CN115793093B - Dam hidden disease diagnosis air-ground equipment - Google Patents

Abstract

The invention relates to dam hidden danger diagnosis air-ground integrated equipment, which comprises a vehicle-mounted platform, a command shelter, a control terminal, an unmanned detection vehicle and an unmanned inspection machine, wherein the command shelter, the control terminal, the unmanned detection vehicle and the unmanned inspection machine are carried on the vehicle-mounted platform; the control terminal classifies the acquired data according to the data sources, marks coordinate labels through coordinate matching, and divides the data according to the length according to the coordinates; establishing a dam model based on the acquired point cloud data; and extracting abnormal region coordinates after identifying the abnormal region and mapping the abnormal region coordinates to a dyke model. The equipment can realize the integrated control of the whole process of multi-source heterogeneous data, realize the rapid association of data such as geophysical prospecting, laser scanning, images and the like through data classification, segmentation and marking of coordinate labels, establish a mapping base through extracting the coordinates of the morphological control points of the dykes and dams, realize the visual display of hidden danger by combining the coordinate labels of all the data, and can rapidly and accurately evaluate the dangerous degree of hidden danger development and realize the precise allocation of emergency resources.

Description

Air-ground integrated equipment for dam concealed disease diagnosis

technical field

The invention belongs to the technical field of hidden danger detection, and in particular relates to integrated equipment for diagnosis of hidden dangers of embankments and dams.

Background technique

At present, manual inspection is still the most common method for hidden dangers of embankments, which is labor-intensive and time-consuming. The continuous development of technologies and equipment such as geophysical prospecting and remote sensing provides effective means for efficient diagnosis and scientific early warning of hidden dangers of embankments. By simply combining the existing geophysical and remote sensing equipment with unmanned vehicles and drones to form structurally integrated inspection equipment is the current cutting-edge technology, which has achieved unmanned operation to a certain extent, but there are still Obvious technical shortcomings:

1) Limited by the remote control capability of unmanned equipment, on-site inspections must be accompanied by professionals, and it is not yet possible to realize personnel station control and automatic inspections, and unmanned vehicles and drones need to be controlled separately, and the information obtained is not sufficient. Insufficient sharing and synergy.

2) The whole process of raw data acquisition, processing, interpretation, and display needs to rely on manual reading, screening, migration, and interpretation. The data flow link is separated, the interpretation software is independent, and the data format is different. It is impossible to realize multi-source heterogeneous data. Process integration control leads to slow data flow.

3) The collected data has a high degree of redundancy, and key information is hidden in it. However, data processing is often "equal" and lacks a method for extracting main key features, resulting in a slow response to data retrieval and a lag in the visual display of embankment behavior, which cannot be timely And prominently present the key features related to the safety of dike structures.

4) Due to the above reasons, data such as geophysical prospecting and remote sensing are still difficult to quickly correlate, and the level of comprehensive diagnosis of hidden dangers of embankments is insufficient, and there is a lack of corresponding judging criteria for embankment risks, resulting in the inability to accurately assess the degree of danger of the development of hidden dangers, which affects the availability of emergency rescue resources. Precise deployment.

Contents of the invention

The object of the present invention is to overcome the above-mentioned problems in the prior art, and provide an air-ground integrated equipment for dam concealed disease diagnosis.

In order to realize the above-mentioned technical purpose, the present invention adopts following technical scheme:

An air-ground integrated equipment for dam concealed disease diagnosis, including a vehicle-mounted platform and a command shelter mounted on the vehicle platform, the command shelter includes a separate equipment room and an operation room;

A control terminal, a server, and a first communication device of the equipment in the equipment room are arranged in the operation room;

Unmanned detection vehicles and unmanned inspection machines are placed in layers in the equipment room;

The unmanned detection vehicle is equipped with a mobile control device, a detection device, a video recording device, a first positioning device and a second communication device; the unmanned patrol machine is equipped with an infrared camera, a visible light camera, a laser scanning device, and a second positioning device , distance measuring device, obstacle avoidance device, third communication device;

The equipment workflow is as follows:

The vehicle-mounted platform moves to the periphery of the embankment section to be detected, starts the unmanned inspection machine to make it inspect the embankment according to the flight route, and collects the shooting data, point cloud data, and synchronous The positioning data is sent to the server through the third communication device, the control terminal plans the driving path of the unmanned detection vehicle according to the ground data returned by the unmanned inspection machine, starts the unmanned detection vehicle to inspect the embankment according to the planned path, and puts the detection device, camera The electromagnetic wave detection data, shooting data, and synchronized positioning data collected by the recording device and the first positioning device are sent to the server through the second communication device;

The control terminal classifies the data obtained by the server according to the data source, marks the coordinate label by coordinate matching, and divides the data according to the length of the coordinate; calls the point cloud data, extracts the coordinates of the shape control point based on the image edge algorithm, and establishes the embankment model; Invoke infrared camera shooting data and electromagnetic wave detection data, identify temperature and electromagnetic wave abnormal areas, extract abnormal area coordinates, and map to the embankment model.

