A bridge detection method, computer and device

Abstract

This invention discloses a bridge and culvert inspection method, comprising: acquiring reference point cloud data; acquiring target point cloud data; matching the target point cloud data and the reference point cloud data in the same coordinate system; comparing and contrasting the target point cloud data and the reference point cloud data in the same coordinate system to obtain difference point cloud data, wherein the difference point cloud data includes reference difference point cloud data and target difference point cloud data; analyzing and processing the reference difference point cloud data and the target difference point cloud data to generate analysis results; converting the reference point cloud data into a reference 3D model; converting the target point cloud data into a target 3D model; converting the reference 3D model into a reference VR panorama; and converting the target 3D model into a target VR panorama. Accordingly, this invention also discloses a computer and equipment related to the above method. Using this invention, there are advantages such as wide detection coverage, high security, high detection efficiency, and high accuracy.

Description

Technical Field

[0001] This invention relates to the field of bridge and culvert inspection technology, and in particular to a bridge and culvert inspection method, computer, and equipment. Background Technology

[0002] Bridge and culvert inspection includes two types of inspection items: bridge appearance inspection and culvert appearance inspection.

[0003] The inspection contents of bridge appearance inspection include: (1) inspecting the appearance quality of the bridge, including the bridge deck system, superstructure, bearings, substructure and other ancillary facilities, with a focus on checking whether there are cracks in the main load-bearing concrete structure; (2) checking whether there are honeycomb, pitting and exposed rebar in the concrete structure; (3) checking whether there are other appearance defects and diseases in the concrete; whether the bearing installation quality meets the requirements, and whether there are voids and abnormal deformation.

[0004] Traditional bridge inspection methods use bridge inspection vehicles as inspection platforms, relying primarily on visual observation, supplemented by inspection tools such as crack width measuring instruments, laser rangefinders, tape measures, feeler gauges, and digital cameras to conduct close-up inspections of the bridge.

[0005] The inspection contents of bridge appearance inspection include: (1) checking the water carrying capacity of the culvert; (2) checking whether the inlet and outlet paving, wing walls, slope protection, water retaining walls, etc. are intact, and whether the connection of the culvert opening is flat and smooth; (3) whether the culvert body and culvert walls are leaking, cracked, deformed or tilted, whether the mortar of the reinforcing body masonry has fallen off, whether the stones are loose, and whether the foundation has been eroded and hollowed out; (4) whether the culvert body, culvert top cover plate or culvert top is cracked, leaking, deformed and deflected; (5) whether the culvert bottom is silted up and blocking water, and whether the culvert bottom paving is intact; (6) whether the backfill near the culvert opening is leaking, eroded or hollow, and whether the backfill is stable; (7) whether the road surface above the culvert is cracked or subsided, and whether driving is safe.

[0006] Traditional culvert inspection methods use bridge inspection vehicles, mobile hanging baskets, ladders, etc. as inspection platforms, mainly relying on visual observation, and carrying inspection tools such as crack width measuring instruments, tape measures, and digital cameras to conduct close-up inspections of the culvert.

[0007] As can be seen from the above, traditional bridge and culvert inspection methods are mainly manual. However, the box ducts of large bridges are poorly lit, the box girders are too high, and the culverts are often poorly lit, have sewage residue, silt accumulation, and toxic gas flow. The inspection environment is extremely harsh, resulting in low inspection efficiency and high inspection risk. Safety accidents such as gas poisoning and falls from heights are likely to occur. Moreover, some dangerous areas cannot be accessed by inspectors, limiting the inspection scope and making it easy to misdiagnose or miss defects. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a bridge and culvert inspection method, computer and equipment, which has the advantages of wide inspection coverage, high safety, high inspection efficiency and high accuracy.

[0009] To address the aforementioned technical problems, this invention provides a bridge and culvert inspection method, comprising: acquiring reference point cloud data; acquiring target point cloud data; matching the target point cloud data and the reference point cloud data in the same coordinate system; comparing the target point cloud data and the reference point cloud data to obtain difference point cloud data, the difference point cloud data including reference difference point cloud data and target difference point cloud data; analyzing and processing the reference difference point cloud data and the target difference point cloud data to generate analysis results; converting the reference point cloud data into a reference 3D model; converting the target point cloud data into a target 3D model; converting the reference 3D model into a reference VR panorama; and converting the target 3D model into a target VR panorama.

