A wind turbine blade surface adaptive crawling robot and a control system thereof

Abstract

The application discloses a wind turbine blade surface self-adaptive crawling robot and a control system thereof, and belongs to the field of wind turbine blade surface cleaning robots. The crawling robot comprises a mechanical structure and a control system. The mechanical structure comprises a main body structure and six under-actuated leg structures. The leg structure is composed of four motors, connecting rods, two-stage transmission structures, passive centering structures and suction cup structures. The main body is provided with a storage battery, an embedded microprocessor, a vacuum generation and destruction module composed of an air pump, an electromagnetic valve, a relay switch and a negative pressure sensor. The robot realizes adsorption and crawling on the blade surface based on the vacuum adsorption principle. The control system is composed of a finite state machine module, a trajectory calculation module, a control execution module and a state sensing module. The control of the robot is realized through the acceptance of instructions, sensing, adaptive adjustment and movement. The under-actuated structure and the modular control system are designed, so that the self-adaptation and omnidirectional movement of the wind turbine blade surface variable curvature are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of mobile robots, in particular to a wind turbine blade surface adaptive climbing robot and a control system thereof. BACKGROUND

[0002] The wind turbine is an important device for converting wind energy into electric energy, and the wind turbine blade is one of the core parts of the wind turbine, and its shape and structure directly determine the conversion efficiency. Because its working environment is usually harsh, it is affected by natural factors such as strong wind, rain and snow, dust, lightning, etc., and the blade is prone to defects such as corrosion, cracking, wear and tear, and fracture. According to statistics, more than 40% of the total failure of the generator in service causes the loss of power generation due to the failure of the blade. In order to ensure the healthy operation of the blade, it needs to be regularly inspected and maintained.

[0003] The current maintenance method is divided into non-contact and contact, the former uses unmanned aerial vehicles and other equipment to shoot and analyze the blade, this method has the characteristics of high efficiency and safety, but it cannot approach the surface of the blade, and can only detect the most surface obvious damage, it cannot identify small cracks and internal defects, and also cannot be repaired. The contact method currently still relies on manual operation, using high-altitude operation, the detection personnel use handheld ultrasonic detection equipment to detect the surface of the blade. This method has great risk and is inefficient. However, the existing high-altitude operation robot can only adapt to the climbing motion of the plane or the surface with insignificant curvature change, and part of the robots only rely on the mechanical structure of the leg to passively adapt to the curvature, and cannot be applied to the scene of large curvature change of the wind turbine blade surface. SUMMARY

[0004] The present application aims at the deficiencies of the prior art, and provides a wind turbine blade surface adaptive climbing robot and a control system, which can realize curvature adaptation and omnidirectional motion on the surface of the wind turbine blade. The robot can carry detection equipment to approach detection and repair work on the blade.

[0005] The purpose of the present application is achieved by the following technical scheme: a wind turbine blade surface adaptive climbing robot, comprising a main body structure and six underactuated leg structures;

[0006] The underactuated leg structure is composed of four motors, a connecting rod driven by the motors, a two-stage transmission structure, a passive centering structure and a suction cup; the two-stage transmission structure is composed of a synchronous belt and a four-bar linkage transmission, and the passive centering structure is connected at the end; the suction cup is installed at the end of the leg structure through the passive centering structure;

[0007] The robot body is composed of a structure and a functional module, the structure is composed of two upper and lower plates and side plates, and the functional module installed in the structure includes a battery for supplying power to the robot, a micro-embedded processor for running a control system, and a vacuum generation and destruction system composed of an air pump, an electromagnetic valve, a switch relay, and a negative pressure sensor.

[0008] Further, the motor includes a hip joint motor, a crotch joint motor, a knee joint motor, and an ankle joint motor, the leg structure is connected with the main body structure through the hip joint motor, the crotch joint motor and the knee joint motor are respectively installed on both sides of the hip link driven by the hip joint motor, and the ankle motor is installed on the crotch link driven by the crotch joint.

[0009] Further, the two-stage transmission structure includes a first-stage transmission system composed of a driving synchronous wheel, a synchronous belt, and a driven synchronous wheel, and a second-stage transmission system composed of a driving link, a connecting link, and a driven link.

[0010] The driving synchronous wheel is installed on a flange shaft between the crotch joint motor and the knee joint motor, and drives the driven synchronous wheel installed on the knee joint link through the synchronous belt.