As a preferred embodiment, the infrared camera and the visible light camera mounted on the unmanned inspection machine are infrared-visible light integrated cameras.

The hull of the command shelter is made of stainless steel pipes as the skeleton, aluminum alloy plates as the shell, and embedded heat insulating material plates;

The command shelter is fixed on the vehicle-mounted platform through a support frame with a jacking function; after the support frame is lifted, the side panels and tail panels of the command shelter are unfolded.

As a preferred implementation, the control terminal determines the road conditions based on the point cloud data sent back by the unmanned patrol machine, including vegetation and external collapse to avoid obstacles, and plans the driving path of the unmanned detection vehicle.

As a preferred implementation, the classification according to the data source, marking the coordinates by coordinate matching, and dividing the data by length according to the coordinates include:

Divide the obtained data into point cloud data, visible light image data, electromagnetic wave detection data and infrared image data according to the source; obtain the coordinates of visible light image data, electromagnetic wave detection data, and infrared image data through coordinate matching; Data and infrared image data are divided according to length.

As a preferred implementation, the control terminal plans the flight route of the unmanned patrol aircraft by calling the electronic map, or manually controls the flight route by using the joystick.

As a preferred embodiment, the unmanned detection vehicle is a crawler chassis, and there is an equipment compartment inside the crawler chassis for placing power supply devices and mobile control devices; a double-layer support platform is set on the upper part of the unmanned detection vehicle chassis, The first positioning device, video recording device and second communication device are installed on the upper layer of the support platform, and the mechanical arm and detection device are installed on the lower layer of the support platform; two sets of wheels are installed on the bottom of the detection device, the front set is universal wheels, and the rear set is Orientation wheels; the mechanical arm is used to release the detection device to be dragged behind the chassis, or retract the detection device to the support platform.

As a preferred embodiment, the unmanned inspection machine is equipped with a side-mounted platform and a bottom-mounted platform, and an infrared camera and a visible light camera are mounted on the side-mounted platform; a laser scanning device is mounted on the bottom-mounted platform; The second positioning device and the ranging device are mounted on the carrier frame of the drone.

As a preferred implementation, the coordinates of the outline of the abnormal area are extracted according to the pseudo-color image accumulated from the electromagnetic wave detection data.

As a preferred embodiment, it also includes, after mapping the coordinates of the abnormal area to the dam model, linking the visible light image data obtained by the unmanned patrol machine and the unmanned detection vehicle by marking the coordinates, and combining the visible light image data to assist the judgment of the disease.

As a preferred implementation, it also includes, based on the dam model, judging the degree of dam disease risk risk:

①Level 1 disease risk: Deformation diseases directly displayed by the dam model constructed by the point cloud, including landslides, surface cracks, and collapses;

②Secondary risk: Leakage disease on the backwater slope of the dam identified by infrared data, and within the straight-line distance of the horizontal projection of the diseased area <50m, there is one or more hidden dangers identified by unmanned detection vehicles, including dams The interior is rich in water and not dense;

③Third-level disease risk: Hidden dangers identified by unmanned detection vehicles, including rich water and lack of compaction inside the embankment; and no obvious leakage diseases were found by unmanned inspection machines on the backwater slope;

When a section of embankment has two or more levels of risk at the same time, it shall be judged according to the highest level of risk.

The air-ground integrated equipment for hidden disease diagnosis of dams of this application can realize the integrated control of the whole process of multi-source heterogeneous data through the data interaction of unmanned detection vehicles and unmanned inspection machines, and realize geophysical exploration, The rapid association of data such as laser scanning and images, the mapping base is established by extracting the coordinates of morphological control points, combined with the coordinate labels of each data to realize the visual display of hidden dangers, which can quickly and accurately evaluate the degree of danger of hidden dangers, and realize emergency rescue Precise allocation of resources.

Description of drawings

Figure 1 is a structural schematic diagram of the vehicle-mounted platform and the command shelter mounted on the vehicle-mounted platform.

Figure 2 is a schematic diagram of the structure of the unmanned probe vehicle.

Figure 3 is a schematic diagram of the structure of the unmanned inspection machine.