[0010] As an improvement to the above scheme, the step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results includes: determining whether the three-dimensional coordinates in the target difference point cloud data are consistent with the three-dimensional coordinates in the benchmark difference point cloud data; if the determination is no, determining that the target area has a defect; if the determination is yes, determining whether the laser reflection intensity in the target difference point cloud data is less than a preset laser reflection intensity; if the laser reflection intensity in the target difference point cloud data is less than the preset laser reflection intensity, determining that the target area has water leakage; if the laser reflection intensity in the target difference point cloud data is greater than or equal to the preset laser reflection intensity, determining that the target area is normal.

[0011] As an improvement to the above scheme, the step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results further includes: when a defect occurs in the target area, acquiring the defect point cloud data in the target difference point cloud data; calculating the three-dimensional size of the defect based on the defect point cloud data; and comparing the three-dimensional size of the defect with a preset size to evaluate the defect level.

[0012] As an improvement to the above scheme, the step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results further includes: when water leakage occurs in the target area, acquiring the leakage point cloud data in the target difference point cloud data; calculating the leakage area based on the leakage point cloud data; and comparing the leakage area with a preset area to assess the leakage level.

[0013] As an improvement to the above scheme, the step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results further includes: generating an analysis report based on the judgment results and the three-dimensional dimensions of the defect, the defect level, the leakage area, and the leakage level.

[0014] As an improvement to the above scheme, the step of calculating the three-dimensional size of the defect based on the defect point cloud data includes: obtaining the point cloud reference plane of the defect point cloud data; dividing the point cloud reference plane into several reference plane grids; counting the number of grids in each row of reference planes and taking the maximum value to obtain a first length value; counting the number of grids in each column of reference planes and taking the maximum value to obtain a second length value; generating several cuboids based on each point cloud and its corresponding reference plane grid in the defect point cloud data; calculating the height value of each cuboid and taking the maximum value to obtain a depth value; calculating the volume of each cuboid and summing the volumes of all cuboids to obtain the defect volume.

[0015] As an improvement to the above solution, the step of calculating the leakage area based on the leakage point cloud data includes: obtaining the point cloud plane of the leakage point cloud data; dividing the point cloud plane into several planar grids; and counting the number of all planar grids to obtain the leakage area.

[0016] Accordingly, the present invention also provides a computer, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the bridge and culvert detection method.

[0017] Accordingly, the present invention also provides a bridge and culvert inspection device, including an inspection robot, a wireless remote control device, a VR display device, and a computer; the inspection robot is wirelessly connected to the wireless remote control device and the computer respectively, and the VR display device is connected to the computer and used to display the reference VR panorama and the target VR panorama; wherein, the inspection robot includes an intelligent mobile robot and a three-dimensional laser scanner, the three-dimensional laser scanner is mounted on the intelligent mobile robot and used to scan the bridge and culvert to generate the reference point cloud data and the target point cloud data.

[0018] As an improvement to the above solution, the intelligent mobile robot is a biomimetic quadruped robot or an amphibious robot; the intelligent mobile robot includes a body, a camera, a light source, a control module, and a wireless communication module. Motion mechanisms are located on both sides of the body. The control module and the wireless communication module are located within the body. The camera and the light source are located on the body. The camera is used to capture real-time images, and the light source is used for illumination. The wireless remote control device is equipped with a display for showing the real-time images. The control module is connected to the motion mechanisms, camera, light source, 3D laser scanner, and wireless communication module, respectively. The wireless communication module is wirelessly connected to the wireless remote control device and a computer, respectively.

