Microwave detection robot and detection system thereof

Abstract

The microwave detection robot comprises a first mounting seat, an electric telescopic rod is hinged to the first mounting seat, a second mounting seat is hinged to the output end of the electric telescopic rod, and machine body assemblies are mounted on the second mounting seat and the first mounting seat; the double-body assembly is designed, the robot is attached to a to-be-detected surface through airflow generated by a first impeller, negative pressure suction force is generated in a ring groove through a second impeller, and the negative pressure suction force is transmitted to the moving wheel assembly through a pipeline structure, so that the road holding force of the moving wheel assembly is enhanced, and the load capacity of the robot is improved; the angle between the two machine body assemblies is adjusted through the electric telescopic rods, so that the robot has the obstacle crossing ability, and the adaptability of the robot during operation in the complex environment is improved; the environment compensation module adopts an intelligent algorithm to compensate original microwave signals, so that the problem of low detection result precision caused by environmental factor interference is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of robot technology, in particular to a microwave detection robot and a detection system thereof. BACKGROUND

[0002] In recent years, with the aging of large infrastructure and the increasing complexity of service environment, the limitations of traditional detection methods in adaptability, efficiency and accuracy are increasingly prominent. In the fields of wind turbines, oil and gas pipelines, bridges, tunnels, nuclear power plants, high-altitude industrial equipment and special military fields, the safety monitoring and detection needs for non-metallic materials and structures are particularly urgent. The integration of microwave non-destructive testing technology and robot technology provides a new detection and maintenance solution for these scenarios. However, the existing related detection equipment still has the following shortcomings: 1. insufficient load capacity: the existing detection equipment mostly uses wall-climbing robots to carry detection components, and their adsorption and movement methods restrict the load performance. For example, robots that move using mechanical legs usually have suction cups at the end of the mechanical legs. During movement, the legs need to be frequently lifted, resulting in a decrease in the number of legs that can maintain adsorption with the surface to be detected, and the load capacity is weak. Robots that move using moving wheels are attached to the surface to be detected by air flow generated by a fan, and the air load capacity is slightly better than that of mechanical leg type, but still cannot meet the carrying needs of high-precision detection equipment; 2. lack of obstacle crossing ability: the existing wall-climbing robots using moving wheels are difficult to cross obstacles such as large protrusions and steps on the surface to be detected due to structural limitations, and have poor adaptability when working on complex non-metallic surfaces; 3. detection accuracy is affected by environmental interference: microwave detection is easily affected by environmental factors such as temperature, humidity and electromagnetic interference, and the existing equipment lacks effective environmental self-adaptive adjustment mechanism, which cannot eliminate these disturbances, resulting in low detection result accuracy. SUMMARY

[0003] The present application aims to provide a microwave detection robot and a detection system thereof to solve the problems raised in the background.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical solution: a microwave detection robot, comprising a first mounting seat, a first electric telescopic rod is hinged on the first mounting seat, a second mounting seat is hinged on the output end of the first electric telescopic rod, a machine body assembly is installed on each of the first mounting seat and the second mounting seat, a first connecting rod is fixedly connected to one of the machine body assemblies, a second connecting rod is fixedly connected to the other machine body assembly, and the second connecting rod is hinged on the first connecting rod, the machine body assembly comprises a shell, a through hole is formed in the shell, a bracket is fixedly connected in the through hole, a first motor is fixedly connected to the bracket, a first impeller is fixedly connected to the output end of the first motor, and the first impeller is arranged at the top end of the through hole.

[0005] Preferably, the first impeller is fixedly connected with a second impeller, the housing is provided with a ring groove at a position corresponding to the second impeller, and the second impeller is arranged in the ring groove.

[0006] Preferably, the lower surface of the housing is fixedly connected with a universal wheel.

[0007] Preferably, the outer walls of the housing are both provided with a moving wheel assembly, the moving wheel assembly comprises a guide rod, a fixed block, a spring, a sliding frame, a second motor, a wheel shell, a second air inlet, a mounting hole, a first fixed pipe, a third air inlet, a closed shell, a through groove, a second fixed pipe, a corrugated pipe, a wheel sleeve, an elastic ring, an elastic strip and a fourth air inlet, both ends of the guide rod are fixedly connected with the fixed block, and the fixed block is fixedly connected to the housing, the sliding frame is slidably connected to the guide rod, the spring is sleeved on the guide rod, one end of the spring is arranged on the sliding frame, and the other end is arranged on the fixed block, the second motor is fixedly connected to the sliding frame.