Figure 4 is a schematic diagram of data interaction between the unmanned detection vehicle and the unmanned inspection machine.

Fig. 5 is a schematic diagram of the working state of the air-ground integrated equipment for hidden disease diagnosis of dams.

Fig. 6 is a schematic diagram of a three-dimensional elevation model of an embankment established based on point cloud data.

Fig. 7 is a schematic diagram of infrared thermal imaging information of an unmanned inspection machine.

Figure 8 is a pseudo-color map of electromagnetic wave signals.

Fig. 9 is a schematic diagram of visually displaying hidden danger areas on the data mapping base.

Detailed ways

Example 1

This embodiment specifically illustrates the equipment structure and work flow of the present invention.

An air-ground integrated equipment for dam hidden disease diagnosis, including a vehicle-mounted platform and a command cabin mounted on the vehicle-mounted platform, an unmanned detection vehicle (at least one) and an unmanned patrol aircraft (at least one).

1. Function description

The command shelter is the core carrier of the whole set of equipment, the on-site operation center for on-site operations, the communication hub and the data interaction platform. The command cabin can provide functions such as charging, maintenance and long-distance maneuvering for unmanned detection vehicles and unmanned inspection aircraft, remotely command unmanned detection vehicles and drones to carry out on-site operations, analyze and process detection data, and make on-site operations and detection The data is communicated with the command center in real time to realize on-site diagnosis and response linkage of hidden dangers of embankments.

The unmanned probe vehicle is equipped with a stepping electromagnetic wave detection device, a video recording device (dual light camera) and a first positioning device. Data information such as detection, image and positioning can be transmitted back to the command shelter to realize high-precision detection of hidden dangers inside the embankment.

The unmanned inspection machine is equipped with optical equipment such as a laser scanning device and an infrared-visible light integrated camera. It can obtain embankment point cloud data to find deformation hazards such as landslides, bank collapses, and subsidences; it can obtain temperature field information of backwater slopes to find hidden dangers such as water leakage points.

2. Structural composition

(1) Command shelter

The command shelter includes: chassis, support frame, shelter. The chassis is a high-mobility chassis, which is the basic carrier of the whole set of equipment; the support frame is installed under the carriage, and can lift the command shelter by raising it to stabilize the shelter and form a temporary on-site control center and data transmission hub; The shelter cabin is divided into equipment room and operation room.

The equipment room is divided into upper and lower floors. The unmanned detection vehicle is placed on the lower floor, and the unmanned inspection machine is placed on the upper floor. There are charging holes on the upper and lower floors, equipped with cleaning tools, which can be used for charging and cleaning the unmanned detection vehicle and the unmanned inspection machine. . The top of the cabin of the equipment room can be put into the top of the operation room, and the sides and rear cabins can be unfolded outwards to form a passage between the upper and lower equipment of the unmanned detection vehicle; after the cabin is unfolded, there is no shelter above the equipment room, and the upper layer is used for unmanned inspection. aircraft apron. There are induction devices in the installation position of the equipment in the equipment room.

The operating room is equipped with an operating console (control terminal), a display device, a server, and a first communication device. The functions and logical relations of each device are as follows: The operating console includes the operating lever/key of the unmanned detection vehicle and the unmanned inspection machine, which can operate unmanned detection Vehicles and unmanned patrol machines patrol operations, and the console can also plan trajectories through electronic maps, and operate unmanned detection vehicles and unmanned patrol machines to perform automatic inspections; the display device can display the data returned by unmanned detection vehicles and unmanned patrol machines Real scene, detection/scanning data, electronic map and other pictures; the server is used to store and transfer the real scene, detection/scanning data, location information and other data returned by unmanned detection vehicles and unmanned patrol machines; the first communication device can send Operation instructions are sent to unmanned patrol aircraft and unmanned detection vehicles, and information such as sound, picture, and data can also be transmitted through mobile networks and satellite networking. The operating console sends the operation instructions to the unmanned detection vehicle and the unmanned inspection machine through the first communication device to control the operation of the two; , The third communication device sends back to the shelter and stores in the server.

In addition, auxiliary facilities such as warning lights and night lights are installed on the cabin body.

(2) Unmanned detection vehicle

The unmanned exploration vehicle includes: a chassis, a power supply device, a mobile control device, a second communication device, a support platform, a detection device, a first positioning device, and a video recording device.