[0019] The beneficial effects of implementing this invention are as follows:

[0020] In the bridge and culvert inspection equipment of this invention, inspectors can wirelessly control the inspection robot to enter the area to be inspected via the wireless remote control device, resulting in a wider inspection range and avoiding inspection risks caused by harsh environments, thus ensuring high safety. Secondly, the inspection robot scans the target area of ​​the bridge and culvert to generate corresponding target point cloud data. The computer compares the target point cloud data with pre-stored benchmark point cloud data to obtain difference point cloud data, processes and analyzes it to generate analysis results, resulting in high inspection efficiency and accuracy. In addition, the computer can convert the target point cloud data into a corresponding target VR panorama and transmit the target VR panorama to the VR display device. Inspectors can observe the pre-stored benchmark VR panorama and the target VR panorama through the VR display device for manual analysis, facilitating the evaluation and judgment of complex projects. Furthermore, the results of the computer analysis can be reviewed to further improve inspection accuracy. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the implementation of the bridge and culvert inspection method of the present invention;

[0022] Figure 2 yes Figure 1 The flowchart illustrates the implementation process of analyzing and processing benchmark difference point cloud data and target difference point cloud data to generate analysis results.

[0023] Figure 3 yes Figure 2 The implementation flowchart for calculating the three-dimensional dimensions of defects based on defect point cloud data;

[0024] Figure 4 yes Figure 2 The implementation flowchart for calculating the leakage area based on leakage point cloud data;

[0025] Figure 5 This is a schematic diagram of the bridge and culvert inspection equipment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

[0027] like Figure 1 As shown, Figure 1 The implementation process of a bridge and culvert inspection method is shown, including:

[0028] S101, acquire baseline point cloud data.

[0029] The point cloud includes three-dimensional coordinates and laser reflection intensity. The reference point cloud data is target point cloud data generated from previous detections, pre-stored in a corresponding memory, from which the processor retrieves the data.

[0030] S102, acquire target point cloud data.

[0031] The target point cloud data is generated by scanning the target area of ​​the bridge and culvert using a 3D laser scanner.

[0032] S103, Match the target point cloud data and the reference point cloud data in the same coordinate system.

[0033] S104, the target point cloud data is compared with the reference point cloud data to obtain the difference point cloud data, the difference point cloud data includes the reference difference point cloud data and the target difference point cloud data.

[0034] After the target point cloud data is superimposed on the reference point cloud data, point cloud data with equal three-dimensional coordinates and laser reflection intensity are removed to obtain the difference point cloud data.

[0035] S105, the benchmark difference point cloud data and the target difference point cloud data are analyzed and processed to generate analysis results.

[0036] S106, convert the reference point cloud data into a reference 3D model.

[0037] The reference point cloud data is converted into a reference 3D model using commercially available point cloud processing software (such as Bentley ContextCapture, BentleyPointools, and Bentley Descartes).

[0038] S107, the target point cloud data is converted into a target 3D model.

[0039] The processing procedure is the same as step S106.

[0040] S108, convert the benchmark 3D model into a benchmark VR panorama.

[0041] At least one origin is created in the reference 3D model; multiple interrelated visual directions are obtained by using the origin as the viewpoint base point and the latitude and longitude coordinate changes as the adjustment direction, resulting in a visual sequence; at least one original view is cropped from the 3D model according to each direction in the view sequence, resulting in a first view set; the corresponding original views in the first view set are pre-cropped according to the latitude and longitude coordinate information obtained from the view sequence, resulting in a second view set; the cropped views in the second view set are VR panoramic stitched together according to the corresponding latitude and longitude coordinate information to obtain a panoramic image; after uploading the panoramic image to the panoramic platform, a 360-degree reference VR panorama is automatically generated.

[0042] S109, convert the target 3D model into a target VR panorama.

[0043] The processing procedure is the same as step S108.

[0044] The baseline VR panorama and the target VR panorama are used for manual analysis.

[0045] It should be noted that in this method, the difference point cloud data is obtained by comparing the target point cloud data with the pre-stored benchmark point cloud data. After processing and analysis, the analysis results are generated, resulting in high detection efficiency and accuracy. Detection personnel can observe the benchmark VR panorama and the target VR panorama through the corresponding VR display device to perform manual analysis, which is convenient for evaluating and judging complex projects. Furthermore, the results of the above machine analysis can be reviewed to further improve the detection accuracy.

[0046] like Figure 2 As shown, specifically, step S105 includes:

[0047] S501, determine whether the three-dimensional coordinates in the target difference point cloud data are consistent with the three-dimensional coordinates in the reference difference point cloud data. If the determination is no, proceed to step S502; if the determination is yes, proceed to step S503.