[0008] Preferably, the wheel shell is provided with a mounting hole, the first fixed pipe is sleeved in the mounting hole, the input end of the first fixed pipe is provided with the third air inlet, the output end of the first fixed pipe is fixedly connected with the second fixed pipe, the second fixed pipe is fixedly connected to the sliding frame, the output end of the second fixed pipe is fixedly connected with the corrugated pipe, and the output end of the corrugated pipe is fixedly connected to the first air inlet.

[0009] Preferably, the first fixed pipe is fixedly connected with the closed shell, and the closed shell is sleeved in the wheel shell, and the lower surface of the closed shell is provided with a through groove.

[0010] Preferably, the wheel shell is fixedly connected with the wheel sleeve, the wheel sleeve is uniformly provided with the elastic strip and the fourth air inlet, the elastic strip and the fourth air inlet are arranged at intervals, the housing is provided with the second air inlet at a position corresponding to the fourth air inlet, the wheel sleeve is fixedly connected with two elastic rings, and both ends of the elastic strip are fixedly connected to the two elastic rings.

[0011] A microwave detection robot detection system, comprising a controller and a cloud, the controller and the cloud are connected, the controller comprises a microwave signal transmitting module, a microwave signal receiving module, an environment data acquisition module, a main control module, a communication module, a positioning and navigation module and a power module, and the main control module is electrically connected with the microwave signal transmitting module, the microwave signal receiving module, the environment data acquisition module, the communication module, the positioning and navigation module and the power module.

[0012] Preferably, the cloud comprises a data preprocessing module, an environment compensation module, a defect identification module, a data storage module, an intelligent scheduling module, a fault diagnosis module, an intelligent feedback module and a remote control module, and the data storage module is in data connection with the data preprocessing module, the environment compensation module, the defect identification module, the intelligent scheduling module, the fault diagnosis module, the intelligent feedback module and the remote control module respectively.

[0013] Preferably, the environment compensation module is realized by using a BP neural network, and the defect identification module is realized by using a CNN neural network.

[0014] Compared with the prior art, the robot has the following advantages: the robot has a double-body assembly, the double-body assembly uses the airflow generated by the first impeller to make the robot adhere to the surface to be detected, the second impeller generates a negative pressure suction force in the annular groove, the negative pressure suction force is transmitted to the moving wheel assembly through the pipeline structure, so that the grip of the moving wheel assembly is enhanced, and the load capacity of the robot is improved; the electric telescopic rod is used to adjust the angle between the two body assemblies, so that the robot has the obstacle crossing ability, and the adaptability of the robot in complex environments is improved; the environment compensation module uses an intelligent algorithm to compensate the original microwave signal, so as to eliminate the problem of low detection result precision caused by environmental interference. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 It is a whole front view structural schematic diagram of the application;

[0016] Figure 2 It is a body assembly three-dimensional structural schematic diagram of the application;

[0017] Figure 3 It is a body assembly three-dimensional structural schematic diagram of the application from another angle;

[0018] Figure 4 It is a shell three-dimensional structural schematic diagram of the application;

[0019] Figure 5 It is a first impeller three-dimensional structural schematic diagram of the application;

[0020] Figure 6 It is a moving wheel assembly side view cut structure schematic diagram of the application;

[0021] Figure 7 It is a wheel sleeve three-dimensional structural schematic diagram of the application;

[0022] Figure 8 It is a controller structure block diagram of the application;

[0023] Figure 9 It is a cloud structure block diagram of the application;

[0024] Figure 10 System flowchart of the present application.