[0011] The driving link is installed on a flange shaft connected with the output shaft of the ankle joint motor, and drives the driven link connected with the ankle joint to control the angle of the suction cup.

[0012] Further, the passive return structure is composed of an ankle link and four tension springs, the ankle link is connected with the end of the transmission structure, a spherical hinge is installed at the end of the ankle link, the spherical hinge is connected with the suction cup through a thread, and the tension springs are used to connect the suction cup and the ankle link.

[0013] Further, in the vacuum generation and destruction system, there are three air pumps, two of which are used to generate negative pressure, and the suction cups of the three leg structures and a negative pressure sensor are connected to form two independent circuits, and the other air pump is used to generate positive pressure and is connected in the two circuits through two electromagnetic valves, the negative pressure pump is opened, the electromagnetic valve and the positive pressure pump are closed to generate vacuum adsorption force, the negative pressure pump is closed, and the electromagnetic valve and the positive pressure pump are opened to quickly destroy the vacuum adsorption force.

[0014] According to another aspect of the description, a wind turbine blade surface adaptive climbing robot control system of the robot is also provided, which includes: a control system hardware composed of a micro-embedded processor and an expansion interface board, the micro-embedded processor runs a control system, and the expansion interface board connects the motor, the switch relay, and the negative pressure sensor with the micro-embedded processor.

[0015] The control system software functions are realized by four modules: finite state machine module, trajectory calculation module, control execution module, and state perception module, which process user instructions, perceive the robot state, calculate adaptive and motion instructions, and realize adaptive and omnidirectional motion of the robot on the variable-curvature surface of the blade.

[0016] Further, the processing flow of each module in the control system software functions includes:

[0017] The finite state machine module accepts user instructions and updates the leg state, the state perception module continuously provides robot joint angles and negative pressure sensor values, the trajectory calculation module accepts information from the previous two modules, first calculates the position of the foot end in the robot body coordinate system, then fits the plane equation of the terrain in the current foot area of the robot, adjusts the body to be parallel to the plane and maintains a fixed height, calculates the foot end trajectory for motion following the input instructions, the control execution module accepts the foot end trajectory, uses a velocity-based position controller to calculate the desired joint angle, the finite state machine module controls the robot to follow the instructions to run for one gait cycle and then stop, waiting for the next instruction; during the cycle, the robot can run at any time to interrupt and execute new instructions.

[0018] Further, the calculation of the position of the foot end in the robot body coordinate system and the fitting of the plane equation of the terrain in the current foot area of the robot include:

[0019]

[0020] where L i represents the length of the ith link, where c i = cos(i), s i = sin(i), c ij = cos(i+j), s ij = sin(i+j), where R e τ , B e τ represent the position of the foot end of the ith leg in the R coordinate system and the B coordinate system, respectively, where R is located at the center of the robot body and B is located at the center of the first motor of the leg; after calculating the positions of the six legs in the R system according to the above formula, the inverse matrix is used to calculate the plane equation:

[0021]

[0022] where the six rows of the A matrix represent the positions of the six leg ends in the R system, and n is the normal vector of the plane equation.

[0023] Further, in the step of adjusting the body to be parallel to the plane and keep the height fixed, the target pose of the body adjustment is represented by a transition matrix, and the calculation formula comprises:

[0024]

[0025] wherein R T R' represents a transition matrix from the target pose R' to the current pose R, x R′ ,y R′ ,z R′ represent the x, y and z axes of the target pose respectively, r and r' represent the origins of the coordinate systems of the current pose and the target pose respectively, p represents the intersection point of the extension line of the z axis of the target pose and the plane calculated in the previous step, and h is set as the height of the robot body from the plane.

[0026] Further, in the step of calculating the foot trajectory for the movement following the input instruction, the control execution module receives the foot trajectory, and calculates the joint expected angle using a velocity-based position controller, and specifically comprises: calculating the foot trajectory for the movement corresponding to the input instruction, and the calculation equation is as follows

[0027] B e τ (t)=( R T B ) -1 ( R T R′ ) -1 ( R e τ )

[0028] wherein R T R' firstly uses the transition matrix calculated in the previous step, and after the calculation, uses the transition matrix calculated according to the input instruction, the input instruction being the movement direction and the yaw angle [v x ,v y ,w z ] of the robot, and the calculation formula is as follows:

[0029]