[0048] S502, it is determined that a defect has occurred in the target area, and steps 506, S507 and S508 are executed.

[0049] When the three-dimensional coordinates in the target difference point cloud data are inconsistent with the three-dimensional coordinates in the reference difference point cloud data, the target area is very likely to have cracks, honeycomb surface, exposed reinforcement, detachment, deformation, or tilting. Therefore, it is determined that the target area has defects.

[0050] S503, determine whether the laser reflection intensity in the target difference point cloud data is less than the preset laser reflection intensity. If the determination is yes, proceed to step S504; if the determination is no, proceed to step S505.

[0051] When the three-dimensional coordinates in the target difference point cloud data are the same as the three-dimensional coordinates in the reference difference point cloud data, their laser reflection intensities are different.

[0052] S504, if water leakage is detected in the target area, proceed with steps S509, S410 and S411.

[0053] Because the color of the leaking area is dark (black, dark gray, or gray), the laser reflection intensity is low. When the laser reflection intensity in the target difference point cloud data is less than the preset laser reflection intensity, the target area is very likely to have leaked water. Therefore, it is determined that the target area has leaked water.

[0054] S505, the target area is determined to be normal.

[0055] S506, Obtain defect point cloud data from the target difference point cloud data.

[0056] S507, Calculate the three-dimensional dimensions of the defect based on the defect point cloud data.

[0057] The three-dimensional dimensions of the defect include its length, width, depth, and volume.

[0058] S508, compare the three-dimensional dimensions of the defect with the preset dimensions to assess the defect level.

[0059] The defect level is assessed by comprehensively comparing the length, width, height, and volume of the defect with the preset dimensions.

[0060] S509, Obtain the seepage point cloud data from the target difference point cloud data.

[0061] S510, calculate the leakage area based on the leakage point cloud data.

[0062] S511, compare the leakage area with the preset area to assess the leakage level.

[0063] S512, Based on the judgment result and the three-dimensional dimensions of the defect, the defect level, the leakage area and the leakage level, an analysis report is generated.

[0064] like Figure 3 As shown, specifically, step S507 includes:

[0065] S711, Obtain the point cloud reference plane of the defect point cloud data.

[0066] When there are defects on the wall, the projection of the defects on the wall is used as the reference plane for the point cloud.

[0067] S712, the point cloud reference surface is divided into several reference surface grids.

[0068] The point cloud reference surface is divided into several reference surface grids according to the set step size (area).

[0069] S713, count the number of grid cells in each row of reference planes, and take the maximum value to obtain the first length value.

[0070] The length of each row is obtained by multiplying the set step size by the number of reference plane grids in each row.

[0071] S714: Count the number of grid cells in each column of reference planes, and take the maximum value to obtain the second length value.

[0072] Of the first length value and the second length value, the larger value is taken as the length of the defect, and the other is taken as the width of the defect.

[0073] S715, generate several cuboids based on each point cloud and its corresponding reference surface mesh in the defect point cloud data.

[0074] When the defect is concave, the reference surface mesh is defined as the bottom surface, and the corresponding point cloud is defined as the top surface, thus generating a corresponding cuboid. When the defect is irregular, one reference surface mesh may correspond to two point clouds. The point cloud closer to the reference surface mesh is used as the bottom surface, and the other point cloud is used as the top surface, thus generating a corresponding cuboid.

[0075] S716 Calculate the height value of each cuboid, and take the maximum value to obtain the depth value.

[0076] The depth value is used as the depth of the defect.

[0077] S717, calculate the volume of each cuboid and sum the volumes of all cuboids to obtain the defect volume.

[0078] like Figure 4 As shown, specifically, step S510 includes:

[0079] S811, Obtain the point cloud plane of the leakage point cloud data.

[0080] The principle is similar to step S711.

[0081] S812, the point cloud plane is divided into several planar grids.

[0082] The principle is similar to step S712.

[0083] S813, count the number of all planar grids to obtain the leakage area.

[0084] The principle is similar to step S713.