The chassis is a crawler chassis, and there is an equipment cabin inside the chassis, where a power supply device and a mobile control device are placed; a support platform is arranged on the upper part of the chassis, and the support platform is divided into upper and lower layers. device and the second communication device; the lower part is equipped with a mechanical arm and a detection device, and the base of the detection device is equipped with two groups of four wheels, the front group is a universal wheel, and the rear group is a directional wheel. The robotic arm can retract the detection device to the chassis, and the detection device can be placed behind the chassis and dragged during work; the first positioning device is installed above the support platform; the recording device includes 2 dual-light cameras, which are respectively installed in front of and behind the support platform .

The functions and logical relationship of each device are as follows: the chassis is the base of the unmanned detection vehicle, and has functions such as forward, backward, and turning; the power supply device is a rechargeable battery pack, which supplies power for each device of the unmanned detection vehicle; The movement of the upper part of the chassis and the rear of the chassis; the communication device is used to receive the operation instructions from the command cabin, and send the position information, detection data and audio and video of the unmanned patrol vehicle; the bracket platform is used to expand the space for equipment placement; Electromagnetic waves are used to obtain the medium information below the inspection path, which can detect hidden dangers such as water-rich, uncompacted, and hollow anomalies; the first positioning device obtains the spatial position coordinates through satellite positioning; the video recording device records front and rear images. The communication device receives the operation instructions from the on-board command cabin, controls the chassis to move forward, backward, and turn through the mobile control device, and controls the detection device to move above and behind the chassis; the detection device, positioning device, and video recording device will detect data, position Coordinates and real-time images are transmitted to the control terminal through the second communication device; the power supply device supplies power to each electrical device of the unmanned detection vehicle.

(3) Unmanned inspection machine

The unmanned inspection machine includes a drone carrier, a suspension platform, an infrared-visible light integrated camera, a laser scanning device, a second positioning device, a distance measuring device, and a third communication device. The UAV carrier is a rotor UAV; the suspension platform includes a side-mounted platform and a bottom-mounted platform, wherein the side-mounted platform is mounted with an infrared-visible light integrated camera, and the bottom-mounted platform is mounted with a laser scanning device; The second positioning device and the ranging device are mounted on the UAV carrier frame; the command cabin (vehicle terminal) can command the UAV inspection operation by sending instructions.

The functions and logical relationship of each device are as follows: the UAV carrier is the basis for the unmanned inspection aircraft, and can also supply power for each electrical device; the side-mounted pan/tilt is mounted on the side of the UAV carrier, and the bottom-mounted pan/tilt is mounted on the Below the UAV, the two gimbals can rotate 360° and have the function of shock absorption; the infrared-visible light camera can record infrared thermal imaging images and visible light images in real time; the laser scanning device can scan the embankment point cloud map; the third positioning device The absolute coordinates of the UAV can be obtained through satellite positioning; the distance measuring device can measure the distance between the UAV and the embankment; the third communication device can send the UAV positioning, scanning and recording data back to the vehicle command cabin; The command cabin can control the flight of the UAV and the work of the mounted devices, and can use the electronic map for route planning.

3. Workflow

(1) Carry unmanned detection vehicles and unmanned patrol aircraft in the command cabin to the periphery of the embankment section to be detected, and select an area that is relatively open and free of communication signal interference to deploy the equipment.

1) Raise the support frame to prop up the command shelter.

2) The top of the cabin of the equipment room is retracted into the top of the operation room, and the sides and rear cabins are expanded outward.

(2) Start all equipment and devices. Check whether the equipment is abnormal and confirm the status of the equipment. Check whether the power of the unmanned detection vehicle is sufficient and whether the devices are normal; check whether the power of the unmanned inspection machine is sufficient, whether the carrier of the drone is intact, and whether the functions of each device are normal; check whether the communication link of the command cabin is complete. After the above problems are confirmed, proceed to the next step.

(3) Inspection route planning. In the operating cabin, plan the route of the unmanned patrol drone by calling the electronic map, or use the joystick to manually control the flight route. This is because the embankment is located in the wild, there are few obstacles in the air, and the UAV is equipped with an obstacle avoidance device (the UAV comes with it), which can plan the flight.

(4) Inspection work. Start the unmanned inspection aircraft to make it inspect the embankment according to the flight route. The display device in the operation room can display the electronic map navigation route of the unmanned inspection aircraft, the image taken by the pan-tilt camera of the unmanned inspection aircraft, the image captured by the infrared camera of the unmanned inspection aircraft, and the laser radar scanning data information of the unmanned inspection aircraft. The returned data is stored in the server in real time. In addition, according to the point cloud data and images sent back by the unmanned inspection machine to confirm the working condition of the embankment, avoid areas with dense vegetation and surface collapse, and plan the travel route of the unmanned detection vehicle. The unmanned detection vehicle conducts embankment inspection according to the planned route, and the display device in the operation room can display the electronic map walking route of the unmanned detection vehicle, the image taken by the front pan-tilt camera of the unmanned detection vehicle, and the rear view of the unmanned detection vehicle. The images captured by the PTZ camera and the data information returned by the detection equipment.

