Insulator outer diameter self-adaptive cleaning method based on unmanned aerial vehicle cluster control

Abstract

The invention discloses an insulator outer diameter self-adaptive cleaning method based on unmanned aerial vehicle cluster control. The method comprises the following steps: analyzing stress and torque conditions during flight based on a mechanical structure of the four-rotor cleaning unmanned aerial vehicle, analyzing dynamic characteristics of the four-rotor cleaning unmanned aerial vehicle on the basis of establishing a proper coordinate system, and establishing a dynamic model of the cleaning unmanned aerial vehicle by using a Newton-Euler equation; based on an unmanned aerial vehicle kinetic model and a cluster cooperative control algorithm, a cluster control method capable of adaptively adjusting the formation of an unmanned aerial vehicle group and the distance between the unmanned aerial vehicle group and an insulator according to measurement and calculation data of a sensor and cooperative information sent between unmanned aerial vehicles is provided, and the control method is applied to different steps in the cleaning process. And the cleaning work of insulators with different outer diameters is realized.

Description

Technical Field

[0001] The present invention relates to a method for cleaning insulators by multi-UAV collaborative operation, and particularly to an adaptive cleaning method for the outer diameter of insulators based on UAV cluster control, belonging to the technical field of power equipment. Background Art

[0002] An insulator is a special insulating control for fixing a live conductor in a transmission line. Since the insulator is in an exposed environment during operation for a long time, dust particles in the surrounding atmosphere are easily attached to its surface to form a fouling layer. The decontamination of substation post insulators mainly relies on manual climbing operations, which have problems such as poor safety and low decontamination efficiency. At the same time, the existing fixed cleaning brush head structure can only clean specific insulators, and it is very necessary to solve the problem of poor adaptability to various types of insulator structures.

[0003] The cluster control of UAVs can be used as one of the solutions for cleaning insulators with different outer diameters. The remote control and flight functions of UAVs can greatly reduce the risk of climbing operations, and the cluster operation of UAVs can perform operations simultaneously by multiple aircraft, shortening the overall task cycle and having dynamic fault tolerance capabilities. For the cleaning tasks of insulators with different outer diameters, different formation shapes or sizes of formation shapes can be formed to execute the cleaning tasks by modifying the desired formation matrix in the formation algorithm.

[0004] Therefore, it is crucial to design an adaptive cleaning method for the outer diameter of insulators based on UAV cluster control. Summary of the Invention

[0005] The purpose of the present invention is to provide an adaptive cleaning method for the outer diameter of insulators based on UAV cluster control. Analyze the dynamic characteristics of a quadrotor UAV with a reasonable cleaning structure and establish a dynamic model. Form a cleaning formation with N cleaning UAVs, and based on the measurement results of UAV sensors and the cluster cooperative control algorithm, complete the cleaning work of insulators with different outer diameters. Solve the technical problem that the existing technology cannot adapt to the cleaning tasks of insulators with different outer diameters.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] An adaptive cleaning method for the outer diameter of insulators based on UAV cluster control includes the following steps:

[0008] Step S1: Form a cleaning formation with N quadrotor cleaning UAVs. Based on the flight principle of the quadrotor cleaning UAV, analyze its dynamic characteristics on the basis of establishing a coordinate system, and use the Newton-Euler equation to establish the dynamic model of the cleaning UAV.

[0009] Step S2: Based on the drone dynamics model, using the cluster cooperative control algorithm, adaptively adjust the formation of the drone swarm and the distance from the insulators according to the measurement data of the sensors and the cooperative information mutually sent between the drones, so as to achieve the cleaning work of insulators with different outer diameters.

[0010] The object of the present invention can also be further realized by the following technical measures:

[0011] Further, in step S1, establish two basic coordinate systems, namely the navigation coordinate system and the vehicle coordinate system, and deduce the transformation matrix for the vehicle coordinate system to be rotated three times to the navigation coordinate system:

[0012]

[0013] where θ is the pitch angle of the drone and ψ is the yaw angle of the drone, is the roll angle.

[0014] Analyze the torque and lift generated by the four rotors in the vehicle coordinate system, comprehensively consider gravity, resistance, water flow reaction force, and the gyroscopic effect of the airframe, and establish a dynamics model by applying Newton's second law:

[0015]

[0016] where I x , I y , I z are the moments of inertia of the cleaning drone when rotating around the three axes of the vehicle coordinate system respectively. U1 represents the resultant force of the lift generated by the four rotors, and U2, U3, and U4 represent the torques on the x, y, and z axes respectively in the vehicle coordinate system. x, y, and z represent the coordinates of the drone in the navigation coordinate system. The water sprayer reaction force coefficients of the X, Y, and Z axes of the navigation coordinate system are taken as K x , K y , K z , m is the total mass of the cleaning drone fuselage plus the cleaning structure, and v x , v y , v z are the components of the centroid velocity of the wall cleaning drone on the X, Y, and Z axes.