[0030] The control execution module receives the foot trajectory, and calculates the angle instruction of the joint according to the current position of the foot, and the angles θ of the first three motors = [θ1, θ2, θ3] are calculated by the following formula:

[0031]

[0032] wherein, is B the derivative of e(t) with respect to time, which represents the theoretical speed of the foot, and Kp is a control parameter that can be customized, B e cur represents the current position of the foot end, represents the desired speed calculated according to the theoretical speed of the foot end and the current actual position of the foot end, represents the joint angular velocity, θ cur represents the current angle of the joint, θ(t) represents the joint angle, J * is the inverse of the Jacobian matrix, calculated by the formula: J * = J T (JJ T +λI) -1 The fourth motor controls the orientation of the suction cup to always be parallel to the suction plane, and is calculated by the following formula:

[0033] θ 4,des =|θ 3,cur |-|θ 2,cur |-π / 2

[0034] Where, θ 2,cur , θ 3,cur represent the current angles of the hip joint motor and the knee joint motor, respectively, θ 4,des represents the desired angle of the ankle joint motor, and the control system is connected to the motor through an expansion interface to control the rotation angle, realizing curvature adaptation and omnidirectional motion.

[0035] Advantages of the present application:

[0036] 1. The present application integrates the air pump, solenoid valve, relay, battery, embedded computing device and expansion board required by the crawling robot on the robot body, so that the robot can move without relying on external air source and power supply, improving the convenience of robot deployment.

[0037] 2. The leg structure proposed in the present application has seven degrees of freedom, four of which are independently driven by motors, and three passive degrees of freedom have automatic centering effect, which can make the suction cup more flexible in adsorption direction, in addition, the orientation of the suction cup is independently controlled by one motor, which improves the adsorption and adaptation ability of the suction cup to the curved surface; and among the four motors, two motors are driven by synchronous belts and connecting rods, so that they can be installed closer to the main body, reducing the moment of inertia during leg movement and facilitating the control of motor angle.

[0038] 3. The variable curvature adaptation method proposed in the present application does not rely on external sensors such as cameras and radars, but characterizes the current robot area and adjusts the pose according to the motor angle, reducing the dependence on external sensors, reducing the power consumption of the robot and improving the load capacity.

[0039] 4. The application provides a modular control system, which can realize variable-curvature self-adaption and omni-directional movement simultaneously, and the robot can automatically adjust the body pose according to the curvature change during movement, so that the control difficulty is reduced, and movement in any direction is supported, and the coverage ability of the inspection area on the blade surface is improved. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 A whole structure diagram of the crawling robot provided for the application example

[0041] Figure 2 A body structure diagram provided for the application example

[0042] Figure 3 A leg structure diagram provided for the application example

[0043] Figure 4 A control system block diagram provided for the application example

[0044] Figure 5 Simulation results in Gazebo software provided for the application example

[0045] Figure 6 Results of the robot body pose change with time provided for the application example

[0046] Figure 7 Calculation results of the robot foot end trajectory provided for the application example DETAILED DESCRIPTION

[0047] The application example provides a wind turbine blade surface self-adaptive crawling robot, which comprises a mechanical structure and a control system.

[0048] The whole mechanical structure of the robot is shown in Figure 1 , and comprises:

[0049] The robot body structure 1; the explosion view of the robot body structure is shown in Figure 2 , and the robot body is composed of structural members and functional modules; the structural members are composed of an upper carbon fiber plate 3, a lower carbon fiber plate 4 and six lateral plates; the lateral plates are composed of a front baffle 5, a rear baffle 6 and four opposite side baffles 7. The functional modules installed in the structural members include a storage battery 13, a micro embedded processor 14, a vacuum generation and destruction system composed of an electromagnetic valve 8, an air pump 9, an expansion board 10, a negative pressure sensor 11 and a switch relay 12.

[0050] Six leg structures 2 installed on the body structure; the leg structure of the robot is shown in Figure 2As shown, the leg structure has four motors: a hip joint motor 15, a waist joint motor 16, a knee joint motor 17, an ankle motor 18, and four connecting rods: a hip connecting rod 19, a waist connecting rod 20, a knee connecting rod 21, and an ankle connecting rod 22. In addition, there is a first-stage transmission system composed of a driving synchronous wheel 30, a synchronous belt 35, and a driven synchronous wheel 33, and a second-stage transmission system composed of a driving connecting rod 36, a connecting rod 27, and a driven connecting rod 38. In addition, there is a passive centering mechanism composed of a tension spring 23 and the ankle connecting rod 22, which is used to connect a suction cup 24.