(5) Return to the home position. The unmanned patrol aircraft returns to the equipment cabin, lands on a fixed position on the apron, and is automatically charged by the induction charging device. Before the unmanned detection vehicle returns to the equipment cabin, the detection device retracts above the chassis, and the unmanned detection vehicle returns to the equipment cabin, and the sensing device will automatically charge it.

(6) Equipment maintenance. Wipe the lens, check if there is any abnormality in the equipment, and check immediately if there is any abnormality.

(7) Data processing. With the help of the server in the equipment cabin, the data returned by the detection is processed, and the results are printed on site or transmitted to the remote end through the communication system. When the data is incomplete or in doubt, repeat steps (3) to (7) for supplementary detection and analysis.

(8) Equipment closed. Close the equipment compartment, retract the support frame, and the vehicle equipment goes to the next section for detection.

The logic relationship of equipment control terminal data collection and interaction:

(1) Acquisition layer: The unmanned detection vehicle and unmanned inspection machine send the detection/scanning/recording/positioning data back to the command shelter through the second and third communication module devices in real time, and store them in the server.

(2) Collaboration layer: The unmanned inspection machine acquires data first, and can determine the vegetation and external subsidence according to the point cloud data obtained by it, and avoid obstacles, and plan the walking route of the unmanned detection vehicle. Unify the data obtained by the unmanned detection vehicle and the unmanned inspection machine through the position coordinates.

(3) Label layer: classify and segment the acquired data, and mark the type label and coordinate label.

Specific operations include:

Classification: by data source, including point cloud data, visible light image data, electromagnetic wave detection data, and infrared image data.

Segmentation: According to the length of the coordinates, the electromagnetic wave detection data, visible light image data, and infrared image data are divided according to the length obtained in each 100m embankment section.

Coordinate label: Obtain visible light image data, electromagnetic wave detection data, and infrared image data through coordinate matching. Subsequent data is mainly traced through coordinates.

(4) Extraction layer: Call the point cloud data in the server, and extract the coordinates of the shape control points (points with sudden changes in elevation) such as the front and rear slope toes, horse paths, and embankment crests based on the image edge algorithm. Call the infrared camera recording data and electromagnetic wave detection data in the server, identify abnormal areas of temperature and electromagnetic waves based on machine learning algorithms, and extract the coordinates of abnormal areas.

(5) Modeling layer: build a visual model based on the extracted point cloud data.

(6) Mapping layer: Map the extracted abnormal area coordinates in different colors/lines according to the coordinate labels in the visualization model.

(7) Access layer: In the visualization model, if you want to obtain the real scene image, you need to access the corresponding image segment in the server according to the coordinate label.

Equipment control terminal data processing process:

(1) The laser scanning device of the unmanned inspection machine obtains the point cloud (including absolute three-dimensional coordinate information) of the embankment surface, establishes the elevation model of the embankment based on the point cloud data, and extracts the shape control points such as the front and rear slope feet, horse paths, and embankment crests of the embankment based on the image edge algorithm coordinate of.

(2) A simplified 3D model of the embankment is formed according to the coordinates of the morphological control points, which serves as the data base for the mapping of abnormal regions.

(3) Obtain the image of the abnormal temperature area on the surface of the backwater slope of the embankment through the infrared thermal imaging information of the unmanned inspection machine, and extract the coordinates of the outline of the abnormal area in the image through the temperature threshold. Extreme value area.

(4) The electromagnetic wave signal obtained by the unmanned detection vehicle is used to extract the coordinates of the outline of the abnormal area according to the pseudo-color map of the electromagnetic wave signal.

(5) Correlation coordinate labels of real-world images (visible light images) obtained by unmanned detection vehicles and unmanned inspection aircraft.

(6) On the data mapping base, the coordinates extracted from (2) and (3) are displayed in different colors, so that hidden danger areas can be quickly visualized.

(7) Both the data mapping base and the segmented image data contain coordinate labels, through which the real scene data stored in the server can be accessed.

Based on the data base, the degree of danger of the dam is judged.

①First-level disease risk: The deformation-type disease directly displayed on the data base constructed by the point cloud, including landslides, surface cracks, and subsidence. This kind of risk indicates that the dam has formed a structural dominant disease due to reasons such as uneven settlement and the development of internal hidden dangers.

