Pipe network inspection integrated robot system

Abstract

A pipe network inspection integrated robot system belongs to the technical field of inspection robots. According to the technical scheme, the system comprises an inspection vehicle used for inspecting indoor pipelines and an inspection unmanned aerial vehicle used for inspecting high-altitude and outdoor pipelines; according to the inspection vehicle, a double-fork-arm independent suspension is connected with a vehicle beam; the side baffle is connected with the vehicle beam; the top-layer cover plate is connected with the side baffles; the depth camera and the battery compartment are arranged on the top-layer cover plate; the wheels are connected with the double fork arms in an independent suspension manner; the 3D laser radar and the vehicle-mounted mechanical arm used for conducting deep inspection on the pipeline group are arranged on the top layer cover plate. The beneficial effects of the utility model are that indoor pipelines are inspected through the inspection vehicle, high-altitude and outdoor pipelines are inspected through the inspection unmanned aerial vehicle, the depth camera carried at the tail end of the mechanical arm carried by the intelligent inspection vehicle is used for shooting a characteristic pattern of a deep area which is difficult to shoot and collect of a pipeline group, and no-dead-corner fault detection is carried out on the pipelines; and omnibearing and multi-angle inspection is realized.

Description

Technical Field

[0001] The utility model belongs to the technical field of inspection robots, and particularly relates to an integrated robot system for pipeline network inspection. Background Art

[0002] Pipes in the air-conditioning pipeline network are used to transport refrigerants, circulating water or air to transfer heat, adjust temperature and humidity inside buildings. In addition to various pipes, the pipeline network also includes valves, accessories and connectors, as well as related control and monitoring equipment, which are crucial for the normal operation of the entire air-conditioning system. Therefore, it is necessary to regularly inspect the pipeline network, and the inspection frequency requirement is high. Affected by environmental limitations and other factors, there are problems such as great difficulty in pipeline network inspection and small investigation intensity.

[0003] Currently, manual inspection is mostly used, which has problems such as high missed inspection rate, low safety, and great difficulty in data collection and analysis, resulting in low efficiency of pipeline network inspection. Using intelligent inspection robots can effectively avoid the above problems.

[0004] Currently, most pipeline network inspection robots are in the conceptual design stage and consider fewer constraint conditions, with problems in operation performance, and the technology of the overall structure and key modules still needs to be upgraded. At the same time, most inspection robots can only complete single inspection, for example, they can only perform single indoor pipe or single high-altitude or single outdoor pipeline inspection, and cannot timely investigate the pipeline network in complex environments.

[0005] Therefore, there is an urgent need for an efficient intelligent inspection robot system that can make autonomous decisions to timely investigate pipeline network problems in complex environments in all directions. Content of the Utility Model

[0006] In order to solve the problems existing in the above-mentioned prior art, the utility model provides an integrated robot system for pipeline network inspection, which can simultaneously and respectively perform comprehensive inspections on indoor pipes, high altitudes and outdoor pipelines.

[0007] To achieve the above object, the utility model provides the following technical solutions:

[0008] An integrated robot system for pipeline network inspection, comprising:

[0009] An inspection vehicle for inspecting indoor pipelines;

[0010] An inspection unmanned aerial vehicle for inspecting high altitudes and outdoor pipelines;

[0011] The inspection vehicle includes:

[0012] Double-fork arm independent suspension, connected to the vehicle beam;

[0013] Side baffle, connected to the vehicle beam;

[0014] The top cover plate, connected to the side baffle;

[0015] The depth camera and the battery compartment are arranged on the top cover plate;

[0016] The wheels are connected to the double-wishbone independent suspension;

[0017] The 3D lidar and the vehicle-mounted robotic arm for deep inspection of the pipeline group are arranged on the top cover plate;

[0018] The inspection drone includes:

[0019] The drone body, which is provided with a drone 3D lidar, a GPS locator and a flight control development board;

[0020] A number of brushless motors are connected to the drone body; each of the brushless motors 34 is respectively connected to its corresponding self-locking propeller;

[0021] The drone depth camera is connected to the drone body;

[0022] The upper computer main control development board is connected to the drone body.