[0025] In the figure: 1, first mounting seat; 11, electric telescopic rod; 12, second mounting seat; 13, first connecting rod; 14, second connecting rod; 2, body assembly; 21, shell; 22, through hole; 23, support; 24, ring groove; 25, first air inlet hole; 26, first motor; 27, first impeller; 28, second impeller; 29, universal wheel; 3, moving wheel assembly; 31, guide rod; 32, fixed block; 33, spring; 34, sliding frame; 35, second motor; 36, wheel shell; 37, second air inlet hole; 38, mounting hole; 39, first fixed tube; 310, third air inlet hole; 311, closed shell; 312, through groove; 313, second fixed tube; 314, corrugated pipe; 315, wheel cover; 316, elastic ring; 317, elastic strip; 318, fourth air inlet hole; 4, controller; 41, microwave signal emitting module; 42, microwave signal receiving module; 43, environmental data acquisition module; 44, main control module; 45, communication module; 46, positioning and navigation module; 47, power module; 5, cloud; 51, data preprocessing module; 52, environmental compensation module; 53, defect identification module; 54, data storage module; 55, intelligent scheduling module; 56, fault diagnosis module; 57, intelligent feedback module; 58, remote control module. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0027] Please refer to the drawings in the embodiments of the present application Figure 1 - the drawings in the embodiments of the present application Figure 10The application provides a microwave detection robot, which comprises a first mounting base 1, a first electric telescopic rod 11 hinged to the first mounting base 1, a second mounting base 12 hinged to the output end of the first electric telescopic rod 11, and a body assembly 2 mounted on the first mounting base 1 and the second mounting base 12, wherein one of the body assemblies 2 is fixedly connected with a first connecting rod 13, the other body assembly 2 is fixedly connected with a second connecting rod 14, and the second connecting rod 14 is hinged to the first connecting rod 13, the body assembly 2 comprises a shell 21, a through hole 22 is formed in the shell 21, a support 23 is fixedly connected in the through hole 22, a first motor 26 is fixedly connected to the support 23, a first impeller 27 is fixedly connected to the output end of the first motor 26, and the first impeller 27 is arranged at the top end of the through hole 22, the first mounting base 1 and the second mounting base 12 are used for mounting the first electric telescopic rod 11, the first connecting rod 13 cooperates with the second connecting rod 14 to realize the hinging of the two body assemblies 2, the relative inclination angle of the two body assemblies 2 is adjusted through the extension and contraction of the first electric telescopic rod 11, so that the body assembly 2 close to the obstacle is raised to enable the body assembly 2 to roll over the obstacle, the first motor 26 on the support 23 drives the first impeller 27, airflow enters from the bottom of the through hole 22, the airflow generated by the first impeller 27 applies pressure to the direction of a detection surface, and the robot is attached to the detection surface, a second impeller 28 is fixedly connected to the first impeller 27, a ring groove 24 is formed in the shell 21 at the position corresponding to the second impeller 28, the second impeller 28 is arranged in the ring groove 24, first air inlets 25 are formed in the inner walls of the two sides of the ring groove 24, the ring groove 24 is used for accommodating the second impeller 28, the second impeller 28 is used for generating negative pressure suction force in the ring groove 24, and the first air inlets 25 are used for air intake, universal wheels 29 are fixedly connected to the lower surface of the shell 21, and the universal wheels 29 are used for providing auxiliary support for the shell 21, mobile wheel assemblies 3 are mounted on the outer walls of the two sides of the shell 21, and the mobile wheel assemblies 3 comprise guide rods 31, fixed blocks 32, springs 33, sliding frames 34, second motors 35, wheel housings 36, second air inlets 37, mounting holes 38, first fixed pipes 39, third air inlets 310, closed shells 311, through grooves 312, second fixed pipes 313, corrugated pipes 314, wheel sleeves 315, elastic rings 316, elastic strips 317 and fourth air inlets 318, the two ends of each guide rod 31 are fixedly connected with the fixed blocks 32, the fixed blocks 32 are fixedly connected to the shell 21, the guide rods 31 are slidingly connected with the sliding frames 34, the guide rods 31 are sleeved with the springs 33, one end of each spring 33 is arranged on the sliding frame 34, and the other end of the spring 33 is arranged on the fixed block 32, the second motors 35 are fixedly connected to the sliding frames 34, the wheel housings 36 are fixedly connected to the output ends of the second motors 35, the fixed blocks 32 are used for mounting the guide rods 31, the guide rods 31 are used for mounting the sliding frames 34 and the springs 33, the springs 33 are used for providing reset elastic force for the sliding frames 34, so that the sliding frames 34 have shock absorption capacity, the second motors 35 are used for driving the wheel housings 36, and the rotation of the wheel housings 36 can drive the robot to move.The wheel shell 36 is provided with a mounting hole 38, the first fixing pipe 39 is sleeved in the mounting hole 38, the input end of the first fixing pipe 39 is provided with a third air inlet hole 310, the output end of the first fixing pipe 39 is fixedly connected with the second fixing pipe 313, the second fixing pipe 313 is fixedly connected to the sliding frame 34, the output end of the second fixing pipe 313 is fixedly connected with the corrugated pipe 314, the output end of the corrugated pipe 314 is fixedly connected to the first air inlet hole 25, the mounting hole 38 is used for accommodating the first fixing pipe 39, the third air inlet hole 310 and the first air inlet hole 25 are used for air inlet, and the first fixing pipe 39, the second fixing pipe 313 and the corrugated pipe 314 are used for conveying airflow; the first fixing pipe 39 is fixedly connected with the closed shell 311, the closed shell 311 is sleeved in the wheel shell 36, the lower surface of the closed shell 311 is provided with a through slot 312, the closed shell 311 only takes in air through the through slot 312, and the suction force generated by the airflow is ensured to be large enough; the wheel shell 36 is fixedly connected with the wheel sleeve 315, the wheel sleeve 315 is uniformly provided with the elastic strips 317 and the fourth air inlet holes 318, the elastic strips 317 and the fourth air inlet holes 318 are arranged at intervals, the wheel shell 36 is provided with the second air inlet hole 37 at a position corresponding to the fourth air inlet hole 318, the wheel sleeve 315 is fixedly connected with two elastic rings 316, and the two ends of the elastic strip 317 are fixedly connected to the two elastic rings 316 respectively, the wheel sleeve 315 is made of elastic material and is used for enhancing the gripping force of the robot, the fourth air inlet hole 318 and the second air inlet hole 37 are used for air suction, the elastic ring 316 and the elastic strip 317 can form a seal between the surface to be detected and the wheel sleeve 315, and thus the adsorption effect is improved.