②Second-level disease risk: Leakage disease on the backwater slope of the dam identified by infrared data, and within the straight-line distance of the horizontal projection of the diseased area <50m, there is one or more rich water inside the dam identified by the unmanned detection vehicle , Not dense and other hidden dangers. Such dangers indicate that there is a leakage channel inside the dam in this area, and it has escaped from the backwater slope.

③Third-level disease risk: hidden dangers such as rich water and loose density inside the embankment identified by the unmanned detection vehicle, and no obvious leakage disease was found by the unmanned inspection machine on the backwater slope. Such risks indicate that there are hidden dangers such as lack of density and looseness inside the dams in this area, but no obvious leakage channels have yet been developed.

When a section of embankment has two or more levels of risk at the same time, it shall be judged according to the highest level of risk.

Classification of disease risk levels: the first-level disease risk level is higher than the second-level disease risk level, and the second-level disease risk level is higher than the third-level disease risk level.

Example 2

This embodiment provides a specific example of equipment composition and workflow.

(1) Command shelter

In this embodiment, the command shelter chassis 1-1 is a Dongfeng flat-head 6×6 off-road truck chassis; the support frame 1-2 uses two hydraulic legs; the cabin body 1-3 is made of stainless steel pipes as the skeleton, and aluminum alloy plates as the shell , custom-made composition with built-in insulation panels. The square cabin is divided into an operation room and an equipment room. The operation room is embedded with 4k high-definition display screens 1-4, operation desks and cabinets 1-5, seats 1-6, and servers 1-7 are installed in the cabinets for data storage and analysis. ; 4G full Netcom routers 1-8 are installed in the cabin, and are connected with servers 1-7 to realize remote communication. Figure 1.

The equipment room is divided into upper and lower floors. The unmanned detection vehicle 2 is placed on the lower floor, and the unmanned inspection machine 3 is placed on the upper floor. There are 220V charging holes 1-9 on the upper and lower floors, and hair dryers and foam cleaners 1-10. The cabin roof of the equipment room is retractable through hydraulic guide rails, and the sides and rear cabins can be unfolded outwards to form a passage between the upper and lower equipment of the unmanned detection vehicle; Patrol aircraft apron. There are limited sensor devices 1-11 in the installation positions of the equipment in the equipment room. After the unmanned detection vehicle and the unmanned inspection machine are in place, they can perform charging, cleaning and other work. Warning lights 1-12 are installed on the sides of the shelter cabin body.

(2) Unmanned detection vehicle

The unmanned detection vehicle includes: chassis 2-1, power supply device 2-2, mobile control device 2-3, support platform 2-4, mechanical arm 2-5, detection device 2-6, video recording device 2-7, Two communication devices 2-8, and a first positioning device 2-9.

The chassis 2-1 is a crawler chassis, and there is an equipment cabin inside the chassis. The power supply device 2-2 and the mobile control device 2-3 are placed in the equipment cabin; It is divided into upper and lower floors. The video recording device 2-7, the second communication device 2-8, and the first positioning device 2-9 are installed on the top. The second communication device 2-8 uses a wifi communication antenna, and the first positioning device 2-9 is Hi-Target RTK V2 Equipment, video recording devices 2-7 use 2 sets of high-definition dual-light cameras, which are respectively installed in front and rear of the bracket platform. The mechanical arm 2-5 and the detection device 2-6 are installed below. The detection device includes a detection device 2-6-1 and a base. The base is equipped with two groups of four wheels before and after. The front group is a universal wheel 2-6-2. 1 group in the rear is directional wheel 2-6-3. The mechanical arm 2-5 can retract the detection device 2-6 onto the chassis, and the detection device can be placed behind the chassis to be dragged during work; the first positioning device is installed above the support platform. Figure 2.

(3) Unmanned inspection machine

The unmanned inspection machine includes a drone carrier 3-1, a suspension platform 3-2, an infrared-visible light integrated camera 3-3, a laser scanning device 3-4, a distance measuring device 3-5, and a second positioning device 3- 6. The third communication device 3-7. UAV carrier 3-1 uses 4-rotor UAV; suspension pan/tilt 3-2 includes side-mounted pan/tilt and bottom-mounted pan/tilt. The laser scanning device 3-4; the second positioning device 3-6 and the distance measuring device 3-5 are mounted on the UAV carrier frame; the vehicle-mounted command cabin can command the UAV inspection operation by sending instructions. Figure 3.

The data collection and interaction of the unmanned detection vehicle and the unmanned inspection machine are shown in Figure 4.

