Underwater submersible robot and its control method and control device

Abstract

Provided are an underwater submersible robot capable of improving its ability to resist interference with underwater turbulence, and a control method and control device for the same. [Solution] A first difference value between the current actual posture and desired posture of the robot body and a second difference value between the actual depth and desired depth are calculated, the first difference value and the second difference value are input into a terminal sliding mode surface to be set, an output value of the terminal sliding mode surface is obtained, the output value is input into a pre-set high-order observer, radial basis function neural network and terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are input into the terminal sliding mode control law to obtain a virtual force of the terminal sliding mode control law output, power allocation is performed for each propeller of the propeller component based on the virtual force, thrust of each propeller is obtained, and the propeller of the underwater submersible robot is operated and controlled based on the thrust.

Description

[Technical field]

[0001] The present invention relates to the technical field of robot control, and more particularly to an underwater submersible robot and a control method and device thereof. [Background technology]

[0002] At present, society has created a huge demand for the construction of various underwater works, such as water-related infrastructure construction and marine resource development facilities, and the construction of underwater works, such as bridges across rivers or seas, marine ranches, offshore wind power generation, and offshore drilling platforms, is essential. These important construction facilities must be regularly maintained and inspected during use to ensure the normal operation of important projects such as traffic transportation and resource development, and to avoid the occurrence of serious casualties and serious economic losses. For example, as the main load-bearing member of the underwater construction structure, the underwater pile foundation directly bears the load transmitted from the superstructure of the underwater construction, so its health condition is extremely important. An efficient and accurate underwater pile foundation inspection technology can provide important construction information to judge the safety and reliability of underwater construction and guide maintenance and repair work. However, the special characteristics and extreme conditions of the underwater environment, such as high pressure and turbulent impact, greatly increase the complexity and risk of construction inspection. A complex flow environment, such as high water flow velocity, is prone to interference and misjudgment of the inspection results.

[0003] The traditional underwater construction inspection method requires sending divers to conduct underwater inspection, which faces several difficulties. First, there are many unknown risks in underwater work, which pose a direct threat to the life safety of divers. Second, the working time of divers is limited and cannot meet the demand for long-term underwater work, and the diving depth is physiologically limited. In addition, it is difficult to find defects in underwater construction using the naked eye inspection method. These defects make it difficult to timely and accurately obtain the health status of the underwater pile foundation, making it difficult to deal with related risks early. Therefore, in recent years, a method of inspecting underwater construction using underwater robots has emerged.

[0004] The rapid development of underwater robot technology has provided reliable engineering equipment for underwater engineering inspection, but the underwater movement control method of the currently common underwater inspection robot mainly moves in the water at a set speed, and has problems such as poor anti-interference ability against underwater turbulence, poor maneuverability, lack of stability, etc. Therefore, an underwater robot that can solve the above problems is needed. Summary of the Invention [Problem to be solved by the invention]

[0005] An embodiment of the present invention provides an underwater submersible robot that can improve the ability of the underwater submersible robot to resist interference with underwater turbulence, and a control method and control device for the underwater submersible robot. [Means for solving the problem]

[0006] One embodiment of the present invention is an underwater submersible robot, The robot body, an inspection device provided on the robot body for inspecting an underwater target; a propeller component provided on the robot body for driving the underwater submersible robot underwater; A sensing device provided on the robot body for sensing the posture and depth of the robot body; A control device is provided on the robot body and connected to the inspection device, the sensing device, and the propeller component; and Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is respectively set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the terminal sliding mode control law output; Performing a power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; and a control device for controlling the propulsion operation of the propeller based on the thrust.

[0007] As an improvement of the above aspect, when the control device is used to allocate power to each propeller of the propeller component based on the virtual force and obtain thrust of each propeller, specifically, A second order filter is used to filter the virtual force and convert the filtered virtual force into thrust for each propeller of the propeller component through power allocation.

[0008] As an improvement to the above aspect, the propeller component includes four propellers, the four propellers being rotatably mounted on both ends of both sides of the robot body and connected to the control device.

[0009] As an improvement of the above aspect, the robot body includes a main body chamber, two alloy skeletons and two streamlined casings, the two alloy skeletons are respectively provided on both sides of the main body chamber so as to be rotatable around the axial direction of the main body chamber, and the two streamlined casings surround the two alloy skeletons in a one-to-one correspondence and are connected to the two alloy skeletons in a corresponding manner.

[0010] As an improvement of the above aspect, the main body chamber includes an inspection head cover, a connection part, a control chamber casing, a special-shaped chamber, a control chamber cover, a battery chamber cover, a battery chamber casing, a chamber cover transition part, and a handle, which are connected and installed in sequence along a longitudinal direction; The inside of the control chamber casing is used to accommodate and install the control device, the inside of the inspection head cover is used to accommodate and install the inspection device, and the inside of the battery chamber cover is used to accommodate and install the battery chamber; The portion of the connecting part that contacts the control chamber casing and the portion that contacts the inspection head lid are both provided with annular grooves along the outer periphery of the connecting part, and an O-seal ring is attached to the annular groove. The portion of the connecting part protrudes into the interior of the inspection head lid and the irregular chamber and is in close contact with the interior. A set screw is provided at the portion of the connecting part that protrudes into the inspection head lid and the irregular chamber. The portion of the special-shaped chamber that contacts the control chamber casing is provided with an annular groove in which an O-seal ring is attached, and a plurality of mesas for attaching water-sealing contacts are recessed and formed on the outer periphery of the special-shaped chamber, the sensing device is provided in the special-shaped chamber, and a lead wire is led out through the water-sealing contact, a receiving groove is formed in the middle of the control chamber cover for accommodating a male plug connected to the battery chamber cover, and a receiving groove is formed in the middle of the battery chamber cover for accommodating a female plug into which the male plug is inserted; an O-seal ring is provided between the battery chamber cover and a portion connected to the battery chamber casing; a battery platform is placed inside the battery chamber cover, and the battery platform is used to mount a battery; An O-seal ring and a set screw are provided at a connection point between one end of the chamber cover transition part and the battery chamber casing, and the handle is provided at the other end of the chamber cover transition part.

[0011] As an improvement of the above aspect, the alloy skeleton includes a head skeleton, a middle skeleton, and a tail skeleton that are sequentially installed along the longitudinal direction of the main body chamber, the head skeleton, the middle skeleton, and the tail skeleton are connected to both ends of the main body chamber via the connecting parts, and the main body chamber is connected to the corresponding streamlined casing via the head skeleton, the middle skeleton, and the tail skeleton; An expansion framework is installed at the bottom of the main chamber, and hole locations are left in the expansion framework for mounting expansion modules.

[0012] As an improvement to the above aspect, the underwater submersible robot further includes a mounting base plate and at least two adaptive magnetization modules; The mounting base plate is used for detachably connecting to the robot body, and at least two of the adaptive magnetization modules are connected to both sides of the mounting base plate; The adaptive magnetization module includes: a lifting plate, a first small hydraulic rod member, a second small hydraulic rod member, a magnetization fixing module, a support plate, a second large hydraulic rod member, a first large hydraulic rod member, an adaptive adjustment hugging mechanism, and a bottom connecting skeleton; The angles formed by the two sides of the support plate and the central part of the support plate are obtuse angles, and the bottom sides of the two sides of the support plate are respectively connected to the magnetization fixing module via at least one pair of the first small hydraulic rod member and the second small hydraulic rod member, and the first small hydraulic rod member is hydraulically driven and connected to the second small hydraulic rod member; The top sides of both sides of the support plate are each connected to the mounting bottom plate via at least one pair of the second large hydraulic rod member and the first large hydraulic rod member, and the first large hydraulic rod member is hydraulically connected to the second large hydraulic rod member.

[0013] As an improvement to the above aspect, the underwater submersible robot further includes two pairs of motion assistance modules, the two pairs of motion assistance modules being provided on both sides of the robot body, The motion assistance module includes a fixed bracket, a motor, and a spoiler, the fixed bracket is provided on one side of the robot body, the spoiler is provided on the fixed bracket so as to be rotatable up and down, and the motor is provided on the robot body and is used to drive the spoiler to move up and down.