[0028] The application discloses a microwave detection robot detection system, which comprises a controller 4 and a cloud 5, the controller 4 is connected with the cloud 5, the controller 4 comprises a microwave signal transmitting module 41, a microwave signal receiving module 42, an environment data collecting module 43, a main control module 44, a communication module 45, a positioning and navigation module 46 and a power module 47, the main control module 44 is connected with the microwave signal transmitting module 41, the microwave signal receiving module 42, the environment data collecting module 43, the communication module 45, the positioning and navigation module 46 and the power module 47, the controller 4 transmits microwave signals to a surface to be detected by the microwave signal transmitting module 41, receives the microwave signals reflected by the microwave signal receiving module 42, the environment data collecting module 43 is used for collecting environment temperature, humidity and electromagnetic intensity data, the main control module 44 is used for controlling the operation of the robot, the communication module 45 is used for establishing communication with the cloud 5, the positioning and navigation module 46 is used for correcting the motion trajectory of the robot, and the power module 47 is used for supplying power to the robot; the cloud 5 comprises a data preprocessing module 51, an environment compensation module 52, a defect identification module 53, a data storage module 54, an intelligent scheduling module 55, a fault diagnosis module 56, an intelligent feedback module 57 and a remote control module 58, the data storage module 54 is connected with the data preprocessing module 51, the environment compensation module 52, the defect identification module 53, the intelligent scheduling module 55, the fault diagnosis module 56, the intelligent feedback module 57 and the remote control module 58, the data storage module 54 is used for storing data, the data preprocessing module 51 is used for preprocessing the obtained data, the environment compensation module 52 is used for correcting original microwave signals, the defect identification module 53 is used for identifying defects of the surface to be detected based on the corrected microwave signals, the intelligent scheduling module 55 is used for task allocation of multiple detection robots, realizes intelligent scheduling, the fault diagnosis module 56 is used for monitoring the operation parameters of each module in real time, identifying potential faults through big data analysis and sending an alarm, when the intelligent feedback module 57 detects abnormal data, firstly triggers a local rapid review process, carries out secondary detection, if the abnormality still exists, sends an alarm, the remote control module 58 is used for online adjustment of the motion and detection parameters of the robot, the environment compensation module 52 is realized by adopting a BP neural network, and the defect identification module 53 is realized by adopting a CNN neural network.