(1) Drive the whole set of equipment to the periphery of a certain embankment section 4 to be investigated (stake number K0+000~K5+000, totaling 5 kilometers), put the on-board command shelter 1 in place, raise the hydraulic outriggers, and unfold the equipment cabin body .

(2) Check whether the power of the unmanned detection vehicle and the unmanned inspection machine is sufficient, and whether each module is normal. The personnel in the operation room are in place, and the server and operation console are turned on.

(3) In the operating cabin, plan the travel routes of the unmanned detection vehicle and the unmanned patrol aircraft through the electronic map.

(4) Start the unmanned patrol aircraft and unmanned detection vehicle. The unmanned inspection aircraft automatically calibrates the take-off point as the landing location, and conducts flight inspection operations according to the planned route. The unmanned inspection aircraft collects laser point cloud data (dyke deformation), infrared image data (embankment surface temperature field), and visible light image data. and position information are sent back to the vehicle-mounted command shelter, displayed on the display device in real time, and stored in the server. According to the data sent back by the unmanned patrol aircraft, avoid areas that are difficult for unmanned detection vehicles such as pits, lush vegetation, slopes and steep slopes, and plan detection routes. The robotic arm of the unmanned detection vehicle is deployed, the detection device is dragged behind, and detection is carried out according to the planned walking route. The detection data, visible light images and location information collected by the unmanned detection vehicle are sent back to the command cabin, displayed on the display device in real time, and stored in the server. Figure 5.

(5) After the inspection, the unmanned inspection machine returns to the landing point, and the unmanned detection vehicle retracts the detection device through the mechanical arm and returns to the equipment cabin. The inductive charging device charges it automatically.

(6) Technicians use hair dryers and foam cleaners to clean equipment.

(7) Use the data interaction platform according to the returned data.

1) Segment the acquired data and mark the coordinate labels.

2) Based on the establishment of a three-dimensional model of embankment elevation based on point cloud data (Figure 6), the coordinates of shape control points such as front and rear slope toes and embankment crests are extracted based on the image edge algorithm.

3) Obtain the image of the temperature anomaly area on the surface of the backwater slope of the dam through the infrared thermal imaging information of the unmanned inspection machine (Figure 7), and extract the coordinates of the outline of the abnormal area in the image.

4) The electromagnetic wave signal obtained by the unmanned detection vehicle is used to extract the coordinates of the contour of the abnormal area according to the pseudo-color map of the electromagnetic wave signal (Figure 8).

5) A simplified 3D model of the embankment is formed according to the coordinates of the control points, which serves as the data base for the mapping of abnormal areas.

6) On the data mapping base, the areas with abnormal coordinates are displayed in different colors, so that hidden danger areas can be quickly visualized. (Figure 9)

7) The data mapping base and both contain coordinate tags, through which the real scene data stored in the server can be accessed. (8) Evaluate the embankment of this section according to the dike risk diagnosis criteria proposed in this application.

1) According to the 3D model of the embankment formed by laser scanning data, combined with the judgment of visible light photography images, there is a landslide on the back water surface of K3+650~K4+020 in the embankment section (K0+000~K5+000) explored this time. In the rest, no deformation risks such as surface cracks and subsidence were found. 2) According to the infrared thermal imaging information of the unmanned inspection aircraft, there is a leakage area near the position of mileage K1+035.

3) According to the detection data obtained by the unmanned detection vehicle, there are hidden dangers of leakage at the mileage K1+010~K1+035 and K1+060~K1+085.

4) According to the principles of disease risk classification proposed in this application:

K0+000~K1+000: no disease insurance.

K1+000~K1+100: Secondary disease insurance. Leakage disease on the backwater slope of the dam identified by infrared data, and within the linear distance of the horizontal projection of the diseased area <50m, there are two diseases such as rich water inside the dam identified by unmanned detection vehicles. Such dangers indicate that there is a leakage channel inside the dam in this area, and it has escaped from the backwater slope. It is comprehensively judged that this section of embankment is a second-class dangerous embankment section.

K1+100~K3+600: No disease risk.

K3+600~K4+100: Class I disease insurance.

K4+100~K5+000: no disease insurance.

(9) Equipment closed. Close the equipment compartment, retract the support frame, and the vehicle equipment goes to the next section for detection.