[0023] Furthermore, the wheels are Mecanum wheels.

[0024] Furthermore, the number of the Mecanum wheels is 4.

[0025] Furthermore, the inspection vehicle further includes a main control development board, and the main control development board uses an STM32F103RCT6 microcontroller.

[0026] Furthermore, it further includes an upper computer main control development board, and the upper computer main control development board is a JETSON NANO B01 module.

[0027] Furthermore, the drone depth camera is a D435 depth camera.

[0028] Furthermore, the upper computer main control development board uses an ETSON ORIN NANO module.

[0029] Furthermore, the self-locking propeller is a 9450 self-locking propeller.

[0030] The beneficial effects of the present utility model:

[0031] Compared with the prior art, the integrated pipeline inspection robot system of the present utility model inspects indoor pipelines through an inspection vehicle, inspects high-altitude and outdoor pipelines through an inspection unmanned aerial vehicle (UAV), and uses a depth camera carried at the end of a robotic arm mounted on an intelligent inspection vehicle to take feature maps of areas that are difficult to photograph and collect in the deep layer of a pipeline group, performing fault detection on the pipelines without dead angles and achieving all-round and multi-angle inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present utility model, the present utility model will be described in detail below with reference to the drawings and specific embodiments. Obviously, the drawings in the following description are only some embodiments of the present utility model, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts. Among them:

[0033] Figure 1 Schematic diagram of the inspection vehicle structure of the present invention;

[0034] Figure 2 Schematic diagram of the inspection UAV structure of the present invention;

[0035] Figure 3 Schematic diagram of the vehicle-mounted robotic arm structure of the present invention;

[0036] Explanation of the reference numerals in the drawings: 11, main body of the robotic arm; 12, side baffle; 13, top cover plate; 14, depth camera; 15, wheels; 16, double-wishbone independent suspension; 17, vehicle beam; 18, battery compartment; 19, 3D lidar;

[0037] 21, 28 stepper motors; 22, planetary reducer 5:1; 23, 42 stepper motors * 48; 24, planetary reducer 20:1; 25, 42 stepper motors * 60; 26, planetary reducer 45:1; 27, 57 stepper motors * 64;

[0038] 31, UAV 3D lidar; 32, GPS locator; 33, flight control development board; 4, brushless motor; 5, UAV main body; 6, UAV depth camera; 7, upper computer main control development board. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] In order to make the objectives, technical solutions and advantages of the present utility model clearer, the present utility model will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present utility model and are not used to limit the present utility model. The following will be further described in conjunction with the attached Figures 1-3 for further illustration.

[0040] Embodiment 1

[0041] An integrated pipeline inspection robot system includes an inspection vehicle for inspecting indoor pipelines and an inspection drone for inspecting high-altitude and outdoor pipelines;

[0042] The inspection vehicle includes: a double-fork arm independent suspension 16 connected to a vehicle beam 17; side baffles 12 connected to the vehicle beam 17; a top cover plate 13 connected to the side baffles 12; a depth camera 14 and a battery compartment 18 arranged on the top cover plate 13; wheels 15 connected to the double-fork arm independent suspension 16; a 3D lidar 19 and a vehicle-mounted robotic arm 11 for deep inspection of a pipeline group arranged on the top cover plate 13;

[0043] The inspection drone includes: a drone main body 35 provided with a drone 3D lidar 31, a GPS locator 21 and a flight control development board 33; a plurality of brushless motors 34 connected to the drone main body 35; each of the brushless motors 34 is respectively connected to its corresponding self-locking propeller; a drone depth camera 36 connected to the drone main body 35; an upper computer main control development board 37 connected to the drone main body 35.