[0029] Based on the above, the advantages of the application are that when the application is used, the cloud 5 is connected with the controller 4, the controller 4 can control the movement of the robot by accessing the cloud 5 through the webpage or the mobile terminal, the controller 4 transmits the microwave signal to the surface to be detected by using the microwave signal transmission module 41, receives the original microwave signal reflected back by using the microwave signal receiving module 42, collects the environmental temperature, humidity and electromagnetic intensity data by using the environmental data collection module 43, the main control module 44 uploads the collected data and the data of the robot itself in real time to the cloud 5 through the communication module 45, the cloud 5 stores the received data by using the data storage module 54, pre-processes the obtained data by using the data preprocessing module 51, such as denoising, normalization and standardization processing, the environmental compensation module 52 corrects the original microwave signal, and the defect identification module 53 identifies the defects of the surface to be detected based on the corrected microwave signal, wherein the main control module 44 has a sensor calibration and self-checking function, automatically calibrates before each detection, ensures that the sensor output data is accurate and correct, the main control module 44 adopts a closed-loop feedback algorithm for the control of the driving mechanism, so as to realize accurate driving, the main control module 44 has a device self-protection function, when there is abnormal vibration, sharp temperature change or adsorption failure, the robot system automatically starts the emergency stop and protection program, the communication module 45 adopts the transmission layer security protocol to ensure data security, sets multiple data checking mechanisms to ensure that the data is complete and correct during transmission, configures local data caching and automatic retransmission mechanism to improve the system anti-packet loss capability, regularly carries out network communication security review, discovers and corrects potential hidden dangers in time, the positioning and navigation module 46 adopts inertial navigation combined with visual positioning, corrects the robot movement track in real time, eliminates the path error through the closed-loop control algorithm, and carries out real-time data feedback, the power module 47 supplies power for the robot, the intelligent scheduling module 55 is used for task allocation of multiple detection robots, realizes intelligent scheduling, the fault diagnosis module 56 monitors the running parameters of each module in real time, identifies potential faults through big data analysis and issues an alarm reminder, the intelligent feedback module 57 detects abnormal data first, triggers the local rapid review process for secondary detection, if the abnormality persists, an alarm is issued, the remote control module 58 is used for online adjustment of the robot movement and detection parameters, the cloud 5 also provides a man-machine interface to intuitively present the detection data in the form of charts, dynamic graphics and the like.During the operation of the robot, the first motor 26 on the support 23 drives the first impeller 27, the airflow enters through the bottom of the through hole 22, the airflow generated by the first impeller 27 exerts pressure in the direction of the surface to be detected, so that the robot is attached to the surface to be detected, the second impeller 28 on the first impeller 27 rotates to generate negative pressure suction in the annular groove 24, under the action of the suction, the gas enters the second air inlet hole 37 through the fourth air inlet hole 318, then enters the closed shell 311 through the through groove 312, then enters the first fixed pipe 39 through the third air inlet hole 310, enters the corrugated pipe 314 through the second fixed pipe 313, enters the annular groove 24 through the first air inlet hole 25, so that the suction force is generated at the bottom of the wheel cover 315, and the elastic ring 316 and the elastic strip 317 form a seal between the surface to be detected and the wheel cover 315, further improving the adsorption effect, wherein the fixed block 32 is used to install the guide rod 31, the guide rod 31 is used to install the sliding frame 34 and the spring 33, the spring 33 is used to provide the sliding frame 34 with a reset elastic force, so that the sliding frame 34 has a shock absorbing capacity, the second motor 35 is used to drive the wheel shell 36, and the rotation of the wheel shell 36 can drive the robot to move; the mounting hole 38 is used to accommodate the first fixed pipe 39; when the robot encounters an obstacle, the adsorption capacity of the closest body component 2 to the obstacle can be controlled to be reduced, that is, the rotating speed of the first motor 26 is reduced, and then the relative inclination angle between the two body components 2 is adjusted through the extension and retraction of the electric telescopic rod 11, so that the body component 2 close to the obstacle is raised, so that the body component 2 can roll over the obstacle, wherein the first mounting seat 1 and the second mounting seat 12 are used to install the electric telescopic rod 11, the first connecting rod 13 cooperates with the second connecting rod 14 to realize the hinging of the two body components 2, and the universal wheel 29 is used to provide auxiliary support for the shell 21; the environmental compensation module 52 is realized by using a BP neural network, the input features of the BP neural network are (T norm ,H norm ,E norm ), T norm is the normalized temperature, H norm is the normalized humidity, E norm is the normalized electromagnetic intensity, and the label value is the deviation value ΔS of the standardized true microwave signal and the test microwave signal; the BP neural network outputs the compensation amount The BP neural network training process is as follows: forward propagation: input environmental parameters, calculate the output compensation amount through the hidden layer