[0085] Accordingly, the present invention also provides a computer 100, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described bridge and culvert detection method.

[0086] like Figure 5 As shown, Figure 5 A bridge and culvert inspection device is shown, including a computer 100, an inspection robot 200, a wireless remote control device 300, and a VR display device 400. The inspection robot 200 is wirelessly connected to both the wireless remote control device 300 and the computer 100. The VR display device 400 is connected to the computer 100 and is used to display a reference VR panorama and a target VR panorama. The inspection robot 200 includes an intelligent mobile robot 201 and a 3D laser scanner 202. The 3D laser scanner 202 is mounted on the intelligent mobile robot 201 and is used to scan the bridge and culvert to generate the reference point cloud data and the target point cloud data. In this embodiment, the VR display device 400 is preferably a VR headset or VR glasses.

[0087] It should be noted that inspectors can wirelessly control the inspection robot 200 to enter the area to be inspected via the wireless remote control device 300, resulting in a wider inspection range and avoiding inspection risks caused by harsh environments, thus ensuring high safety. Secondly, the inspection robot 200 scans the target area of ​​the bridge and culvert to generate corresponding target point cloud data. The computer 100 compares the target point cloud data with the pre-stored reference point cloud data to obtain the difference point cloud data, processes and analyzes it to generate analysis results, resulting in high inspection efficiency and accuracy. In addition, the computer 100 can convert the target point cloud data into a corresponding target VR panorama and transmit the target VR panorama to the VR display device 400. Inspectors can use the VR display device 400 to observe the pre-stored reference VR panorama and the target VR panorama for manual analysis, facilitating the evaluation and judgment of complex projects. For example, the defects can be identified as cracks, honeycomb pitting, exposed reinforcement, detachment, deformation, or tilting. Furthermore, the results of the computer analysis can be reviewed to further improve the inspection accuracy.

[0088] Specifically, the intelligent mobile robot 201 is a biomimetic quadruped robot (such as the existing BigDog robot) or an amphibious robot, capable of agile movement and suitable for various harsh environments. The intelligent mobile robot 201 includes a body, a camera 211, a light source 212, a control module 213, and a wireless communication module 214. Motion mechanisms 215 are located on both sides of the body. The control module 213 and the wireless communication module 214 are housed within the body. The camera 211 and the light source 212 are mounted on the body. The camera 211 is used to capture real-time images, and the light source 212 is used for illumination. The wireless remote control device 300 is equipped with a display for showing the real-time images. The control module 213 is connected to the motion mechanism 215, the camera 211, the light source 212, the 3D laser scanner 202, and the wireless communication module 214. The wireless communication module 214 is wirelessly connected to the wireless remote control device 300 and the computer 100.

[0089] In summary, the present invention has the following advantages:

[0090] (1) The testing personnel can wirelessly control the testing robot 200 to enter the area to be tested through the wireless remote control device 300, which has a wider testing range and can avoid the testing risks caused by harsh environments, thus ensuring high safety.

[0091] (2) The detection robot 200 scans the target area of ​​the bridge and culvert to generate corresponding target point cloud data. The computer 100 obtains the difference point cloud data by comparing the target point cloud data with the pre-stored reference point cloud data. After processing and analysis, the analysis results are generated. The detection efficiency is high and the detection accuracy is high.

[0092] (3) The computer 100 can convert the target point cloud data into a corresponding target VR panorama and transmit the target VR panorama to the VR display device 400. The inspection personnel can observe the pre-stored benchmark VR panorama and the target VR panorama through the VR display device 400 to perform manual analysis, so as to facilitate the evaluation and judgment of complex projects.