[0044] The wheels 15 are Mecanum wheels. The number of the Mecanum wheels is 4. The inspection vehicle further includes a main control development board, and the main control development board adopts an STM32F103RCT6 microcontroller. It further includes an upper computer main control development board, and the upper computer main control development board is a JETSON NANO B01 module. The drone depth camera 36 is a D435 depth camera. The upper computer main control development board 37 adopts an ETSON ORIN NANO module.

[0045] Embodiment 2

[0046] As a new embodiment or a supplement to Embodiment 1.

[0047] (1) Intelligent inspection system with autonomous decision-making

[0048] The present utility model designs a set of efficient intelligent inspection robot system capable of autonomous decision-making, and realizes comprehensive and efficient inspection of pipelines in complex environments by integrating binocular depth cameras, three-dimensional lidars, machine vision technologies and SLAM mapping technologies.

[0049] (2) Multi-robot cooperation

[0050] Through the collaborative work of the intelligent inspection vehicle and the drone, dual ground and high-altitude inspection work is provided for indoor pipelines, and comprehensive inspection work is provided for outdoor pipelines.

[0051] (3) Deep inspection through the vehicle-mounted robotic arm

[0052] The intelligent inspection vehicle uses a depth camera carried at the end of the robotic arm to capture feature maps of areas deep in the pipeline group that are difficult to photograph and collect, thereby performing fault detection on the pipeline without blind spots.

[0053] 3.1 Design of robot autonomous inspection system

[0054] This inspection robot system consists of two robot bodies, namely the intelligent inspection vehicle and the intelligent inspection drone. Tasks are allocated through the different characteristics of the two robots to realize a robot inspection system suitable for any pipeline inspection scenario.

[0055] First, the operator provides the robot inspection system with the inspection range and working mode. There are two working modes: the first is for indoor inspection, where the inspection smart car and the drone work together, and the second working mode is for outdoor inspection, where the drone performs the inspection alone.

[0056] First, for the first working mode, the intelligent inspection vehicle uses the binocular depth camera and 3D laser radar according to the control logic to quickly build a 3D map of the indoor pipeline (using the three-dimensional mapping method in the existing technology), builds a 3D point cloud model of the relative layout of the indoor pipeline and the indoor movable space through point cloud processing technology, plans the inspection path for the intelligent inspection vehicle, and uses the vehicle-mounted mechanical arm 11 to shoot the internal feature map of the pipeline group, and uploads it to the cloud server through the Internet technology to troubleshoot the pipeline through machine vision technology. Through logical control, all the above operations are completed autonomously by the robot inspection system without manual operation.

[0057] For the second working mode, the inspection drone will perform the same inspection operation according to the inspection range set in advance, so as to realize the comprehensive inspection of the entire indoor pipeline. For outdoor pipelines, the inspection drone of the utility model can also perform visual SLAM mapping through the binocular depth camera to inspect the outdoor pipelines.

[0058] 3.2 Inspection vehicle

[0059] In the mechanical structure design of the inspection vehicle, the chassis of the inspection vehicle is designed to be square, and a double wishbone independent suspension 16 system equipped with Mecanum wheels 15 is selected. Each wheel is connected through the suspension system, which reduces the mass of the suspension part, effectively mitigates the impact on the vehicle body, and improves the ground adhesion of the wheels. This design provides good vehicle body stability for the inspection vehicle when performing fault type identification and vehicle-mounted mechanical arms. The design fully considers the small space, narrow terrain, and complex pipelines of the inspection site, combined with the characteristics of Mecanum wheels 15 that can be translated horizontally, so that the inspection vehicle can adjust the inspection posture without turning. Compared with traditional ordinary wheel inspection vehicles, it can increase space utilization to 100%.