[0030]

[0031] Wherein, W1, W2 are weight matrix, b1, b2 are bias term, f1, f2 are hidden layer and output layer activation function, loss function adopts mean square error to measure the gap between predicted compensation and real compensation; Back propagation: update weight and bias through gradient descent method, minimize loss function, formula as follows:

[0032]

[0033] Wherein, W is weight, b is bias, η is learning rate; After completing the training, the original microwave signal is corrected by compensation:

[0034]

[0035] Wherein, S corr is the corrected microwave signal, S raw is the original microwave signal, is the compensation;

[0036] The defect recognition module 53 is realized by CNN neural network, and the time-frequency domain features of the corrected microwave signal are extracted by CNN to realize the classification and positioning of defects, as follows: signal time-frequency conversion:

[0037] The short-time Fourier transform is performed on the corrected signal S corr (t) to obtain the time-frequency matrix F(m,n):

[0038] F(m,n)=∑ t S corr (t)·w(t-nT)·e -j2πmt / N

[0039] Wherein, w() is window function, T is time step, m, n are frequency and time index respectively;

[0040] The CNN network structure is as follows: the input layer is 2D time-frequency graph F(m,n); The convolution layer extracts local features through convolution kernel, and the output of the lth convolution layer is:

[0041] C l (i,j)=σ(∑ p,q K l (p,q)·C l-1 (i+p,j+q)+b l )

[0042] Wherein, K l is convolution kernel, σ is activation function, C l-1 is the output of the previous layer;

[0043] The pooling layer reduces the dimension and retains the key features:

[0044] P l(i,j) = max p,q C l (i·s+p,j·s+q)

[0045] wherein s is a pooling step length;

[0046] The full connection layer and the output layer output defect category probability through a softmax function:

[0047]

[0048] wherein z k is an output score of the kth category, and C is a defect category number, such as no defect, crack, etc.

[0049] It is apparent for a person skilled in the art that the present application is not limited to the details of the above exemplary embodiments, but can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all respects as illustrative and not restrictive, the scope of the present application being defined by the appended claims rather than the above description, and it is intended to embrace all changes and modifications that fall within the meaning and scope of the equivalent elements of the claims. Any reference signs in the claims should not be considered as limiting the claims involved.

Claims

1. A microwave detection robot, comprising a first mounting base (1), characterized in that: An electric telescopic rod (11) is hingedly connected to the first mounting seat (1), and an output end of the electric telescopic rod (11) is hingedly connected to a second mounting seat (12). A fuselage assembly (2) is mounted on both the second mounting seat (12) and the first mounting seat (1). One fuselage assembly (2) is fixedly connected to a first connecting rod (13), and the other fuselage assembly (2) is fixedly connected to a second connecting rod (14), and the second connecting rod (14) is hingedly connected to the first connecting rod (13). The fuselage assembly (2) includes a shell (21), a through hole (22) is provided on the shell (21), a bracket (23) is fixedly connected in the through hole (22), a first motor (26) is fixedly connected to the bracket (23), and a first impeller (27) is fixedly connected to the output end of the first motor (26), and the first impeller (27) is arranged at the top of the through hole (22).