[0051] Specifically, the leg structure of the robot is connected to the upper carbon fiber plate 3 of the robot main body structure through the hip joint motor 15, is installed on the lower carbon fiber plate 4 through the hip connecting piece 25, one end of the hip connecting rod 19 is directly connected to the output shaft of the hip joint motor 15

[0052] The other end is installed at the screw bearing 26 to realize the rotation control of the hip connecting rod 19 by the hip joint motor 15; the waist joint motor 16 and the knee joint motor 17 are respectively installed on both sides of the hip connecting rod; the output shaft of the waist joint motor is directly connected to the waist connecting rod 20 to realize the rotation control; the flange shaft 28 is supported and fixed by the rib bearing 29 and the rib bearing 27, connects the output shaft of the knee joint motor 17 and the driving synchronous wheel 30, the driving synchronous wheel 30 drives the driven synchronous wheel 33 installed on the knee joint connecting rod 21 through the synchronous belt 35, and the rotation of the knee joint 21 is realized by the knee joint motor 17 through the synchronous belt transmission; the ankle joint motor is installed on the waist connecting rod 20, the output shaft is connected to the small flange shaft 32 supported by the rib bearing 31 and the ball bearing 33; the other end of the small flange shaft 32 is connected to the driving connecting rod 36 in the four-bar linkage transmission, the driving connecting rod 36 drives the driven connecting rod 38 connected to the ankle joint 22 through the connecting rod 27 to realize the control of the angle of the suction cup. The end of the ankle connecting rod 22 is a spherical hinge, which is connected to the suction cup 24 through a threaded connection, and the four tension springs 23 are connected between the suction cup 24 and the ankle connecting rod 22 to realize the passive self-adaptive effect of external force. After the external force is removed, it will automatically return to the center.

[0053] Specifically, there are three air pumps 9, two electromagnetic valves 8, and two negative pressure sensors 11 for controlling the generation and destruction of vacuum. The air pumps 9 located on both sides are respectively connected to the suction cups 24 of the three leg structures 2, one negative pressure sensor 11, and one electromagnetic valve 8, the other end of the electromagnetic valve 8 is connected to the air pump 10 in the middle, and the opening and closing of the air pump 9 and the electromagnetic valve 8 are controlled by the switch relay 12; when the air pumps 9 on both sides are opened and the electromagnetic valve 8 is closed, negative pressure will be generated inside the suction cup 24, forming an adsorption force; when the air pumps 9 on both sides are closed and the electromagnetic valve 8 is opened, the air pump 9 in the middle is opened, positive pressure will be generated inside the suction cup 24, the vacuum is destroyed, and the suction cup 24 can be smoothly detached from the surface.

[0054] According to the device described above, the application also provides a robot control system, comprising:

[0055] The robot control system hardware includes a micro-embedded processor 14, an expansion board 10; wherein the micro-embedded processor 14 runs the control system of the robot, and the expansion board 10 connects all the motors, negative pressure sensors 11, switch relays 12 with the embedded microprocessor.

[0056] The control system software framework is as shown in Figure 4 The control system software functions are realized by four modules: a finite state machine module, a trajectory calculation module, a control execution module, and a state perception module, which can process user instructions, perceive the state of the robot, calculate adaptive and motion instructions, and realize the adaptive and omnidirectional motion of the robot on the variable-curvature surface of the blade.

[0057] Specifically, the control system is composed of four functional modules as shown in Figure 4 which realizes the adaptive omnidirectional motion of the robot on the variable-curvature surface, and the implementation process includes: the finite state machine module accepts user instructions and updates the leg state, the state perception module continuously provides the robot joint angle and negative pressure sensor value, the trajectory calculation module accepts the information of the previous two modules, first calculates the position of the foot end in the robot body coordinate system, then fits the plane equation of the terrain in the current foot area of the robot, and adjusts the body to be parallel to the plane and maintains a fixed height, calculates the foot end trajectory following the input instruction, the control execution module accepts the foot end trajectory, uses a velocity-based position controller to calculate the desired joint angle, the finite state machine module controls the robot to follow the instruction to run for one gait cycle and then stop, waiting for the next instruction, during this period, the robot can be interrupted at any time and the new instruction can be executed again;