Claims (5)

1. A space-ground integrated equipment for hidden disease diagnosis of embankments, characterized in that it comprises a vehicle-mounted platform and a command shelter carried on the vehicle-mounted platform, and the command shelter includes a separate equipment room and an operation room; The cabin is composed of stainless steel pipes as the skeleton, aluminum alloy plates as the outer shell, and heat-insulating material panels embedded; the command shelter is fixed on the vehicle-mounted platform through a support frame with a jacking function; after the support frame is lifted, the command shelter The side panels and tail panels are deployed; A control terminal, a server, and a first communication device of the equipment in the equipment room are arranged in the operation room; Unmanned detection vehicles and unmanned inspection machines are placed in layers in the equipment room; The unmanned detection vehicle is equipped with a mobile control device, a detection device, a video recording device, a first positioning device and a second communication device; the unmanned patrol machine is equipped with an infrared camera, a visible light camera, a laser scanning device, and a second positioning device , distance measuring device, obstacle avoidance device, third communication device; The equipment workflow is as follows: The vehicle-mounted platform moves to the periphery of the embankment section to be detected, starts the unmanned inspection machine to make it inspect the embankment according to the flight route, and collects the shooting data, point cloud data, and synchronous The positioning data is sent to the server through the third communication device, and the control terminal determines the road conditions according to the point cloud data returned by the unmanned patrol machine, including vegetation, vegetation, and external collapse conditions to avoid obstacles, and plans the driving path of the unmanned detection vehicle. Start the unmanned detection vehicle to inspect the embankment according to the planned path, and send the electromagnetic wave detection data collected by the detection device, video recording device, and first positioning device, shooting data, and synchronized positioning data to the server through the second communication device; The control terminal classifies the data obtained by the server according to the data source, marks the coordinate label through coordinate matching, and divides the data according to the length according to the coordinate, including: The obtained data is divided into point cloud data, visible light image data, electromagnetic wave detection data and infrared image data according to the source; the coordinates of visible light image data, electromagnetic wave detection data and infrared image data are obtained through coordinate matching; electromagnetic wave detection data, visible light Image data and infrared image data are divided according to length; Call the point cloud data, extract the coordinates of the morphological control points based on the image edge algorithm, and establish the dam model; the morphological control points refer to the elevation mutation points of the dam, including the front and rear slope feet, horse paths, and embankment crests of the dam; use infrared camera shooting data and electromagnetic waves Detect data, identify temperature and electromagnetic wave abnormal areas, extract abnormal area coordinates, and map to the dam model; after mapping the abnormal area coordinates to the dam model, access the visible light image data obtained by unmanned patrol machines and unmanned detection vehicles through marking coordinates, combined with Visible light image data assists in dangerous diagnosis; The diagnosis results are judged based on the degree and type of sickness mapped in the embankment model: ①Level 1 disease risk: Deformation diseases directly displayed by the dam model constructed by the point cloud, including landslides, surface cracks, and collapses; ②Secondary risk: Leakage disease on the backwater slope of the dam identified by infrared data, and within the straight-line distance of the horizontal projection of the diseased area <50m, there is one or more hidden dangers identified by unmanned detection vehicles, including dams The interior is rich in water and not dense; ③Third-level disease risk: Hidden dangers identified by unmanned detection vehicles, including rich water and lack of compaction inside the embankment; and no obvious leakage diseases were found by unmanned inspection machines on the backwater slope; When a section of embankment has two or more levels of risk at the same time, it shall be judged according to the highest level of risk. 2. The equipment according to claim 1, wherein the control terminal plans the flight route of the unmanned patrol aircraft by calling the electronic map, or manually controls the flight route by using the joystick. 3. The equipment according to claim 1, wherein the unmanned detection vehicle is a crawler chassis, and there is an equipment compartment inside the crawler chassis for placing power supply devices and mobile control devices; the unmanned detection vehicle chassis The upper part is provided with a double-layer bracket platform, the upper layer of the bracket platform is equipped with a first positioning device, a video recording device and a second communication device, and the lower layer of the bracket platform is equipped with a mechanical arm and a detection device; two sets of wheels are installed at the bottom of the detection device, and the front set is a million Directional wheels, the rear group is directional wheels; the mechanical arm is used to release the detection device and drag it behind the chassis, or retract the detection device to the support platform. 4. The equipment according to claim 1, wherein the unmanned inspection machine is provided with a side-mounted pan-tilt and a bottom-mounted pan-tilt, and an infrared camera and a visible light camera are mounted on the side-mounted pan-tilt; The laser scanning device is mounted on the upper part; the second positioning device and the distance measuring device are mounted on the UAV carrier frame. 5. The equipment according to claim 1, characterized in that the coordinates of the outline of the abnormal area are extracted according to the pseudo-color image accumulated from the electromagnetic wave detection data.

Images