[0014] As an improvement to the above aspect, the inspection device includes an underwater vision matrix module, the underwater vision matrix module includes a matrix sleeve ring and at least two sets of camera adjustment components, the matrix sleeve ring is used for removably mounting to the robot body, and the at least two sets of camera adjustment components are uniformly distributed on the same side of the matrix sleeve ring; The camera adjustment component includes a push rod motor, a clamp plate, a waterproof motor, a waterproof camera, a movement frame and a hinge movement frame, the bottom end of the hinge movement frame is hingedly connected to the row and column sleeve ring, the waterproof motor is provided at the top end of the hinge movement frame, the waterproof camera is provided at the free end of the rotation shaft of the waterproof motor, one end of the clamp plate is connected to the middle of the hinge movement frame, the other end of the clamp plate is connected to one end of the movement frame, the other end of the movement frame is connected to one end of the clamp plate of an adjacent another camera adjustment component, the other end of the clamp plate of the another camera adjustment component is connected to the middle of the hinge movement frame of the camera adjustment component, the push rod motor is provided on the row and column sleeve ring, and the free end of the push rod of the push rod motor is connected to the middle of the movement frame.

[0015] As an improvement to the above aspect, the underwater submersible robot further comprises a variable stiffness-flexible embracing self-stabilizing module, the variable stiffness-flexible embracing self-stabilizing module comprising a fixed main frame, a pair of hydraulic grippers, a variable stiffness covering array, a gap-filling covering pneumatic unit, and a carbon fiber intermediate layer; The fixed main frame is used to be detachably connected to the robot body, a pair of hydraulic grippers are respectively arranged on both sides of the fixed main frame so as to be movable toward or away from each other, the variable stiffness covering array is provided on both sides of the inner free ends of the pair of hydraulic grippers, the gap-filling covering pneumatic unit is provided in the middle of the inner sides of the pair of hydraulic grippers, and the gap-filling covering pneumatic unit is covered with the carbon fiber intermediate layer.

[0016] Another embodiment of the present invention is a control method for an underwater submersible robot used in the underwater submersible robot described in any one of the above, comprising: Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, respectively, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the terminal sliding mode control law output; performing power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; and controlling the propulsion operation of the propeller based on the thrust.

[0017] As an improvement of the above aspect, the power allocation of each propeller of the propeller component based on the virtual force and obtaining the thrust of each propeller is The method includes filtering the virtual force using a second-order filter, and converting the filtered virtual force into thrust for each propeller of the propeller component through power allocation.

[0018] Another embodiment of the present invention provides a control device for an underwater submersible robot, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, which, when the processor executes the computer program, realizes the control method for an underwater submersible robot described in the above embodiment of the invention. Effect of the Invention

[0019] Compared with the prior art, one of the above embodiments has the following advantages.

[0020] A first difference value between the current actual attitude and the desired attitude of the robot body and a second difference value between the actual depth and the desired depth are calculated, and the first difference value and the second difference value are input to a terminal sliding mode surface to be set, and an output value of the terminal sliding mode surface is obtained. The output value is respectively input to a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are respectively used as compensation inputs of the terminal sliding mode control law, so that data deviation compensation can be performed for the current underwater turbulence, so that the terminal sliding mode control law can finally output a more accurate virtual force, and power allocation of each propeller of the propeller component is performed according to the virtual force, the thrust of each propeller is obtained, and the operation control of the propeller of the underwater submersible robot is performed according to the thrust, so that a more reasonable motion control can be performed for the current underwater turbulence. As can be seen, the embodiment of the present invention can improve the underwater submersible robot's ability to resist interference with underwater turbulence, thus realizing the autonomous stabilization of the robot body in a complex environment and the autonomous escape operation in an emergency environment, effectively realizing a strong autonomous anti-interference function, improving the operational safety and stability of the equipment, reducing the probability of equipment accidents, and realizing the smart operation of the underwater drone based on the control policy, greatly improving the inspection efficiency of underwater engineering, effectively improving the safety of underwater engineering inspection work, eliminating the need for divers to directly participate in the work, and greatly reducing the risks that workers face when working in an underwater environment. Of course, any product implementing the present invention does not necessarily need to achieve all of the above advantages at the same time. [Brief description of the drawings]

[0021] [Figure 1] 1 is a schematic structural diagram of an underwater submersible robot according to an embodiment of the present invention, taken from one viewing angle; [Diagram 2] FIG. 2 is a schematic diagram of another viewing angle direction structure of an underwater submersible robot according to an embodiment of the present invention. [Diagram 3] 1 is a schematic diagram of the structure of a robot body of an underwater submersible robot according to an embodiment of the present invention; [Figure 4] 1 is a cross-sectional view of a robot body of an underwater submersible robot according to an embodiment of the present invention. [Diagram 5] 1 is a schematic diagram showing an electronic device according to an embodiment of the present invention being fixed to a skeleton of a robot body; [Figure 6] FIG. 2 is a structural schematic diagram of an adjustable battery fixing skeleton of a robot body according to an embodiment of the present invention. [Figure 7] 2 is a structural schematic diagram of a low fluid resistance inspection head cover of a robot body according to an embodiment of the present invention; FIG. [Figure 8] 2 is a structural schematic diagram of a special-shaped chamber of a robot body according to an embodiment of the present invention; FIG. [Figure 9] 2 is a structural schematic diagram of a motion assistance module of an underwater submersible robot according to an embodiment of the present invention; FIG. [Figure 10] 1 is a schematic diagram showing the installation of a motion assistance module of an underwater submersible robot according to an embodiment of the present invention. [Figure 11] 1 is a structural schematic diagram of an adaptive magnetization module of an underwater submersible robot according to an embodiment of the present invention; [Figure 12] FIG. 2 is a detailed schematic diagram of an adaptive magnetization module of an underwater submersible robot according to one embodiment of the present invention. [Figure 13] 1 is a schematic diagram of the installation of an adaptive magnetization module of an underwater submersible robot according to one embodiment of the present invention. FIG. [Figure 14] 1 is an execution flowchart of an adaptive magnetization module of an underwater submersible robot according to an embodiment of the present invention. [Figure 15] 1 is a structural schematic diagram of an underwater vision module of an underwater submersible robot according to an embodiment of the present invention; [Figure 16] 1 is a schematic diagram showing the installation of an underwater vision module of an underwater submersible robot according to an embodiment of the present invention. [Figure 17] 4 is an execution flowchart of an underwater vision module of an underwater submersible robot according to one embodiment of the present invention. [Figure 18] 4 is a flowchart showing a processing algorithm of an underwater vision module of an underwater submersible robot according to an embodiment of the present invention. [Figure 19] FIG. 2 is a structural schematic diagram of a variable stiffness and flexibility hugging self-stabilizing module of an underwater submersible robot according to one embodiment of the present invention. [Figure 20] FIG. 2 is a detailed schematic diagram of a variable stiffness-flexibility hugging self-stabilizing module for an underwater submersible robot according to one embodiment of the present invention. [Figure 21] FIG. 2 is an operational schematic diagram of a variable stiffness-flexibility hugging self-stabilizing module for an underwater submersible robot according to one embodiment of the present invention. [Figure 22] FIG. 2 is a schematic diagram of the implementation logic of a variable stiffness-flexibility hugging self-stabilizing module for an underwater submersible robot according to one embodiment of the present invention. [Figure 23] FIG. 2 is a principle diagram of a smart self-anti-interference control policy of an underwater submersible robot according to an embodiment of the present invention. [Figure 24] 1 is a structural schematic diagram of a control system for an underwater robot according to an embodiment of the present invention; [Diagram 25] 1 is a flowchart of a control method for an underwater submersible robot according to an embodiment of the present invention. [Figure 26] 1 is a structural schematic diagram of a control device for an underwater submersible robot according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments that can be obtained by those skilled in the art without creative labor belong to the scope of protection of the present invention.

[0023] In the description of this specification and the claims, it should be understood that the orientations or positional relationships indicated by the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inside," "outside," etc. are based on the orientations or positional relationships shown in the drawings and are intended merely to facilitate the description of the embodiments of the present invention, and do not indicate or imply that the indicated devices or parts must be configured and operated in a particular orientation or in a particular direction, and cannot be understood as limitations on the embodiments of the present invention.

[0024] It should be noted that the terms "first," "second," etc. in the specification and claims are merely intended to distinguish between descriptions of identical technical features, and do not indicate or imply relative importance or implicitly indicate the number of the indicated technical features, and do not necessarily describe an order or chronology. Where appropriate, the terms are interchangeable. Thus, a feature qualified with "first" or "second" may explicitly or implicitly include at least one of the said features.

[0025] 1 to 24, one embodiment of the present invention provides an underwater submersible robot, the underwater submersible robot including a robot body 1, an inspection device 2, a propeller component 7, a sensing device 8 and a control device 9, the inspection device 2 is provided on the robot body 1 and used to inspect an underwater target, the propeller component 7 is provided on the robot body 1 and used to drive the underwater submersible robot underwater, the sensing device 8 is provided on the robot body 1 and used to sense the attitude and depth of the robot body 1, the control device 9 is provided on the robot body 1 and connected to the inspection device, the sensing device and the propeller component 7, and used to execute a self-anti-interference smart control policy, the self-anti-interference smart control policy being: Calculating a first difference value between a current actual posture and a desired posture of the robot body 1 and a second difference value between an actual depth and a desired depth of the robot body 1; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, respectively, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the terminal sliding mode control law output; performing power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; and controlling propulsive actuation of the propeller based on the thrust.