[0060] In the electronic control design part of the inspection vehicle, the present utility model selects the STM32F103RCT6 microcontroller as the main control development board of the inspection vehicle. The present utility model selects the JETSON NANO B01 as the main control development board of the upper computer. Its powerful AI computing ability and rich interfaces enable the inspection vehicle to autonomously plan paths through multi-sensor fusion for autonomous inspection. For the motor that drives the inspection vehicle to work, the present utility model selects a high-load planetary reduction Hall encoder motor with a reduction ratio of 1:72 to ensure the mobility of the inspection vehicle in case of complex terrains and accurately lock the position of the fault point.

[0061] For the bottom-layer control code of the inspection vehicle, considering that the inspection vehicle conducts inspection work with high-precision requirements in the factory and it is itself a highly closed-loop robot system, the present utility model integrates a variety of advanced algorithms (all existing control algorithms) in the bottom-layer control code, and transplants FreeRTOS (an open-source real-time operating system) to integrate complex and numerous algorithms and tasks. After writing the basic task list, FreeRTOS uniformly schedules the task handles, making the bottom-layer control code of the inspection vehicle overall reliable, efficient, and real-time.

[0062] The present utility model conducts forward kinematics and inverse kinematics of the Mecanum wheel chassis according to the actual parameters of the inspection vehicle body, and deduces the code parameters and motion control logic dedicated to the inspection vehicle of the present utility model. In the writing of the bottom-layer logic of the code, the present utility model uses the advanced theory of the PID controller, combines the feedback data of the Hall encoder at the end of the motor and the on-vehicle gyroscope, conducts closed-loop control of the speeds of the four wheels and the overall vehicle body speed, adjusts the PID parameters suitable for the inspection vehicle, and uses the bus protocol of the CAN controller of the STM32F103RCT6 to allocate and publish the current of the four wheels to control the rotation of the motor, preventing obvious speed differences in the rotations of the four wheels caused by the large current interference of the inspection vehicle's own circuit and external interference in the factory. At the same time, it also enables the inspection vehicle chassis to have a faster response speed to the upper computer instructions. At the same time, the present utility model also conducts relevant operations on the automatic parking calculation of the inspection vehicle through the PID controller in the code.

[0063] The present utility model adjusts the overall motion state of the inspection vehicle according to the acceleration, angular velocity and other parameters obtained by the MPU6050 gyroscope during the movement of the inspection vehicle to keep it stable. And in the code, according to the parameter information measured by the gyroscope, the data filtering and estimation process of the Kalman filtering algorithm is used to improve the state estimation accuracy of the inspection vehicle system by combining the measured value and the predicted value, reducing the influence of measurement noise on the estimation result.

[0064] In order to enable the inspection vehicle to perform high-precision automatic inspection work in a complex scenario such as a factory full of pipeline networks, the utility model fully combines artificial intelligence algorithms to equip the inspection vehicle with a ROS system, which is deployed on the NVIDIA JETSON NANO B01 development board to serve as the upper computer for obtaining the position information of the inspection vehicle and making motion decisions. It conducts serial communication with the STM32F103RCT6 development board, which serves as the lower computer and mainly controls the movement of the inspection vehicle chassis. A Unitree L1 3D lidar is installed on the inspection vehicle to obtain the 3D point cloud information of the inspection site. In the early stage of project implementation, the utility model uses the robot simulation system of ROS for simulation debugging and feasibility verification of the inspection vehicle. The inspection vehicle model is created through the URDF toolkit, the lidar sensor is simulated through Gazebo, and finally the point cloud data of the lidar on the inspection vehicle during inspection is displayed in Rviz. In order for the inspection vehicle to better perform path planning in the pipe network factory, the utility model uses the 3D laser SLAM algorithm to enable the inspection vehicle to perform SLAM mapping. The path planning algorithm encapsulated in the move_base function package is used to call the path planning node according to the actual parameters of the inspection vehicle, and in the Rviz environment, ROS subscribes to the global_costmap and local_costmap cost maps to enable the inspection vehicle to perform path planning.