[0058] First, the foot end position is calculated, and the plane equation of the area under the current foot of the robot is fitted, and the calculation formula is as follows:

[0059]

[0060] wherein L i represents the length of the i-th connecting rod, wherein c i = cos(i), s i = sin(i), c ij = cos(i+j), s ij = sin(i+j), wherein R e τ , B e τ represents the position of the foot end of the τ-th leg in the six legs in the R coordinate system and the B coordinate system, respectively, wherein R is located at the center of the robot body, and B is located at the center of the first motor of the leg; after calculating the positions of the six legs in the R system according to the above formula, the inverse matrix is used to calculate the plane equation:

[0061]

[0062] where the six rows of matrix A represent the positions of the six leg ends in the R system, n is the normal vector of the plane equation, which can be expressed as n·[x, y, z] T = 1, and 1 represents a vector composed of six 1s; then the target pose of the body adjustment can be obtained according to the plane equation, and a transition matrix is used to represent the calculation formula as follows:

[0063]

[0064] where R T R' represents the transition matrix from the target pose R' to the current pose R, and the x R′ ,y R′ ,z R′ of the transition matrix represent the x, y and z axes of the target pose, respectively, and r and r' represent the origins of the current pose and the target pose coordinate systems, respectively, and p represents the intersection of the extension line of the z axis of the target pose and the plane calculated in the previous step, and h is set as the height of the robot body and the plane;

[0065] The foot end trajectory corresponding to the input instruction is calculated, and the calculation equation is as follows

[0066] B e τ (t)=( R T B ) -1 ( R T R′ ) -1 ( R e τ )

[0067] where R T R' First, the transition matrix calculated in the previous step is used to complete the calculation, and then the transition matrix calculated according to the input instruction is used, and the input instruction is the motion direction and yaw angle [v x ,v y ,w z ] of the robot, and the calculation formula is as follows:

[0068]

[0069] The control execution module accepts the foot end trajectory and calculates the angle instruction of the joint according to the current position of the foot end, and the angles θ = [θ1, θ2, θ3] of the three motors of the hip joint motor 15, the waist joint motor 16 and the knee joint motor 17 are calculated by the following formula:

[0070]

[0071] wherein is B e(t) is the derivative of the theoretical speed of the foot end, K p is a control parameter that can be defined by the user, B e cur represents the current position of the foot end, represents the desired speed calculated from the theoretical speed of the foot end and the current actual position of the foot end, represents the joint angular velocity, θ cur represents the current angle of the joint, θ(t) represents the joint angle, J * is the inverse of the Jacobian matrix, calculated by the formula: J * = J T (JJ T + λI) -1 The angle of the fourth motor controls the orientation of the suction cup, which is always parallel to the suction plane, calculated by the following formula:

[0072] θ 4,des = |θ 3,cur | - |θ 2,cur | - π / 2

[0073] wherein, θ 2,cur , θ 3,cur represent the current angles of the hip joint motor 16 and the knee joint motor 17, respectively, θ 4,des represents the desired angle of the ankle joint motor 18, the control system is connected to the motor through an expansion interface to control its rotation angle, realizing curvature adaptation and omnidirectional motion. When controlling the motor angle, the execution control module checks the control command at a frequency of 100 Hz, and if a stop command is issued, the current motion will be stopped, and after receiving the stop command, the above steps will be re-executed according to the newly received command.

[0074] In this embodiment, the robot mechanism and control system are also simulated; a multi-rigid-body dynamics simulation software is used to establish a robot model and a wind turbine blade segment model to verify the pose adjustment and omnidirectional motion capability of the control system. The main simulation process is as follows:

[0075] (1) Simplify the robot model, use a rigid body to represent the joint composed of multiple parts, retain the original mass, center of mass, etc. characteristics, import the model into the dynamics simulation software;

[0076] (2) Establish a blade segment model, the curvature radius of the blade model varies from 1.4m to 20m;

[0077] (3) Place the robot model on the leading edge of the blade and input three control commands: linear velocity only, angular velocity 0; angular velocity only, linear velocity 0; and both linear velocity and angular velocity are not 0. The simulation diagram and the robot's motion process are as follows: Figure 5 As shown;

[0078] (4) When the robot's foot contacts the blade, a force perpendicular to the suction cup is set to simulate the magnitude of the suction force;