[0026] For example, the sensing device 8 may include a depth sensor and a posture detection sensor, and transmits related actual posture data and actual depth data to the control device 9 upon detection.

[0027] Specifically, the attitude deviation and the depth deviation are input to the terminal sliding mode surface to be set, and the terminal sliding mode surface is respectively the input of the high-order observer, the radial basis function neural network, and the terminal sliding mode control law that are set in advance. Since chattering is likely to occur in the output of the terminal sliding mode control law, the high-order observer can more accurately observe the chattering state and estimate the deviation, and thus can be used to suppress the chattering of the output of the terminal sliding mode control law. The radial basis function neural network is a highly efficient feedforward neural network that can perform nonlinear function approximation and realize real-time optimization of the system, so that the stability and anti-interference ability of the sliding mode control system can be improved by introducing the radial basis function neural network. Therefore, the output of the high-order observer and the output of the radial basis function neural network are used as the compensation input of the terminal sliding mode control law, and finally the virtual force of the output of the terminal sliding mode control law can be made more reasonable, which finally improves the ability of the underwater submersible robot to resist turbulence.

[0028] In an embodiment of the present invention, a first difference value between the current actual posture and the desired posture of the robot body and a second difference value between the actual depth and the desired depth are calculated, the first difference value and the second difference value are input to a terminal sliding mode surface to be set, and an output value of the terminal sliding mode surface is obtained. The output value is respectively input to a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are respectively used as compensation inputs of the terminal sliding mode control law, so that data deviation compensation can be performed for the current underwater turbulence, so that the terminal sliding mode control law can finally output a more accurate virtual force, and the power allocation of each propeller of the propeller component is performed based on the virtual force, the thrust of each propeller is obtained, and the operation control of the propeller of the underwater submersible robot is performed based on the thrust, so that a more reasonable motion control can be performed for the current underwater turbulence. As can be seen, in the embodiments of the present invention, the underwater submersible robot can improve its ability to resist interference with underwater turbulence, and can realize the autonomous stabilization of the robot body in such a complex environment and the autonomous escape operation in an emergency environment, effectively realizing a strong autonomous anti-interference function, improving the operational safety and stability of the equipment, and reducing the probability of equipment accidents. Furthermore, based on the control policy, the underwater drone can be operated intelligently, greatly improving the inspection efficiency of underwater construction, effectively improving the safety of underwater construction inspection work, eliminating the need for divers to directly participate in the work, and greatly reducing the risks faced by workers when working in an underwater environment.

[0029] As an improvement of the above embodiment, the control device 9 performs power allocation of each propeller of the propeller component based on the virtual force described above, and determines the location used to obtain the thrust of each propeller, specifically, A second order filter is used to filter the virtual force and convert the filtered virtual force into thrust for each propeller of the propeller component through power allocation.

[0030] Specifically, referring to FIG. 23, a specific implementation process of the self-anti-interference smart control policy is as follows:

[0031] Dynamics Modeling: First, two basic coordinate systems for the underwater robot are established: the fixed coordinate system E-ξηζ (fixed system) and the moving coordinate system G-xyz (moving system).

[0032] J(η)∈R 6×6 is a Jacobian matrix, and its expression is as follows:

[0033]

number

number

[0034] The transformation matrix T2 from the moving system of angular velocity to the fixed system is given by:

number

[0035] Based on existing modeling experience, we make the following assumptions:

[0036] (1) The underwater robot can be regarded as a rigid body with a constant mass, and changes in conditions such as density and pressure in the water can be ignored.

[0037] (2) When the underwater submersible robot is operating, the hardware inside the robot body remains fixed, i.e., the positions of the center of gravity and center of buoyancy of the underwater submersible robot do not change.

[0038] (3) Neglect towing force and moment interference due to buoyant umbilicals.

[0039] (4) The underwater robot is in water during operation and is wet.

[0040] (5) The pitch angle θ is set to prevent the singularity of J(η).

number

number

number

[0041] Based on the above assumptions, the six-degree-of-freedom nonlinear dynamics equations of the underwater robot are established as follows:

number

[0042] M∈R 6×6 is the inertia matrix, C(v)∈R 6×6 are the Coriolis and centripetal matrices, D(v)∈R 6×6 is the damping matrix, g(η)∈R 6 are the restoring forces and moment matrices, τ∈R 6 are the thrust and moment, τd1∈R 6 is the model uncertainty interference.

[0043] Dual loop control policy:

number

[0044] Then, we define the velocity tracking error sliding mode surface as follows:

number

[0045] (4) together with ν c ∈R 6 is designed as the speed control command for the reference speed.

number

[0046] however,

number

number

number

[0047] Based on (7), the inner loop terminal sliding mode surface is designed.

number

[0048] The derivative of ν is:

number

[0049] Differentiating (7) and substituting it into (5) gives us the following:

number

[0050] Substituting (10) into (8) we get the following:

number

[0051] S T Differentiate as follows:

number

number

[0052] Gaussian radial basis function neural network RBF has function approximation ability and can be used to control complex nonlinear systems with uncertainty. In the present invention, six similar Gaussian radial basis function neural network RBF are designed, and the ideal value of the unknown function f to be approximated can be written as follows:

number

[0053] where φ(S T )∈R 30 and

number

number

number

number

number

number

number

[0054] C ik and b ik represent the center and width of the Gaussian function, respectively.

[0055] The vector of estimated weights is expressed as follows:

number

[0056]

number

[0057] Unknown function f∈R 6 Let be the output of the RBF and express it as follows.

number

[0058] The proposed dual-loop neural network sliding mode controller design is as follows:

number

[0059] however,

number

number

[0060] however,

number

[0061] Select the following Lyapunov functions: V = V1 + V2 + V3, (20) however,

number

number

number

[0062] Substituting the update rule (26) into (25), we get

number

[0063]

number

[0064]

number

number

[0065] Differentiating (22), we get:

number

[0066] Substituting (13) and (17) into (32), we get:

number

[0067] From (33), the terminal sliding mode surface error obtained is:

number

[0068] Using (33), (34) can be turned into an inequality.

number

[0069] The differential term of the terminal sliding mode surface corresponds to the dynamics equation (4) of the underwater submersible robot, and (4) is divided into a reference term and an uncertainty term. Because the underwater submersible robot has high nonlinearity, high coupling degree and strong interference, the dynamics equation of the reference term is not accurate, and the terminal sliding mode surface requires high model accuracy, so a high-order observer needs to be designed to compensate for the uncertainty term. By using (8), the following can be obtained.

number

[0070] s in (36) i of

number

number

[0071] However, π i0 ,π i1 ,π i2 ,...,π im are respectively

number

number

number

number

[0072] Combining (37) and (38), we derive the following.

number

[0073] Then, select the following positive definite function (23):

number

[0074]

number

number

number

[0075] Differentiating V3 gives us:

number

[0076] Substituting (43) into (44) we get:

number

[0077] Differentiating (20) and calling (31), (35) and (45), we get

number

[0078] Then, by invoking Lemma 1, we get:

number

[0079] λ min (·) represents the minimum feature value of the matrix.

number

[0080] The proof is completed.

[0081] Theorem 1: For the described underwater robotic system, with the control law (18) and the update law (26), if the initial conditions are bounded, then the sliding mode surface error and its derivatives are semi-globally bounded.

[0082] Proof:(47)e βt Multiplying by gives:

number

[0083] Integrating (50) gives us:

number

[0084] Substituting (21), (22), and (23) into (51), we get the following:

number

[0085] Based on Lemma 1, for the inner loop terminal sliding mode surface, there is a finite time t reach Inside S T =0, we get the following:

number

[0086] Using (58), we can obtain the following two forms:

number

[0087]

number

number

number

number

number

[0088] With the same proof steps, we can prove (59),

number

number

number

[0089] Therefore, from the above analysis, the velocity tracking error converges to zero within a finite time, and the initial value of the total time taken is

number

[0090] t>t sum Then, from (7) we get ν=ν c (65)

[0091] (65)

number

number

[0092] (66)

number

[0093] To understand, we consider (66) as a second-order filter system with zero input, and define its natural frequency and damping ratio as

number

number

[0094] As described above, the self-anti-interference smart control policy enables the robot to autonomously compensate for water current disturbances when following a track, thereby achieving accurate path tracking and navigation control.