[0065] 3.2.1 On-vehicle robotic arm

[0066] The robotic arm control system based on stepper motors is a complex system, mainly composed of three parts: mechanical structure, circuit control, and software control. The mechanical structure includes stepper motors, transmission devices, and robotic arms, and the movement of the robotic arm is achieved by driving the transmission device with stepper motors.

[0067] The circuit control part includes stepper motor drivers, sensors, and controllers. The speed of the stepper motor is real-time feedback by a magnetic encoder and fed back to the control system. After fitting the reduction ratio, the control system obtains the joint angular velocity of the robotic arm. It is used to achieve precise control of the stepper motor and detection of the position state of the robotic arm.

[0068] Spatial planning is carried out for the robotic arm. The motion space planning of the robotic arm refers to determining the appropriate joint angles or Cartesian coordinates of the robotic arm under a given task to achieve the required motion trajectory. In joint space planning and Cartesian space planning, Cartesian space planning is selected for analysis because joint space planning cannot accurately control the pose change of the end effector of the robotic arm. By setting the position and pose of the end effector as the target values and using the Jacobian matrix to calculate the joint angular velocity, the motion control of the robotic arm is achieved.

[0069] The specific implementation steps are as follows. First, set the target position and posture. Through the position detection transmitted back by the sensor, use the forward kinematic equation to solve the joint angles of the robotic arm, and obtain the current posture of the robotic arm. According to the analyzed joint angles of the current robotic arm, calculate the Jacobian matrix. Divide the path of the robotic arm into multiple discrete points, and then use the interpolation algorithm to calculate the joint angles of each point, so as to obtain the joint angle sequence of the entire path. According to the Jacobian matrix, target position and posture, inversely solve to obtain the joint angular velocity. Then output the angular velocity, convert it into a pulse signal to drive the motor, and reach the target position while maintaining the set posture.

[0070] To ensure the stable clamping of the end camera and the gripper, during the movement of the robotic arm, analyze its pose, lock and keep the end pose unchanged, so as to achieve the effect of smooth operation while maintaining the pose unchanged to make the camera and clamping stable.

[0071] During the operation, due to the complex and irregular changes in the movement environment of the equipment, the robotic arm needs to have the basic ability to avoid obstacles during the execution process. Detect obstacles through the radar sensor, convert them into spatial positions and store them. When performing spatial planning, convert them into workspace limitations. When analyzing obstacles, convert Cartesian coordinates into spatial joint point position information for spatial analysis, so as to achieve the function of avoiding obstacles. The overall path planning is carried out around the stable trajectory of the end pose and smooth trajectory to meet the operation requirements.

[0072] Considering the impact of movement obstacles on the robotic arm, the data is synchronized at a particularly short period. Since the force of the robotic arm is large, an accident is likely to affect or damage the overall equipment. If the operating current of the motor suddenly increases, control the robotic arm to move in the opposite direction, stop and wait for the environmental data to be updated before proceeding to the next step.

[0073] After kinematic analysis and calculation, the robotic arm can achieve smooth movement of the trajectory and stable movement of the end pose, so that the camera can maintain a stable direction of movement, which is convenient for obtaining environmental information. In addition, the system can also ensure that the end effector accurately reaches the target position, so as to achieve precise operation and control. The design and optimization of the entire system need to comprehensively consider knowledge in multiple aspects such as machinery, electricity, and software to achieve efficient, stable, and precise control effects.

[0074] For the situation of image acquisition around a cylinder. Take the center of the object cross-section as the coordinate of the end of the robotic arm, and take the distance from the center of the circle photographed by the camera as the length of the newly added connecting rod on the basis of its original end, that is

[0075] α6' = α6 + α

[0076] In this way, the end position is limited to the center coordinates. After the end attitude is parallel to the cross-section, the attitude of the camera will always be perpendicular to the outer surface of the circular pipe, which is convenient for obtaining images that meet the requirements. Considering that although three points on the circumference can cover the entire circle of the image, in order to make the image recognition more accurate and improve its accuracy, the present utility model uses six points to acquire images.