[0079] (5) Observe the change in the robot's center of mass pose over time to see if it proceeds as planned. The result of the robot's center of mass change over time is as follows: Figure 6 As shown, observe whether the robot's foot trajectory is installed with the desired trajectory following. Select one foot of the robot and plot its trajectory change over time. Figure 7 As shown, the robot's omnidirectional adaptive motion capability with variable curvature is verified. Figure 6 This demonstrates how the robot's body posture changes over time during movement, θ real ,θ des φ represents the robot's actual pitch angle and desired pitch angle. real ,φ des Represents the robot's actual roll angle and expected roll angle. Figure 6 The first and second graphs on the left represent robots. Figure 5 The subplot below shows the changes in actual and expected pitch and roll angles over time as the robot moves from point A to point C. To avoid the singularity of Euler angles around 90°, the pitch and roll angles in the second graph from the left are referenced to the robot's attitude at point B. Figure 6 As can be observed in the two graphs on the left, the robot continuously adjusts its posture as the curvature changes during movement. Figure 6 The curve on the right is Figure 6 The enlarged view of area A in the first image on the left shows that the robot can control its body posture to follow changes in the desired posture angle, while maintaining the current posture during two adjustments.

[0080] Figure 7 This demonstrates how the position of the robot's feet changes over time during its movement. des x(t) represents the x-components of the desired and actual positions of the robot's foot over time, and y(t) represents the x-components of the desired and actual positions of the robot's foot. des (t),y(t),z des z(t) and z(t) represent the changes of the y-component and z-component of the expected and actual foot positions over time, respectively. Figure 7 The left sub-image represents the robot's left front leg. Figure 5 The curve showing the foot position when moving from point B to point C in the upper sub-image. Figure 7The right sub-diagram represents the corresponding foot end position curve of the robot's right front leg when moving from point C to point D in the upper sub-diagram. Figure 5 The right sub-diagram represents the corresponding foot end position curve of the robot's right front leg when moving from point C to point D in the upper sub-diagram.

[0081] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0082] It is to be understood that the above general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application. The application is not restricted to the precise construction and combinations of parts and steps described above and shown in the accompanying drawings. Various modifications and changes can be made in the structures and steps without departing from the scope of the application. The scope of the application should be determined by the following claims.

Claims

1. An adaptive crawling robot for wind turbine blade surfaces, characterized in that, Includes the main structure and six underactuated leg structures; The underactuated leg structure consists of four motors and their driven linkages, a two-stage transmission structure, a passive centering structure, and a suction cup; the two-stage transmission structure consists of a synchronous belt and a four-bar linkage, with the passive centering structure connected at the end; the suction cup is installed at the end of the leg structure through the passive centering structure. The robot body consists of structural components and functional modules. The structural components consist of upper and lower plates and side plates. The functional modules installed therein include a battery that powers the robot, a micro embedded processor for the operation control system, and a vacuum generation and destruction system consisting of an air pump, solenoid valves, switch relays, and a negative pressure sensor. The motor includes a hip joint motor, a hip joint motor, a knee joint motor, and an ankle joint motor. The leg structure is connected to the main structure through the hip joint motor. The hip joint motor and the knee joint motor are respectively installed on both sides of the hip linkage driven by the hip joint motor, and the ankle motor is installed on the hip linkage driven by the hip joint. The two-stage transmission structure includes a first-stage transmission system consisting of a driving synchronous pulley, a synchronous belt, and a driven synchronous pulley, and a second-stage transmission system consisting of a driving connecting rod, a connecting connecting rod, and a driven connecting rod. The active synchronous pulley is mounted on the flange shaft between the hip joint motor and the knee joint motor, and drives the driven synchronous pulley mounted on the knee joint linkage via a synchronous belt. The active linkage is mounted on the flange shaft connected to the output shaft of the ankle joint motor. The angle of the suction cup is controlled by driving the driven linkage connected to the ankle joint through the connecting linkage.

2. The adaptive crawling robot for wind turbine blades according to claim 1, characterized in that, The passive centering structure consists of an ankle link and four tension springs. The ankle link is connected to the end of the transmission structure, and a ball joint is installed at the end of the ankle link. The ball joint is connected to the suction cup via a thread, and the tension springs are used to connect the suction cup and the ankle link.