[0095] As an improvement to the above embodiment, referring to Figures 1 and 2, the propeller component 7 includes four propellers, which are rotatably mounted on both ends of the robot body 1 in all directions, and the four propellers are connected to the control device 9.

[0096] As an improvement of the above embodiment, referring to Figures 1 to 4, the robot body 1 includes a main body chamber 10, two alloy skeletons 200, and two streamlined casings 206, the two alloy skeletons 200 are rotatably arranged in the main body chamber 10 around the axial direction of the main body chamber 10, and the two streamlined casings 206 surround the two alloy skeletons 200 in a one-to-one correspondence and are connected to the two alloy skeletons 200 in a corresponding manner.

[0097] Specifically, the main chamber 10 includes an inspection head cover 101, a connection part 102, a control chamber casing 103, a special-shaped chamber 104, a control chamber cover 105, a battery chamber cover 106, a battery chamber casing 107, a chamber cover transition part 108, and a handle 109, which are connected in sequence along the longitudinal direction. The inside of the control chamber casing 103 is used to install the control device 9, the inside of the inspection head cover 101 is used to install a camera, and the inside of the battery chamber cover 106 is used to install a battery. The connecting part 102 is used to accommodate and install a chamber, and a portion of the connecting part 102 that contacts the control chamber casing 103 and a portion of the connecting part that contacts the inspection head cover 101 are provided with annular grooves along the outer periphery of the connecting part, and an O-seal ring is attached to the annular groove. The connecting part 102 is inserted into the inspection head cover 101 and the irregular chamber 104 and is tightly attached to the inside of the irregular chamber 104. A set screw is installed at the portion of the connecting part 102 that projects into the inspection head cover 101 and the irregular chamber 104, and the control chamber of the irregular chamber 104 is attached to the control chamber casing 103. A ring-shaped groove is formed at a portion in contact with the chamber casing 103 to receive an O-seal ring, and eight mesas are formed on the outer periphery of the irregular chamber 104 to receive water-seal contacts. A sensing device including a depth sensor and a switch is provided in the irregular chamber, and a lead wire is led out through the water-seal contact. A receiving groove is formed at the middle of the control chamber cover 105 to receive a male plug connected to the battery chamber cover 106, and a receiving groove is formed at the middle of the battery chamber cover 106 to receive a female plug into which the male plug is inserted. An O-seal ring is provided at a portion of the battery chamber cover 106 connected to the battery chamber casing 107, and a battery platform is placed inside the battery chamber cover 106. The battery platform is used to mount a battery. An O-seal ring and a set screw are provided at a connection point between one end of the chamber cover transition part 108 and the battery chamber casing 107, and the handle 109 is provided at the other end of the chamber cover transition part 108.

[0098] As an improvement of the above embodiment, referring to Figures 3 to 4, the alloy skeleton 200 includes a head skeleton 201, a middle skeleton 202 and a tail skeleton 203 which are arranged in order along the longitudinal direction of the main chamber 10, and the head skeleton 201, the middle skeleton 202 and the tail skeleton 203 are connected to both ends of the main chamber 10 via the connecting parts 102, the main chamber 10 is connected to the streamlined casing 206 via the head skeleton 201, the middle skeleton 202 and the tail skeleton 203, and an expansion skeleton is installed at the bottom of the main chamber 10, and a hole position for attaching an expansion module is left in the expansion skeleton.

[0099] For ease of understanding, the robot body of the underwater submersible robot will be described in more detail below.

[0100] 1 to 8, in the robot body of the underwater submersible robot, three modules, namely, the main body chamber 10, the high-strength lightweight alloy skeleton 200, and the streamlined casing 206, constitute the drone body, and the main body chamber 10 module includes nine parts, namely, the inspection head cover 101, the connecting part 102, the control chamber casing 103, the irregular chamber 104, the control chamber cover 105, the battery chamber cover 106, the battery chamber casing 107, the chamber cover transition part 108, and the handle 109. Inside the inspection head cover 101, a lens cover 117, an O-ring 122, a camera mounting plate 120, an EVA foam 119, a tempered glass 118, and a super-resolution camera 121 are provided. The super-resolution camera 121 is mounted on the camera mounting plate 120, and the camera mounting plate 120 is fixed to the inspection head cover 101 via a screw 123 for image identification.

[0101] An electronic component arrangement plate is installed inside the control chamber casing 103, and the electronic component arrangement plate is fixed to the inner wall of the connecting part 102 through the copper pillar 124, so that various control components and electronic components can be installed and arranged. The electronic component arrangement plate is divided into a left damper 128, a right damper 125, and an intermediate support plate 126, and the left and right dampers are connected and fixed to the connecting part 102 through the copper pillar 124, and the intermediate support plate 126 is sandwiched between the left and right dampers, and both sides of the intermediate support plate 126 can be inserted into the notches 127 of the left and right dampers to achieve fixing. The left and right dampers adopt a hollow design in the middle, which facilitates the passage of lead wires, and the notches 127 are installed only at the position where they are fitted with the intermediate support plate 126. Various notches and openings are installed inside the intermediate support plate 126 for mounting electronic devices.

[0102] The parts of the connecting part 102 that come into contact with the control chamber casing 103 and the inspection head lid 101 have notches 129 for fitting O-rings. When connected, part of the wall of the connecting part 102 penetrates into the interior of the inspection head lid 101 and the irregular chamber 104 and comes into close contact, achieving a water-sealed installation. A set screw 111 is installed at the position where they penetrate and overlap, fixing the two chamber bodies together and preventing them from falling off. The subsequent installation of the chamber body is also performed in this way.

[0103] Connected to the control chamber casing 103 is the irregular chamber 104, the contact portion is also grooved to mount an O-ring, and the irregular chamber 104 is provided with eight small mesas, the small mesas 132 are designed to be recessed and are used to mount the water seal contact 113, the depth sensor 1140 and the switch 115, and the lead wires are led out through the water seal contact 113 to achieve waterproofing and quick connection and disconnection. The water seal contact 113 is mounted on the water seal base 133, which is further mounted on the six small mesas 132 in the irregular chamber 104. The mesa 131 is used to mount the water seal contact 113 and the switch 114, and the mesa 131 is deeper than the mesa 130. Similarly, the water seal contact 113 and the switch 114 are mounted on the irregular chamber 104 through the water seal base 133.

[0104] The irregular chamber 104 and the control chamber lid 105 are similarly connected by a water seal, and a water seal O-ring 134 is installed at the connection site, and the middle of the control chamber lid 105 is recessed to accommodate a male plug 135, which is connected to the battery chamber lid 106. A similar structure for mounting a female plug 136 is also provided in the middle of the battery chamber lid 106, and when attached, the protruding part in the middle of the battery chamber lid 106 can be inserted into the control chamber lid 105, and the recesses in the middle of the control chamber lid 105 and the battery chamber lid 106 are used to accommodate the male plug 135 and the female plug 136, respectively. The battery chamber cover 106 is connected to the battery chamber casing 107, and an O-ring 137 is also provided at the connection site. A battery platform 138 is placed inside the battery chamber, and the battery platform 138 is used to mount a battery 139. It is connected to a hole position 145 in the battery chamber cover 106 via a copper pillar 144, and includes a battery frame 146, a lead wire EVA board 147, a battery flat plate support frame 148, a battery folding support frame 149, and a battery top plate 150. The battery frame 146 is connected to the battery chamber lid 106 via the copper pillar 144, the middle opening of the battery frame 146 is used for the battery lead wire to pass through, the lead wire EVA board 147 is disposed on the battery frame 146 and used for guiding the battery wiring, the battery bending support frame 149 and the battery flat plate support frame 148 are disposed on the battery frame 146 and fixed via screws and used for restricting the radial movement of the battery, and the battery top plate 150 is connected to the top of the battery bending support frame 149 and the battery flat plate support frame 148 to prevent the battery from moving along the axial direction.

[0105] The chamber lid transition part 108 is connected to the battery chamber casing 107 , and an O-ring 140 and a set screw 141 are installed at the connection part, and the chamber lid transition part 108 is connected to a handle 143 .

[0106] The main chamber 10 is connected to the casing via a head skeleton 201, a middle skeleton 202, and a tail skeleton 203. The head skeleton 201, the middle skeleton 202, and the tail skeleton 203 are symmetrically distributed on the left and right sides of the main chamber 10, and are fixed to the upper surface of the connecting part 102 in the main chamber via screws. The head skeleton 201, the middle skeleton 202, and the tail skeleton 203 are connected to a streamlined casing 206, the head skeleton 201 is connected to the main control chamber and the streamlined casing 206, and the tail skeleton 203 is connected to the power chamber and the low fluid resistance casing 204. The casing has four brushless motor propellers 207 that provide lifting power arranged vertically, and four duct propellers 208 arranged vector-shaped horizontally, and their lead wires enter the irregular chamber 104 through the water-sealed contact 113 to drive the machine and realize functions such as rapid movement and levitation.