[0077] Different from the calculation of the working range of a normal robotic arm, to ensure that all positions around the pipeline can be photographed and the robotic arm will not touch the object when it needs to work on the back of the object. When the present utility model conducts Cartesian space planning, the skew line distance between each connecting rod and the pipeline axis needs to be solved.

[0078]

[0079] is the direction vector, is the vector determined by two fixed points on the straight line. The cross product of the two direction vectors determines the normal vector perpendicular to these two direction vectors. The projection of the vector formed by the two fixed points on the straight line onto this vector is the distance between the two straight lines.

[0080] If the distance from each connecting rod except the end to the pipeline axis satisfies di > r, it is considered that the robotic arm can reach the working position and will not touch the pipeline. To make it safer and more stable (perceiving that there will be a slight deviation in its radius), when restricting, it is necessary to restrict di > 2 / 3r.

[0081] The calculation of the working area is planned within the working area based on the pipeline radius sensed before work, so as to obtain the maximum distance when the end is located on the back of the pipeline. After the inspection vehicle reaches the working area, the end trajectory attitude of the six points on the pipeline circumference is planned, and the attitude is parallel to the cross-section. Then, by combining the target point and the limiting conditions after α6', the robotic arm can capture pictures that meet the requirements.

[0082] 3.3 Inspection UAV

[0083] In the electronic control design part of the inspection UAV, the present utility model selects the STM32F427VIT6 microcontroller as the flight control development board 33 of the inspection UAV.

[0084] This utility model selects JETSON ORIN NANO as the host computer main control development board 37. Its powerful AI computing ability and rich interfaces enable the inspection UAV to autonomously plan paths through the fusion of multi-sensors such as depth cameras and IMUs for autonomous inspection. For the motors that drive the inspection UAV to work, this utility model selects high-speed brushless motors 34 with a kv value of 1000kv and 9450 self-locking propellers to ensure the flexibility and stability of the inspection UAV in the face of complex terrains, and at the same time make the movement control of the inspection UAV more precise, capable of accurately locking the position of the fault point.

[0085] Considering that the inspection UAV conducts inspection work with high-precision requirements in the factory and it is itself a highly closed-loop robot system, this utility model integrates a variety of advanced algorithms (all algorithms in the prior art, directly used) in the underlying control code of the inspection UAV, and ported NuttX (an open-source real-time operating system) to integrate complex and numerous algorithms and tasks, and writes a main task for processing the main functions of the UAV and sub-tasks for processing pending functions, making the overall underlying control code of the inspection UAV reliable, efficient and real-time. This utility model predicts the future state of the UAV under a given control input through the kinematic forward solution based on the motor thrust and control input of the inspection UAV. In the writing of the underlying logic of the code, this utility model uses a cascade PID controller, combines the feedback data of gyroscopes, accelerometers, geomagnetic sensors, and barometers, and performs closed-loop control on the speeds of the four motors and the overall fuselage to adjust the PID parameters suitable for the inspection UAV, and uses the DShot protocol between STM32F427VIT6 and the electronic speed controller to adjust the power of the motor through digital signals to achieve precise control of the motor. This utility model adjusts the overall motion state of the inspection UAV according to the acceleration, angular velocity, magnetic heading and other parameters obtained by the gyroscope, accelerometer, geomagnetic sensor, and barometer during the movement of the inspection UAV to keep it stable. And in the code, according to the angular velocity information measured by the gyroscope, the acceleration information measured by the accelerometer, and the heading information measured by the geomagnetic sensor, etc., a Kalman filter or complementary filter is used to fuse the data of different sensors, thereby improving the accuracy and reliability of attitude estimation.