[0093] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for detecting bridges and culverts, characterized in that, include: Acquire baseline point cloud data; Acquire target point cloud data, wherein both the reference point cloud data and the target point cloud data include three-dimensional coordinates and laser reflection intensity; Match the target point cloud data and the reference point cloud data into the same coordinate system; The target point cloud data in the same coordinate system is compared with the reference point cloud data. The target point cloud data with the same three-dimensional coordinates and laser reflection intensity are removed from the reference point cloud data to obtain the difference point cloud data. The difference point cloud data includes the reference difference point cloud data and the target difference point cloud data. The baseline difference point cloud data and the target difference point cloud data are analyzed and processed to generate analysis results; The benchmark point cloud data is converted into a benchmark 3D model; Convert the target point cloud data into a target 3D model; Convert the benchmark 3D model into a benchmark VR panorama; Convert the target 3D model into a target VR panorama; The step of analyzing and processing the baseline difference point cloud data and the target difference point cloud data to generate analysis results includes: determining whether the three-dimensional coordinates in the target difference point cloud data are consistent with the three-dimensional coordinates in the baseline difference point cloud data; if the determination is no, determining that a defect exists in the target area; if the determination is yes, determining whether the laser reflection intensity in the target difference point cloud data is less than a preset laser reflection intensity; if the laser reflection intensity in the target difference point cloud data is less than the preset laser reflection intensity, determining that water leakage exists in the target area; if the laser reflection intensity in the target difference point cloud data is greater than or equal to the preset laser reflection intensity, determining that the target area is normal.

2. The bridge and culvert inspection method according to claim 1, characterized in that, The step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results further includes: When a defect occurs in the target area, the defect point cloud data is obtained from the target difference point cloud data; Calculate the three-dimensional dimensions of the defect based on the defect point cloud data; The three-dimensional dimensions of the defect are compared with preset dimensions to assess the defect level.

3. The bridge and culvert inspection method according to claim 2, characterized in that, The step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results further includes: When water seepage occurs in the target area, the seepage point cloud data is acquired from the target difference point cloud data; Calculate the leakage area based on the leakage point cloud data; The leakage area is compared with a preset area to assess the leakage level.

4. The bridge and culvert inspection method according to claim 3, characterized in that, The step of analyzing and processing the benchmark difference point cloud data and the target difference point cloud data to generate analysis results further includes: An analysis report is generated based on the judgment results, as well as the three-dimensional dimensions of the defect, the defect level, the leakage area, and the leakage level.

5. The bridge and culvert inspection method according to claim 2, characterized in that, The step of calculating the three-dimensional dimensions of the defect based on the defect point cloud data includes: Obtain the point cloud reference surface of the defect point cloud data; The point cloud reference surface is divided into several reference surface grids; Count the number of grid cells in each row of reference planes, and take the maximum value to obtain the first length value; Count the number of grid cells in each column of reference planes, and take the maximum value to obtain the second length value; Several cuboids are generated based on each point cloud and its corresponding reference surface mesh in the defect point cloud data. Calculate the height of each cuboid, and take the maximum value to obtain the depth value; Calculate the volume of each cuboid and sum the volumes of all cuboids to obtain the defect volume.

6. The bridge and culvert inspection method according to claim 3, characterized in that, The step of calculating the leakage area based on the leakage point cloud data includes: Obtain the point cloud plane of the leakage point cloud data; The point cloud plane is divided into several planar grids; The number of all planar grids is counted to obtain the leakage area.

7. A computer, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the bridge and culvert detection method according to any one of claims 1 to 6.

8. A bridge and culvert inspection device, characterized in that, Includes a detection robot, a wireless remote control device, a VR display device, and the computer as described in claim 7; The detection robot is wirelessly connected to the wireless remote control device and the computer respectively, and the VR display device is connected to the computer and used to display the baseline VR panorama and the target VR panorama; The detection robot includes an intelligent mobile robot and a 3D laser scanner. The 3D laser scanner is mounted on the intelligent mobile robot and is used to scan the bridge and culvert to generate the reference point cloud data and the target point cloud data.

9. The bridge and culvert testing equipment according to claim 8, characterized in that, The intelligent mobile robot is a biomimetic quadruped robot or an amphibious robot. The intelligent mobile robot includes a body, a camera, a light source, a control module, and a wireless communication module. The body has a moving mechanism on both sides. The control module and the wireless communication module are located inside the body. The camera and the light source are located on the body. The camera is used to capture real-time images, and the light source is used for illumination. The wireless remote control device is equipped with a display for displaying the real-time images. The control module is connected to the moving mechanism, camera, light source, 3D laser scanner and wireless communication module respectively. The wireless communication module is wirelessly connected to the wireless remote control device and computer respectively.

Images