[0107] The robot body is used for carrying tasks using a lightweight alloy handle 204, and conventional expansion modules such as a DVL, a single-degree-of-freedom underwater arm, a sound bar, a binocular camera, and a laser distance measuring device can be attached using holes left in the expansion skeleton 205, and are connected and used via water-sealed contacts 113 left in the robot body.

[0108] As an improvement of the above embodiment, referring to Figs. 11 to 14, the underwater submersible robot further includes a mounting bottom plate 410 and at least two adaptive magnetization modules 4, the mounting bottom plate 410 is used for detachably connecting to the robot body 1, and at least two of the adaptive magnetization modules 4 are connected to both sides of the mounting bottom plate 410, the adaptive magnetization module 4 includes a lifting plate 401, a first small hydraulic rod member 402, a second small hydraulic rod member 403, a magnetization fixing module 404, a support plate 405, a second large hydraulic rod member 406, a first large hydraulic rod member 407, an adaptive adjustment hugging mechanism 408 and a bottom connection skeleton 409, and each of the two sides of the support plate 405 is The angle formed between the bottom plate 404 and the middle of the support plate 405 is an obtuse angle, and each of the bottom sides of both sides of the support plate 405 is connected to the magnetization fixing module 404 via multiple sets of the first small hydraulic rod member 402 and the second small hydraulic rod member 403, the first small hydraulic rod member 402 is hydraulically connected to the second small hydraulic rod member 403 and can move linearly relative to the second large hydraulic rod member 406, and each of the top sides of the support plate 405 is connected to the mounting bottom plate 410 via the second large hydraulic rod member 406 and the first large hydraulic rod member 407, and the first large hydraulic rod member is hydraulically connected to the second large hydraulic rod member.

[0109] Specifically, the adaptive magnetization module 4 is mainly used for fixed adsorption and stable movement along the pole, and can effectively resist the strong ocean current interference encountered during the inspection process.

[0110] The bottom of the underwater robot body and the adaptive controllable magnetized attachment module are tightly connected through screws and fixing holes, the mounting bottom plate 410 is tightly attached to the bottom of the underwater robot body, effectively playing the role of supporting and stabilizing, and the adaptive adjustment hugging mechanism 408 adopts a tilted current guide mode to realize a stable connection, and effectively reduce the influence of water resistance that the equipment experiences during operation, and reduce excessive energy loss. When the equipment is to identify and inspect the crack defects on the surface of the underwater construction, when the equipment approaches the underwater construction, the underwater robot can accurately adjust the rotation speed and rotation direction of each propeller to realize the equipment operating up to a vertical state. In addition, with the drive of the propeller, the equipment will gradually approach the crack defects on the surface of the underwater construction in a vertical state, at this time, the adaptive controllable magnetized attachment module attached to the bottom of the equipment, and the magnetized fixation module 404 attached to the lower end of the receiving plate 405 will realize the rapid magnetization and demagnetization repeated inside the magnetized fixation module 404 by short electric pulses, generating huge magnetic force, and since the inside of the underwater construction support column is full of iron bars, the magnetized fixation module 404 can be rapidly and stably attached to the surface of the underwater construction support column, so that the whole equipment can be stably and tightly fixed on the surface of the underwater construction support column. When the equipment is attached to the surface of the underwater construction support column, the horizontal propeller of the equipment will increase the rotation speed, at this time, the equipment will autonomously check the adsorption status of the adaptive controllable magnetized attachment module, perform a stability early warning test, and reach the rated inspection rotation speed. When the equipment can be stably adsorbed on the surface of the underwater construction support column, the propeller rotation speed will return to normal, and continue to move along the column surface and perform defect inspection. In addition, during attachment, the equipment adjusts the opening and closing status of the bottom connecting skeleton 409 in real time according to the radius of the underwater construction support pillar. When the underwater construction support pillar is thick, the bottom connecting skeleton 409 will be flipped upward to expand the middle fixing space, effectively increasing the attachment radius of the adaptive controllable magnetized attachment module, making the magnetized fixing module 404 better adhere to the surface of the underwater construction support pillar, effectively improving the adaptability of the magnetic attraction attachment module, and accurately expanding the application range of the module.In addition, when the magnetized clamping module 404 is fixed on the surface of the underwater construction support column, the lifting plate located in the middle gradually descends, and the magnetized clamping module 404 at the bottom of the lifting plate 401 adheres to the surface of the underwater construction support column, and the first large hydraulic rod member 407 and the second large hydraulic rod member 406 located above the lifting plate adjust in real time to smoothly adhere the magnetized clamping module 404 to the surface of the underwater construction support column. After the equipment is stably attached to the surface of the underwater construction support column, the multiple propellers located on the surface of the robot stably provide forward and backward power to the adaptively controllable magnetized attachment module, so that the equipment can stably operate on the column surface and realize the function of stably searching for crack defects on the column surface. The first large hydraulic rod member 407 and the second large hydraulic rod member 406 located above the lifting plate 401 adjust in real time, and through the buffer adjustment of the hydraulic rod members, mainly achieve the stabilization performance of the adaptive controllable magnetized attachment module of the equipment, and play the role of reducing the pressure on the surface of the underwater engineering support column, effectively sharing the pressure on the surface of the underwater engineering support column caused by the lateral water pressure and its own weight, and avoiding other damage and influence on the support column surface. Meanwhile, the first small hydraulic rod member 402 and the second small hydraulic rod member 403 located at the lower end of the receiving plate 405 adjust in real time when the magnetized fixing module 404 is fixed, and when the equipment is subjected to water flow impact during fixing, the small hydraulic rod member module is favorable for real time adjustment and buffering, effectively reducing the impact and destruction of the water flow impact force on the entire equipment, which is more favorable to the construction of a stable inspection environment, and the equipment can stably perform defect inspection at the crack defect position. When the equipment needs to perform micro-angle adjustment in the rotation direction, the magnetization fixing module 404 reduces the magnetic force to generate a small gap between the module and the surface of the underwater engineering support column, so that the propeller of the equipment can adjust in real time to realize the rotation adjustment relative to the equipment body. When the equipment adjustment is completed, the magnetization fixing module 404 strengthens the short electric pulse intensity and greatly increases the maximum magnetic force, so that the equipment can be quickly re-adsorbed on the surface of the underwater engineering support column to complete any micro-angle rotation adjustment during operation, thereby realizing the accuracy and precision of the defect inspection position, and effectively improving the accuracy and reliability of underwater engineering defect inspection.

[0111] As shown in FIG. 13, this is the execution flowchart of the adaptive controllable magnetization adhesion module. The execution flow of the adaptive controllable magnetization adhesion module can be divided into a macro angle adjustment module, a hydraulic clamping module, and a short electric pulse trigger. During the operation of the adaptive controllable magnetization adhesion module, the macro angle adjustment module adjusts the position of the equipment at a large angle. At this time, the hydraulic clamping module synchronously performs multi-stage adjustment. Through the multi-stage response of the large hydraulic rod member adjustment module and the small hydraulic rod member adjustment module, it can effectively realize adsorption at a fixed angle and can also perform rapid adaptive adjustment. At the bottom, through the action of the short electric pulse trigger, the lower end magnetization fixing module rapidly repeats magnetization and demagnetization, thereby effectively achieving the effect of tight adsorption and fixing by magnetic force, and finally achieving rapid and stable adaptive adsorption fixing. When the identification module is operating, the magnetization fixing module is in an operating state, and can firmly fix the equipment to the surface of the underwater construction support pillar, and fix the equipment in a stable inspection environment. At this time, the super-resolution camera 121 can identify defects in the underwater construction below the equipment, and quickly backhaul the images. After the inspection is completed, the equipment can quickly respond to search for the next defect site.

[0112] As an improvement of the above embodiment, referring to Figures 9 to 10, the underwater submersible robot further includes two pairs of motion assistance modules 3, the two pairs of motion assistance modules 3 are provided on both sides of the robot body 1, and the motion assistance module 3 includes a fixed bracket 301, a motor 302 and a spoiler 303, the fixed bracket 301 is provided on one side of the robot body 1, the spoiler 303 is provided on the fixed bracket 301 so as to be able to rotate up and down, and the motor 302 is provided on the robot body 1 and drives the spoiler 303 to move up and down.