[0086] The utility model fully combines artificial intelligence algorithms to equip the inspection UAV with a ROS system, which is deployed on a JETSON ORIN NANO development board to serve as the upper computer for obtaining the position information of the inspection UAV and making motion decisions. It communicates with the STM32F427VIT6 development board, which serves as the flight controller for mainly controlling the power system of the inspection UAV, using the MAVLink protocol (a communication protocol designed for micro air vehicles). At the same time, by utilizing the distributed communication mechanism of ROS, the UAV and the unmanned vehicle can achieve efficient mutual transmission of other data such as position information and point cloud data. This data sharing is crucial for realizing the collaborative operation of the multi-robot system. In specific implementation, the message passing architecture can be designed through the publish / subscribe mode, enabling the UAV to publish the aerial perspective point cloud data it collects to specific topics, and the unmanned vehicle can subscribe to these topics to obtain information. Similarly, the unmanned vehicle can also send out its ground perspective position information for the UAV to use. In this way, the entire multi-robot system can achieve real-time sharing of information, thereby better coordinating actions, optimizing task allocation, and improving the overall operation efficiency. In addition, the message passing mechanism of ROS also supports setting message filters and buffers to ensure the reliability and stability of data communication. A D435 depth camera 36 is equipped on the inspection UAV so that it can capture the three-dimensional structure information around the machine. Combining the information of the GPS locator 32, the system can not only make accurate navigation decisions within a local range but also plan waypoints in the global coordinate system to achieve long-distance autonomous navigation. Using multi-sensor state estimation technology to fuse and process this information from different sources can further optimize the decision-making process of the system.

[0087] The above description is only the preferred specific implementation manner of the utility model, but the protection scope of the utility model is not limited thereto. Any person skilled in the art within the technical scope disclosed by the utility model, according to the technical solution and the inventive concept of the utility model, making equivalent replacements or changes should be covered within the protection scope of the utility model.

Claims

1. A pipe network inspection integrated robot system, characterized in that: include: Inspection vehicles used to inspect indoor pipelines; Inspection drones for inspecting high-altitude and outdoor pipelines; The inspection vehicle comprises: A double wishbone independent suspension (16) connected to a vehicle beam (17); A side baffle (12) connected to the vehicle beam (17); A top cover plate (13) connected to the side baffle plate (12); A depth camera (14) and a battery compartment (18) are arranged on the top cover plate (13); Wheels (15) connected to the double wishbone independent suspension (16); A 3D laser radar (19) and a vehicle-mounted mechanical arm (11) for performing deep inspection of a pipeline group are arranged on the top cover plate (13); The inspection drone comprises: The drone body (35) is provided with a drone 3D laser radar (31), a GPS locator (21) and a flight control development board (33); A plurality of brushless motors (34) are connected to the drone body (35); each of the brushless motors (34) is connected to a corresponding self-locking propeller; A drone depth camera (36) connected to the drone body (35); The host computer main control development board (37) is connected to the drone body (35).

2. The integrated pipeline network inspection robot system according to claim 1, characterized in that: The wheel (15) is a Mecanum wheel.

3. The integrated pipeline network inspection robot system according to claim 2, characterized in that: The number of the Mecanum wheels is 4.

4. The integrated pipeline network inspection robot system according to claim 1, characterized in that: The inspection vehicle also includes a main control development board, and the main control development board adopts an STM32F103RCT6 microcontroller.

5. The integrated pipe network inspection robot system according to claim 4, characterized in that: It also includes a host computer main control development board, and the host computer main control development board is a JETSON NANO B01 module.

6. The integrated pipe network inspection robot system according to claim 1, characterized in that: The drone depth camera (36) is a D435 depth camera.

7. The integrated pipeline network inspection robot system according to claim 1, characterized in that: The host computer main control development board (37) adopts ETSON ORIN NANO module.

8. The integrated pipe network inspection robot system according to claim 1, characterized in that: The self-locking propeller is a 9450 self-locking propeller.

Images