the real-time position of each vehicle and each link node with the passing direction according to the real-time position of the non-traffic accident vehicle in the tunnel, the tunnel entrance and exit positions and the positions of all transverse passage ports to generate a plurality of feasible paths; and

and the path screening submodule is used for screening the feasible paths with the shortest distance between each vehicle and the tunnel entrance and exit from the feasible paths to form evacuation paths, and further forming evacuation path information of each vehicle.

4. The intelligent 10 KM-class ultra-long highway tunnel emergency rescue lighting control system according to claim 1, wherein the collecting device comprises:

the cameras are uniformly arranged along the length direction of the tunnel and are used for acquiring real-time images of all road sections in the tunnel, wherein critical areas of effective monitoring ranges of two adjacent cameras are mutually connected or have overlapped parts; and

and the image splicing submodule is used for splicing the real-time images of all road sections in the tunnel to form complete and continuous image information in the tunnel.

5. The intelligent 10 KM-class ultra-long highway tunnel emergency rescue lighting control system according to claim 1, wherein the identification module comprises:

the image preprocessing submodule is used for extracting the moving vehicles in the image information by adopting a background difference method and marking the outlines of the moving vehicles to form a tracking target;

the target tracking submodule is used for segmenting image information according to frames to obtain a video frame sequence of a tracking target, respectively tracking each moving vehicle by taking the same tracking target as a unit for each frame image of the video frame sequence, and predicting the position of the next frame of the tracking target by combining a Kalman filtering algorithm;

the abnormal information identification model is used for identifying the movement speed and the movement direction of a moving vehicle in the video frame sequence by taking the video frame sequence as input, judging that the running state of the moving vehicle is abnormal when the movement speed and the movement direction of the moving vehicle are not matched with a preset parameter change threshold value, and outputting a vehicle running abnormal signal;

the abnormal information extraction submodule is used for extracting the running track and/or the site traffic condition of the moving vehicle with abnormal running state in the video frame sequence according to the vehicle running abnormal signal to form the abnormal information; and

and the model training submodule is used for performing iterative training on the abnormal information identification model by taking the acquired historical traffic accident data as a training set until the set loss function tends to be stable or the maximum iteration times is reached, and finishing the training.

6. The intelligent 10 KM-class ultra-long highway tunnel emergency rescue lighting control system according to claim 1, wherein the segment dividing module comprises:

the tunnel segmentation sub-module is used for segmenting the tunnel into a plurality of sub-tunnel segments according to the position of the transverse channel according to a construction design drawing of the tunnel;

the statistic and calculation submodule is used for extracting the image of each sub-tunnel section from the image information, counting the proportion of all vehicles in each image in the image and calculating the cross section traffic occupancy of each sub-tunnel section;

the comparison submodule is used for comparing the section traffic occupancy with a set first threshold and a set second threshold to obtain the influence degree of each sub-tunnel section when a traffic accident occurs, and the first threshold is smaller than the second threshold; and

and the section determining submodule is used for determining the section category of each sub-tunnel section into an accident section, an influence section and a non-influence section according to the comparison result of the comparison submodule to form the tunnel section information.

7. The intelligent 10KM lighting control system for emergency rescue in ultra-long highway tunnel according to claim 6, wherein the statistic and calculation sub-module calculates the cross-sectional traffic occupancy rate in each sub-tunnel segment by using the following formula:

wherein: o is_iIs the section traffic occupancy, L, of the ith sub-tunnel segment_iIs the total area occupied by all vehicles in the image of the ith sub-tunnel segment, D_iThe total area corresponding to the image of the ith sub-tunnel segment, i ═ 1, 2.

8. The intelligent 10 KM-class ultra-extra-long highway tunnel emergency rescue lighting control system according to claim 1, wherein the emergency lighting system comprises emergency lights uniformly arranged on two side walls of the tunnel along the length direction of the tunnel, and the emergency lights have a plurality of lighting colors and flickering frequencies which can be combined to form corresponding lighting rules; the emergency lighting rules are set in one-to-one correspondence with the traffic accident types.

9. The intelligent 10KM lighting control system for emergency rescue in super-extra-long road tunnel according to claim 1, further comprising a traffic management server and an alarm module;

the traffic management server is used for realizing data interaction with the control module, storing and displaying the accident information, and generating an emergency triggering signal based on a confirmation request of a manager for the accident information or a system delay automatic confirmation request;

the instruction generating module is also used for generating an alarm instruction according to the emergency trigger signal;

and the alarm module is used for executing corresponding warning according to the alarm instruction.

10. The intelligent 10 KM-class ultra-long highway tunnel emergency rescue lighting control system according to claim 9, wherein the control module comprises:

the upper computer is used for displaying and storing accident information;

the controller host is used for respectively sending control instructions to each accident zone, each affected zone and each non-affected zone according to the emergency trigger signals and the corresponding emergency instructions, evacuation instructions and holding instructions; and

and the zone control extensions are used for controlling emergency lighting systems in zones of the controller to provide emergency lighting according to set emergency lighting rules, provide evacuation lighting according to set evacuation lighting rules or keep the current lighting state according to control instructions sent by the controller host.

Images