[0113] Specifically, when the device (i.e., the underwater submersible robot) operates underwater, the motion assistance module always maintains a standby state, and when the device needs to be braked quickly during operation, the propeller decelerates and pushes backwards, and at the same time, the motor 302 attached to the fixed bracket 301 operates synchronously, and the spoiler 303 located around the robot is driven by the motor 302 to expand the water stopping area along with the robot, increasing the operating resistance of the device, and rapidly reducing the operating speed of the device, thereby realizing a high-speed and highly maneuverable evacuation function. When the device needs to make a sharp turn, the operating speed of the propeller is adjusted, and when the device makes a sharp turn to the right, the side spoiler 303 is rapidly activated, and under the action of the spoiler 303, the flow speed on the right side increases instantaneously, the pressure is rapidly reduced, and under the drive of the pressure difference, the device body also rapidly turns to the right, and under the cooperative action of the spoiler 303 and the propeller, the device can make a sharp turn to evacuate. The same can be said for sharp turns to the left, enabling the underwater robot to move with high maneuverability.

[0114] As an improvement of the above embodiment, referring to Figures 15 to 18, the inspection device 2 includes an underwater vision matrix module 5, the underwater vision matrix module 5 includes a matrix sleeve ring 503 and at least two sets of camera adjustment components, the matrix sleeve ring 503 is used for being detachably attached to the robot body 1, the at least two sets of camera adjustment components are uniformly distributed on the same side of the matrix sleeve ring 503, the camera adjustment components include a push rod motor 501, a clamp plate 502, a waterproof motor 504, a waterproof camera 505, a motion frame 506 and a hinge motion frame 507, the bottom end of the hinge motion frame 507 is hingedly connected to the matrix sleeve ring 503, and the hinge motion frame 507 has a The waterproof motor 504 is provided at the top end, the waterproof camera 505 is provided at the free end of the rotation shaft of the waterproof motor 504, one end of the clamp plate 502 is connected to the middle of the hinge movement frame 507, the other end of the clamp plate 502 is connected to one end of the movement frame 506, the other end of the movement frame 506 is connected to one end of the clamp plate 502 of an adjacent another camera adjustment component, and the other end of the clamp plate 502 of the another camera adjustment component is connected to the middle of the hinge movement frame 507 of the camera adjustment component, the push rod motor 501 is provided on the matrix sleeve ring 503, and the free end of the push rod of the push rod motor 501 is connected to the middle of the movement frame 506.

[0115] Specifically, the underwater vision matrix module 5 is used to obtain a higher quality data set and a larger viewing area.

[0116] When the equipment is to identify and inspect the crack defects on the surface of the underwater work, when the equipment approaches the underwater work, the push rod motors 501 located at the upper and lower ends are started according to the distance between the equipment and the inspection target, and push the motion frame 506 to move forward. At the same time as the motion frame 506 moves forward, the hinge motion frame 507 connected to the motion frame 506 via the clamp plate 502 also moves forward accordingly, gradually approaching the target object to be inspected. When the hinge motion frame 507 moves until it approaches the inspection target, the waterproof motor 504 located at the end of the hinge motion frame 507 is started and rotates in accordance with the distance relationship and angle relationship between the waterproof camera 505 and the target to be inspected, so as to rotate the waterproof camera 505 synchronously, so as to realize the field of view of the waterproof camera 505 to inspect the inspection target in all directions, thereby improving the inspection coverage and accuracy, and effectively increasing the inspection field of view and inspection speed, and greatly improving the work efficiency and inspection accuracy.

[0117] As shown in Figure 16 and Figure 17, the underwater matrix module based on the biomimetic compound eye and the execution flow chart of its algorithm, the underwater matrix module based on the biomimetic compound eye obtains the identification sense of the position and distance, and then obtains the relevant position and distance information as the operating condition of the position and attitude adjustment module, which drives the push rod motor adjustment module and the small waterproof motor angle adjustment module to accurately adjust and change the position and attitude of the equipment, thereby realizing the rapid adaptive adjustment and switching to the optimal defect modeling position. During the adjustment, the image acquisition and identification module located at the front end combines the vision expansion algorithm and the feature fusion method to build an accurate identification system of defects, thereby realizing the rapid and accurate identification and inspection of defects of underwater construction. In addition, the equipment can perform the expansion switching of multi-modules including the exploration modules such as underwater exploration and water body inspection according to the required application, and finally effectively realize the omnidirectional, full coverage, and high-precision target identification inspection of defects of underwater construction. The feature fusion module is used to splice the collected omnidirectional target images into a target image with more complete and obvious features, and the module includes five operations: pre-processing, feature extraction, feature matching, image alignment, and image fusion. In the images taken underwater, phenomena such as scattering and color distortion exist, for which pre-processing operations including color calibration and scattering removal are required before feature fusion. In terms of color correction, the method of histogram equalization is used to improve the color accuracy of the image and make the features of the original dark image more obvious. To remove scattering, the method of wavelet denoising is used to reduce the effect of scattering in the underwater environment, taking into account the scattering noise and background noise contained in the image. The feature extraction operation is used to perform feature extraction for each image, in which the SIFT algorithm is used to extract key points with scale invariance and rotation invariance, and generate corresponding descriptors. After obtaining the feature descriptors, the image is subjected to feature matching, and the least squares algorithm based on the descriptor is used to find the correspondence between these features, thereby determining the relative position and attitude between the features. After determining the relative information between the features, image registration is performed, and based on the matching results, the affine transformation model is used to perform image registration.Finally, the aligned features are image fused, and after aligning the image features, a weighted average pixel-level fusion method is used to perform image fusion, and an average smoothing method is used to reduce the obvious and excessive edges at the junctions, which further improves the quality of the image.

[0118] As an improvement of the above embodiment, referring to Figures 19 to 22, the underwater submersible robot further includes a variable stiffness / flexible hugging self-stabilizing module 6, the variable stiffness / flexible hugging self-stabilizing module 6 including a fixed main frame 601, a pair of hydraulic grippers 60, a variable stiffness covering array 607, a gap-filling covered pneumatic unit and a carbon fiber intermediate layer 609, the fixed main frame 601 is used for being detachably connected to the robot body 1, the pair of hydraulic grippers 60 are respectively provided on both sides of the fixed main frame 601 so as to be capable of moving toward or away from each other, the variable stiffness covering array 607 is provided on both sides of the inner free ends of the pair of hydraulic grippers 60, the gap-filling covered pneumatic unit is provided in the middle of the inner sides of the pair of hydraulic grippers 60, and the gap-filling covered pneumatic unit is covered with the carbon fiber intermediate layer 609.

[0119] Specifically, the variable stiffness and flexibility self-stabilizing module 6 is used for highly efficient inspection of pile defects and resistance to ocean current interference.

[0120] The variable stiffness and flexibility holding self-stabilizing module 6 is mounted on the inspection head cover 101 in the main body chamber 10, and the hydraulic gripper 60 is connected and fixed to the inspection head cover 101 via a fixed main frame 601. The inspection head cover 101 is provided with clamp bosses 151, and the clamp bosses 151 are arranged in four circumferential directions, and have holes 150 drilled therein to fix the fixed main frame 601 via screws. The fixed main frame 601 is connected to the hydraulic gripper 60, and the hydraulic gripper 60 includes a front hydraulic part 603, a rear hydraulic part 604, a gripper head part 602, a gripper middle part 605, and a gripper tail part 606. When operated, the front hydraulic part 6 03 extends, and then the hydraulic component 604 drives the gripper head part 602, gripper middle part 605, and gripper tail part 606 of the variable stiffness-flexible hugging self-stabilizing module 6 to extend and close, and after the gripper body wraps the object, the carbon fiber middle layer 609 located in the middle part contacts the surface of the target object and plays a supporting role, and after stabilization, the gap filling coating unit 608 drives the gear ejector to let air in, during this process, the surface of the gap filling coating unit 608 comes into close contact with the target object, realizing flexible and complete coating adhesion to the surface of the target object, and improving the adhesion stability. After the gap filling coating unit 608 performs one-stage adhesion, the variable stiffness coating array 607 performs a two-stage flexible adhesion operation, and the variable stiffness coating array further adheres to the surface of the target object and increases the rigidity of the coating contact surface, and the effect of multi-stage adhesion greatly improves the stability of the equipment gripping the object. There are two sets of variable stiffness and flexibility hugging self-stabilizing modules 6, which are distributed at the front end of the equipment and can non-destructively clamp large underwater objects of any shape to complete the sampling operation. In addition to the non-destructive sampling function, in order to prevent the ocean current from disturbing the robot when performing the defect inspection work of underwater construction, which would result in poor imaging quality of the defective parts, the hydraulic component 604 drives the gripper to hug the underwater pile foundation, and uses the gap filling coating unit 608 and the variable stiffness coating array 607 to realize the non-destructive tight hug of the surface of the pile foundation, thereby realizing the stable imaging of the robot body.