2. A microwave detection robot according to claim 1, characterized in that: A second impeller (28) is fixedly connected to the first impeller (27), an annular groove (24) is provided on the housing (21) at a position corresponding to the second impeller (28), and the second impeller (28) is arranged in the annular groove (24), and first air inlet holes (25) are provided on both sides of the inner wall of the annular groove (24).

3. The microwave detection robot according to claim 1, characterized in that: The lower surface of the housing (21) is fixedly connected with a universal wheel (29).

4. A microwave detection robot according to claim 3, characterized in that: The outer walls of both sides of the housing (21) are each provided with a moving wheel assembly (3), the moving wheel assembly (3) comprising a guide rod (31), a fixing block (32), a spring (33), a sliding frame (34), a second motor (35), a wheel housing (36), a second air inlet (37), a mounting hole (38), a first fixing pipe (39), a third air inlet (310), a closing shell (311), a through groove (312), a second fixing pipe (313), a bellows (314), a wheel sleeve (315), an elastic ring (316), an elastic strip (317), and a plurality of other components. 17) and a fourth air inlet (318), both ends of the guide rod (31) are fixedly connected to fixed blocks (32), and the fixed block (32) is fixedly connected to the housing (21), the guide rod (31) is slidably connected to a sliding frame (34), a spring (33) is sleeved on the guide rod (31), and one end of the spring (33) is arranged on the sliding frame (34), and the other end is arranged on the fixed block (32), a second motor (35) is fixedly connected to the sliding frame (34), and an output end of the second motor (35) is fixedly connected to the wheel housing (36).

5. The microwave detection robot according to claim 4, characterized in that: The wheel housing (36) is provided with a mounting hole (38), a first fixed tube (39) is sleeved in the mounting hole (38), a third air inlet (310) is provided at the input end of the first fixed tube (39), a second fixed tube (313) is connected and fixed to the output end of the first fixed tube (39), and the second fixed tube (313) is fixedly connected to the sliding frame (34), a bellows (314) is connected and fixed to the output end of the second fixed tube (313), and the output end of the bellows (314) is connected and fixed to the first air inlet (25).

6. The microwave detection robot according to claim 5, characterized in that: A sealing shell (311) is fixedly connected to the first fixed tube (39), and the sealing shell (311) is sleeved in the wheel housing (36). A through groove (312) is provided on the lower surface of the sealing shell (311).

7. The microwave detection robot according to claim 5, characterized in that: The wheel housing (36) is fixedly connected to a wheel sleeve (315), and elastic strips (317) and fourth air inlet holes (318) are evenly distributed on the wheel sleeve (315), and the elastic strips (317) and the fourth air inlet holes (318) are arranged at intervals. A second air inlet hole (37) is opened at a position corresponding to the fourth air inlet hole (318) on the wheel housing (36), and two elastic rings (316) are fixedly connected to the wheel sleeve (315), and two ends of the elastic strip (317) are respectively fixedly connected to the two elastic rings (316).

8. A microwave detection robot detection system, comprising a controller (4) and a cloud (5), characterized in that: The controller (4) establishes a data connection with the cloud (5), and the controller (4) includes a microwave signal transmitting module (41), a microwave signal receiving module (42), an environmental data acquisition module (43), a main control module (44), a communication module (45), a positioning and navigation module (46), and a power module (47), and the main control module (44) establishes an electrical connection with the microwave signal transmitting module (41), the microwave signal receiving module (42), the environmental data acquisition module (43), the communication module (45), the positioning and navigation module (46), and the power module (47).

9. The microwave detection robot detection system according to claim 8, characterized in that: The cloud (5) includes a data preprocessing module (51), an environmental compensation module (52), a defect recognition module (53), a data storage module (54), an intelligent scheduling module (55), a fault diagnosis module (56), an intelligent feedback module (57) and a remote control module (58), and the data storage module (54) establishes data connections with the data preprocessing module (51), the environmental compensation module (52), the defect recognition module (53), the intelligent scheduling module (55), the fault diagnosis module (56), the intelligent feedback module (57) and the remote control module (58), respectively.

10. The microwave detection robot detection system according to claim 9, characterized in that: The environmental compensation module (52) is implemented using a BP neural network, and the defect recognition module (53) is implemented using a CNN neural network.

Images