3. The adaptive crawling robot for wind turbine blades according to claim 1, characterized in that, In the vacuum generation and destruction system, there are three air pumps. Two of them are used to generate negative pressure, connecting the suction cups of the three leg structures and a negative pressure sensor to form two independent circuits. The third air pump is used to generate positive pressure, which is connected to these two circuits through two solenoid valves. When the negative pressure air pump is turned on, the solenoid valve and the positive pressure air pump are turned off to generate vacuum suction force. When the negative pressure air pump is turned off, the solenoid valve and the positive pressure air pump are turned on to quickly destroy the vacuum suction force.

4. A wind turbine blade surface adaptive crawling robot control system for the robot of claim 1, characterized in that, The control system includes: The control system hardware consists of a micro embedded processor and an expansion interface board. The micro embedded processor runs the control system, and the expansion interface board connects the motor, switch relay, negative pressure sensor and micro embedded processor. The control system software functions are implemented by four modules: finite state machine module, trajectory calculation module, control execution module, and state perception module. The module processes user commands, senses and manages the robot's state, calculates adaptive and motion commands, and realizes the robot's adaptive and omnidirectional motion on the variable curvature surface of the blade.

5. The control system according to claim 4, characterized in that, The processing flow of each module in the control system software includes: The finite state machine module receives user commands and updates the leg state. The state perception module continuously provides robot joint angles and negative pressure sensor values. The trajectory calculation module receives information from the first two modules, first calculates the position of the foot in the robot's body coordinate system, then fits the plane equation of the terrain under the robot's current feet, adjusts the body to be parallel to the plane and maintains a fixed height, and calculates the foot trajectory to follow the input command. The control execution module receives the foot trajectory, uses a velocity-based position controller to calculate the desired joint angle, and the finite state machine module controls the robot to follow the command for one gait cycle before stopping and waiting for the next command. During the cycle, the robot's operation can be interrupted at any time and a new command can be re-executed.

6. The control system according to claim 5, characterized in that, The process of calculating the position of the foot in the robot's body coordinate system and then fitting the planar equation of the terrain in the area currently under the robot's feet includes: ； in Representing the The length of each link, of which , ,in Each represents the first of the six legs. The foot of one leg at coordinate system and Position in coordinate system, where Located at the center of the robot's main body, Located at the center of the first motor in the leg; the six legs are calculated according to the above formula. After determining the position of the plane, the plane equation is calculated using the inverse matrix: ； in The six rows of the matrix represent the ends of the six legs. Its position in the system, It is the normal vector of the plane equation.

7. The control system according to claim 5, characterized in that, The process of adjusting the body to be parallel to the plane and maintaining a fixed height involves a target pose represented by a transition matrix, the calculation formula of which includes: ； in Represents the target pose To the current pose The transition matrix, the transition matrix The x, y, and z axes represent the target pose, respectively. These represent the origins of the coordinate systems for the current pose and the target pose, respectively. This represents the intersection point of the extended z-axis line of the target pose with the plane calculated in the previous step. Set as the height of the robot's body relative to the plane.

8. The control system according to claim 5, characterized in that, The calculation of the foot trajectory following the input command, and the control execution module receiving the foot trajectory and using a velocity-based position controller to calculate the desired joint angle, specifically includes: calculating the foot trajectory corresponding to the input command, with the following calculation equation. ； in First, the transfer matrix calculated in the previous step is used. After the calculation is completed, the transfer matrix calculated based on the input command is used. The input command is the robot's motion direction and yaw angle. The calculation formula is as follows: ； The control execution module receives the foot trajectory and calculates the joint angle command based on the current position of the foot, including the angles of the first three motors. Calculated using the following formula: ； in, yes The derivative with respect to time represents the theoretical velocity of the foot. These are customizable control parameters. Represents the current position of the foot. This represents the expected speed calculated based on the theoretical speed of the foot and the current actual position of the foot. Represents joint angular velocity. Represents the current angle of the joint. Represents joint angle, The converse of the Jacobian matrix is ​​obtained through the formula: The angle of the fourth motor controls the orientation of the suction cup, ensuring it remains parallel to the suction plane. This is calculated using the following formula: ； in, These represent the current angles of the hip joint motor and the knee joint motor, respectively. The desired angle of the ankle joint motor is represented by the control system, which connects to the motor via an expansion interface to control its rotation angle, achieving curvature adaptation and omnidirectional motion.

Images