[0121] During operation, after the variable stiffness and flexibility embracing self-stabilizing module 6 embraces the designated pile foundation, the battery 139 provides power, the super-resolution camera 121 can identify the image in front of the equipment and backhaul the image, the brushless motor propeller is activated to realize the equipment moving vertically to the defect position, and then the super-resolution camera 121 performs the inspection work and obtains a high-quality pile foundation defect data set.

[0122] As shown in FIG. 21, the execution logic schematic diagram of the variable stiffness and flexible hugging self-stabilizing module is shown. The hydraulic gripper is processed with high-strength and stiffness composite material, which can effectively resist the poor working environment underwater. The hydraulic rod system precisely adjusts the opening and closing width and size of the hydraulic gripper, realizing adaptive adjustment to the target object, and realizing rapid and stable gripping of the target object body part. During the operation of the hydraulic gripper, the gap filling and covering unit works synchronously, and based on the data fed back by the pressure sensor, the gear ejector acts to fill air into the unit, realizing full-angle flexible hugging and gripping of the target object. The variable stiffness covering array precisely fills the remaining gap, so that the inner side of the hydraulic gripper fully covers the surface of the target object, realizing micro-angle hugging and wrapping of the gap, and finally realizing full-angle flexible hugging and gripping.

[0123] 23 and 25 are flowcharts showing a control method for an underwater submersible robot according to one embodiment of the present invention. The control method for the underwater submersible robot is executed by an underwater submersible robot control device 9 used for the underwater submersible robot described in any of the above embodiments, and includes steps S10 to S14.

[0124] S10, calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body.

[0125] S11, inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface.

[0126] S12, the output value is respectively set as the input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the terminal sliding mode control law output.

[0127] S13, performing power allocation for each propeller of the propeller component based on the virtual force to obtain thrust of each propeller.

[0128] S14, controlling the propulsion operation of the propeller based on the thrust.

[0129] In an embodiment of the present invention, a first difference value between the current actual posture and the desired posture of the robot body and a second difference value between the actual depth and the desired depth are calculated, the first difference value and the second difference value are input to a terminal sliding mode surface to be set, and an output value of the terminal sliding mode surface is obtained. The output value is respectively input to a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are respectively used as compensation inputs of the terminal sliding mode control law, so that data deviation compensation can be performed for the current underwater turbulence, so that the terminal sliding mode control law can finally output a more accurate virtual force, and the power allocation of each propeller of the propeller component is performed based on the virtual force, the thrust of each propeller is obtained, and the operation control of the propeller of the underwater submersible robot is performed based on the thrust, so that a more reasonable motion control can be performed for the current underwater turbulence. As can be seen, in the embodiments of the present invention, the underwater submersible robot can improve its ability to resist interference with underwater turbulence, and can realize the autonomous stabilization of the robot body in such a complex environment and the autonomous escape operation in an emergency environment, effectively realizing a strong autonomous anti-interference function, improving the operational safety and stability of the equipment, and reducing the probability of equipment accidents. Furthermore, based on the control policy, the underwater drone can be operated intelligently, greatly improving the inspection efficiency of underwater construction, effectively improving the safety of underwater construction inspection work, eliminating the need for divers to directly participate in the work, and greatly reducing the risks faced by workers when working in an underwater environment.

[0130] As an improvement of the above aspect, the power allocation of each propeller of the propeller component based on the virtual force and obtaining the thrust of each propeller is The method includes filtering the virtual force using a second-order filter, and converting the filtered virtual force into thrust for each propeller of the propeller component through power allocation.

[0131] It should be noted that the contents of the relevant aspects of the embodiment of the control method for the underwater submersible robot can be referred to the contents of the embodiment of the control device 9 for the underwater submersible robot, and the explanation thereof will be omitted here.

[0132] 26 is a schematic diagram of a control device 9 for an underwater submersible robot according to one embodiment of the present invention. The control device 9 for an underwater submersible robot of this embodiment includes a processor 1000, a memory 1001, and a computer program stored in the memory 1001 and executable by the processor 1000, such as a control program for an underwater submersible robot. When the processor 1000 executes the computer program, it realizes the steps of the embodiments of the control method for each of the underwater submersible robots described above.

[0133] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to accomplish the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the computer program in the control device 9 of the underwater submersible robot.

[0134] The underwater submersible robot control device 9 may include, but is not limited to, a processor and a memory. As can be understood by those skilled in the art, the schematic diagram is merely an example of the underwater submersible robot control device 9 and does not constitute a limitation of the underwater submersible robot control device 9, and may include more or fewer parts than those shown in the diagram, or a combination of some parts, or different parts, for example, the underwater submersible robot control device 9 may further include input / output devices, network access devices, buses, etc.

[0135] The so-called processor may be a Central Processing Unit (CPU), other general-purpose processor, Digital Signal Processor (DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The general-purpose processor may be a microprocessor, or any general processor, etc., and the processor is the control center of the control device 9 of the underwater submersible robot, and connects each part of the control device 9 of the entire underwater submersible robot using various interfaces and lines.

[0136] The memory can be used to store the computer programs and / or modules, and the processor operates or executes the computer programs and / or modules stored in the memory and calls up the data stored in the memory to realize various functions of the control device 9 of the underwater submersible robot. The memory may mainly include a program storage area and a data storage area, and the program storage area can store an operating system, an application program required for at least one function, and the data storage area can store data created by the use of a mobile phone. The memory may include a high-speed random access memory, and may further include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk memory device, a flash memory device, or other volatile solid-state memory device.

[0137] The modules / units integrated in the control device 9 of the underwater submersible robot may be realized in the form of software functional units and stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the present invention may realize all or part of the processes in the methods of the above embodiments, and may be completed by instructing related hardware by a computer program, and the computer program may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of each of the above method embodiments can be realized. Here, the computer program includes a computer program code, and the computer program code may be in a source code format, an object code format, an executable file, or some intermediate format, etc. The computer readable medium may include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier wave signal, an electrical communication signal, a software distribution medium, and the like. Note that what is included in the computer readable medium may be increased or decreased as required by the legislation and patent practice in a jurisdiction. For example, in some jurisdictions, the legislation and patent practice of the computer readable medium may not include electrical carrier wave signals and electrical communication signals.

[0138] It should be noted that the above-described embodiment of the device is merely illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed among multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. In addition, in the drawings of the embodiment of the device according to the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically realized as one or more communication buses or signal lines. Those skilled in the art can understand and implement the present invention without any creative effort.

[0139] It should be pointed out that the above are preferred embodiments of the present invention, and those skilled in the art can make some improvements and refinements without departing from the principle of the present invention, and these improvements and refinements can also be regarded as the protection scope of the present invention. [Explanation of symbols]

[0140] Robot body 1, inspection device 2, propeller component 7, sensing device 8, control device 9, main body chamber 10, inspection head cover 101, connection part 102, control chamber casing 103, irregular chamber 104, control chamber cover 105, battery chamber cover 106, battery chamber casing 107, chamber cover transition part 108, handle 109, alloy skeleton 20, head skeleton 201, middle skeleton 202, tail skeleton 203, streamlined casing 206, motion assistance module 3, fixed bracket 301, motor 302, spoiler 303, adaptive magnetization module 4, lifting plate 401, first small hydraulic rod member 40 2, second small hydraulic rod member 403, magnetized fixed module 404, receiving plate 405, second large hydraulic rod member 406, first large hydraulic rod member 407, adaptive adjustment hugging mechanism 408, bottom connection skeleton 409, mounting bottom plate 410, underwater vision matrix module 5, push rod motor 501, clamp plate 502, matrix sleeve ring 503, waterproof motor 504, waterproof camera 505, motion frame 506, hinge motion frame 507, variable stiffness and flexibility hugging self-stabilizing module 6, fixed main frame 601, hydraulic gripper 60, variable stiffness covering array 607, gap filling covering unit 608, carbon fiber intermediate layer 609

Claims

1. An underwater submersible robot, The robot body, an inspection device provided on the robot body for inspecting an underwater target; a propeller component provided on the robot body for driving the underwater submersible robot underwater; A sensing device provided on the robot body for sensing the posture and depth of the robot body; A control device is provided on the robot body and is connected to the inspection device, the sensing device, and the propeller component; and Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is respectively set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the output of the terminal sliding mode control law; Performing a power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; a control device for controlling a propulsive operation of the propeller based on the thrust, The robot body includes a main body chamber, two alloy skeletons, and two streamlined casings, the two alloy skeletons are provided on both sides of the main body chamber so as to be rotatable around the axial direction of the main body chamber, and the two streamlined casings surround the two alloy skeletons in a one-to-one correspondence and are connected to the two alloy skeletons in a corresponding manner; The main body chamber includes an inspection head cover, a connection part, a control chamber casing, a special-shaped chamber, a control chamber cover, a battery chamber cover, a battery chamber casing, a chamber cover transition part, and a handle, which are connected in sequence along a longitudinal direction, The inside of the control chamber casing is used to accommodate and install the control device, the inside of the inspection head cover is used to accommodate and install the inspection device, and the inside of the battery chamber cover is used to accommodate and install the battery chamber; The portion of the connecting part that contacts the control chamber casing and the portion that contacts the inspection head lid are both provided with annular grooves along the outer periphery of the connecting part, and an O-seal ring is attached to the annular groove. The portion of the connecting part protrudes into the inspection head lid and the irregular chamber and is in close contact with the interior of the irregular chamber. A set screw is provided at the portion of the connecting part that protrudes into the inspection head lid and the irregular chamber. The portion of the special-shaped chamber that contacts the control chamber casing is provided with an annular groove in which an O-seal ring is attached, and a plurality of mesas for attaching water-sealing contacts are recessed and formed on the outer periphery of the special-shaped chamber, the sensing device is provided in the special-shaped chamber, and a lead wire is led out through the water-sealing contact, a receiving groove is formed in the middle of the control chamber cover for accommodating a male plug connected to the battery chamber cover, and a receiving groove is formed in the middle of the battery chamber cover for accommodating a female plug into which the male plug is inserted; an O-seal ring is provided between the battery chamber cover and a portion connected to the battery chamber casing; a battery platform is placed inside the battery chamber cover, and the battery platform is used to mount a battery; an O-seal ring and a set screw are installed at a connection point between one end of the chamber lid transition part and the battery chamber casing, and the handle is installed at the other end of the chamber lid transition part.

2. When the control device is used to allocate power to each propeller of the propeller component based on the virtual force and obtain thrust of each propeller, specifically, 2. The underwater submersible robot according to claim 1, wherein a secondary filter is used to filter the virtual force, and the filtered virtual force is converted into thrust of each propeller of the propeller component by power allocation.

3. 2. The underwater submersible robot according to claim 1, wherein the propeller component includes four propellers, the four propellers being rotatably mounted on both ends of each side of the robot body and connected to the control device.

4. the alloy skeleton includes a head skeleton, a middle skeleton, and a tail skeleton that are sequentially installed along a longitudinal direction of the main body chamber, the head skeleton, the middle skeleton, and the tail skeleton are connected to both ends of the main body chamber via the connecting parts, and the main body chamber is connected to the corresponding streamlined casing via the head skeleton, the middle skeleton, and the tail skeleton; 2. The underwater submersible robot according to claim 1, wherein an expansion skeleton is installed at the bottom of the main chamber, and hole positions are left in the expansion skeleton for mounting expansion modules.

5. An underwater submersible robot, The robot body, an inspection device provided on the robot body for inspecting an underwater target; a propeller component provided on the robot body for driving the underwater submersible robot underwater; A sensing device provided on the robot body for sensing the posture and depth of the robot body; A control device is provided on the robot body and is connected to the inspection device, the sensing device, and the propeller component; and Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is respectively set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the output of the terminal sliding mode control law; Performing a power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; a control device for controlling a propulsive operation of the propeller based on the thrust, The underwater submersible robot further includes a mounting base plate and at least two adaptive magnetization modules; The mounting base plate is used for detachably connecting to the robot body, and at least two of the adaptive magnetization modules are connected to both sides of the mounting base plate; The adaptive magnetization module includes a lift plate, a first small hydraulic rod member, a second small hydraulic rod member, a magnetization fixing module, a support plate, a second large hydraulic rod member, a first large hydraulic rod member, an adaptive adjustment hugging mechanism, and a bottom connecting skeleton; The angles formed by the two sides of the support plate and the central part of the support plate are obtuse angles, and the bottom sides of the two sides of the support plate are respectively connected to the magnetization fixing module via at least one pair of the first small hydraulic rod member and the second small hydraulic rod member, and the first small hydraulic rod member is hydraulically driven and connected to the second small hydraulic rod member; An underwater submersible robot, characterized in that the top sides of both sides of the support plate are connected to the mounting bottom plate via at least one pair of the second large hydraulic rod member and the first large hydraulic rod member, and the first large hydraulic rod member is hydraulically driven and connected to the second large hydraulic rod member.

6. The underwater submersible robot further includes two pairs of motion assistance modules, the two pairs of motion assistance modules being provided on both sides of the robot body; 6. The underwater submersible robot of claim 1, wherein the motion assistance module includes a fixed bracket, a motor, and a spoiler, the fixed bracket is provided on one side of the robot body, the spoiler is provided on the fixed bracket so as to be rotatable up and down, and the motor is provided on the robot body and is used to drive the spoiler to move up and down.

7. An underwater submersible robot, The robot body, an inspection device provided on the robot body for inspecting an underwater target; a propeller component provided on the robot body for driving the underwater submersible robot underwater; A sensing device provided on the robot body for sensing the posture and depth of the robot body; A control device is provided on the robot body and is connected to the inspection device, the sensing device, and the propeller component; and Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is respectively set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the output of the terminal sliding mode control law; Performing a power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; a control device for controlling a propulsive operation of the propeller based on the thrust, The inspection device includes an underwater vision matrix module, the underwater vision matrix module includes a matrix sleeve ring and at least two sets of camera adjustment components, the matrix sleeve ring is used for removably mounting on the robot body, and the at least two sets of camera adjustment components are uniformly distributed on the same side of the matrix sleeve ring; the camera adjustment component includes a push rod motor, a clamp plate, a waterproof motor, a waterproof camera, a motion frame and a hinge motion frame, the bottom end of the hinge motion frame is hingedly connected to the row and column sleeve ring, the waterproof motor is provided at the top end of the hinge motion frame, the waterproof camera is provided at the free end of the rotation shaft of the waterproof motor, one end of the clamp plate is connected to the middle of the hinge motion frame, the other end of the clamp plate is connected to one end of the motion frame, the other end of the motion frame is connected to one end of the clamp plate of an adjacent another camera adjustment component, and the other end of the clamp plate of the another camera adjustment component is connected to the middle of the hinge motion frame of the camera adjustment component, the push rod motor is provided on the row and column sleeve ring, and the free end of the push rod of the push rod motor is connected to the middle of the motion frame.

8. An underwater submersible robot, The robot body, an inspection device provided on the robot body for inspecting an underwater target; a propeller component provided on the robot body for driving the underwater submersible robot underwater; A sensing device provided on the robot body for sensing the posture and depth of the robot body; A control device is provided on the robot body and is connected to the inspection device, the sensing device, and the propeller component; and Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is respectively set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the output of the terminal sliding mode control law; Performing a power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; a control device for controlling a propulsive operation of the propeller based on the thrust, The underwater submersible robot further includes a variable stiffness flexible hugging self-stabilizing module, the variable stiffness flexible hugging self-stabilizing module including a fixed main frame, a pair of hydraulic grippers, a variable stiffness covering array, a gap-filling covering pneumatic unit, and a carbon fiber intermediate layer; an underwater submersible robot, characterized in that the fixed main frame is used for being detachably connected to the robot body, a pair of hydraulic grippers are respectively provided on both sides of the fixed main frame so as to be capable of moving toward or away from each other, the variable stiffness covering array is provided on both sides of the inner free ends of the pair of hydraulic grippers, the gap-filling covering pneumatic unit is provided in the middle of the inner sides of the pair of hydraulic grippers, and the gap-filling covering pneumatic unit is covered with the carbon fiber intermediate layer.

9. A method for controlling an underwater submersible robot used in the underwater submersible robot according to any one of claims 1 to 8, comprising the steps of: Calculating a first difference value between a current actual posture and a desired posture of the robot body and a second difference value between an actual depth and a desired depth of the robot body; inputting the first difference value and the second difference value into a terminal sliding mode surface to be set, and obtaining an output value of the terminal sliding mode surface; The output value is set as an input of a pre-set high-order observer, a radial basis function neural network and a terminal sliding mode control law, respectively, and the output of the high-order observer and the output of the radial basis function neural network are set as compensation inputs of the terminal sliding mode control law, to obtain a virtual force of the output of the terminal sliding mode control law; performing power allocation for each propeller of the propeller component based on the virtual force to obtain a thrust for each of the propellers; and controlling the propulsion operation of the propeller based on the thrust.

10. 10. A control device for an underwater submersible robot, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the control method for an underwater submersible robot according to claim 9 is realized.

Images