[0017] Further, in step S2, the cleaning drone swarm consists of N four-rotor cleaning drones, where one drone is the leading drone that directly receives the control command signal. The control command signal includes the desired roll angle and the velocity or desired coordinate point (x d , y d , z d ) in the navigation coordinate system; the other drones are following drones that rely on the cluster cooperative control algorithm for flight movement and pose adjustment:

[0018]

[0019] where i is the number of the drone being followed, j is the neighbor drone that establishes a communication link with the i-th drone, and N i represents the set of neighbor drones of the i-th drone, n is the number of neighbor drones, and v i and v j are the velocity magnitudes of the i-th drone and its neighbor robot in the xoy plane of the navigation coordinate system, respectively. ψ i and ψ j are the roll angles of the i-th drone and its neighbor robot in the navigation coordinate system, respectively. z i and z j are the heights of the i-th drone and its neighbor robot in the navigation coordinate system, respectively. v * and ψ * are the desired velocity magnitude and the desired roll angle, respectively. v i c and ψ i c and z i c are the command signals for the following robot i. a ij is the weighting value of the communication topology graph. (x i ’, y i ’) are the horizontal and vertical coordinates in the vehicle coordinate system. x ij ’r and y ij ’r are the predetermined desired distances between the i-th drone and the j-th drone. c i and b i and k i and γ are all control coefficients. k i v and k i ψ are the control gains for adjusting the drone interval.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] The existing fixed cleaning brush head structure can only perform cleaning work on specific insulators, and has poor adaptability to insulators with different outer diameters, and cannot complete the cleaning task for insulators with a large difference in outer diameters. The insulator outer diameter adaptive cleaning method based on drone swarm control of the present invention uses drone swarms to cooperate to complete the cleaning task. According to the swarm control algorithm, it can adapt to insulators with different outer diameters and can accurately complete the cleaning task. <## BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1Schematic diagram of a quadrotor cleaning drone, where 1 is the sensor mounting platform, 2 is the drone rotor, 3 is the flight control system, 4 is the nozzle and water tank, and 5 is the cleaning brush head;

[0023] Figure 2 Schematic diagram of the coordinate system and flight principle of the cleaning drone;

[0024] Figure 3 Top view of the insulator cleaning by drone swarm control, where 1 is the ascending task point of the leader drone, 2 is the cleaning task point of the leader drone, 3 is the outer diameter of the insulator, and 4 is the follower drone;

[0025] Figure 4 Flow chart of the insulator cleaning by drone swarm control;

[0026] Figure 5 Flow chart of the design for cleaning insulators with different outer diameters adaptively. Specific implementation mode

[0027] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0028] As shown in the quadrotor cleaning drone Figure 1 First, analyze the structure and functional modules of the quadrotor drone. The quadrotor aircraft is mainly composed of a frame, a flight control board, a fixed platform, a sensor part, and a power supply part. The frame is composed of rigid brackets and satisfies axial symmetry and central symmetry. The fixed platform is located at the intersection of the two brackets and is used to place the flight control board, battery, and various sensors, etc. The sensors can mainly complete functions such as ranging between the drone and the insulator and calculating the pose information of the drone, and mainly include accelerometers, GPS positioning, laser ranging, etc. This can basically ensure that the weight of the drone is concentrated at the center of mass of the fuselage without affecting the basic flight of the drone. In addition, the four motors, electronic speed controllers, and propellers of the drone are respectively installed at the top of the two brackets, and the electronic speed controllers are installed on the arms of the frame corresponding to the motors. The cleaning drone is realized by loading cleaning devices such as nozzles, water tanks, and electric rotary brushes on the basis of the quadrotor aircraft. The water tank is placed below the fixed platform to ensure that the weight of the wall cleaning drone is concentrated at the center of mass of the fuselage. The electric rotary brush is installed at the front end of the frame, which can facilitate the control to complete the cleaning work.

[0029] In the process of analyzing the dynamics of the drone, first establish as Figure 2The shown vehicle coordinate system and navigation coordinate system, where the Y-axis, X-axis, and Z-axis of the navigation coordinate system are distributed as east-north-up. Set the point where the initial position of the cleaning drone is located as the origin. The navigation coordinate system is a reference coordinate system for describing the translational and rotational motions of an object, defined as OXYZ; the origin of the vehicle coordinate system is the center of mass of the cleaning drone, defined as oxyz. Deduce the transformation matrix R for the vehicle coordinate system to be rotated three times to the navigation coordinate system:

[0030]

[0031] where θ is the pitch angle of the drone and ψ is the yaw angle of the drone, is the roll angle. They are the angles of rotation along the Y, Z, and X axes of the navigation coordinate system during the movement of the drone respectively.

[0032] As Figure 2 shown, this paper uses a quadrotor aircraft as the fuselage of the wall-cleaning drone, and the input quantity is the 4 lift forces generated by the motor-driven propellers. The cleaning drone has a 6-degree-of-freedom motion mode during the movement, that is, translational motion along the three axes of X, Y, and Z and rotational motion around the three axes of X, Y, and Z. The cleaning drone relies on the power supply to continuously supply energy to the four motors, driving the four propellers to rotate to generate forces and torques. The changes in the forces and torques generated by the rotation of the four propellers are used to achieve the 6-degree-of-freedom motion. Therefore, the cleaning drone is an underactuated system, with more control quantities than input quantities. Changing the state of the cleaning drone is achieved by adjusting the rotational speed differences of the four motors to change the attitude angle of the fuselage.

[0033] The construction of the dynamic model of the cleaning drone can be regarded as the result of linear motion plus angular motion. Therefore, the model can be established based on Newton's equations of motion. The Newton-Euler equations are shown in Equation (2):

[0034]

[0035] where F i is the external force received by the i-th wing of the cleaning drone, m is the total mass of the cleaning drone, and v is the velocity of the center of mass of the wall-cleaning drone; M i is the torque generated by the i-th wing of the cleaning drone relative to the rotation axis, and l is the angular momentum of the cleaning drone.

[0036] In the vehicle coordinate system, the lift force F B given by the propellers received by the cleaning drone is shown in Equation (3):

[0037]

[0038] where the lift forces provided by the four rotors of the cleaning drone are represented by F1, F2, F3, and F4 respectively, and U1 is used to represent their sum, that is:

[0039] U1 = F1 + F2 + F3 + F4 (4)

[0040] Thus, the force vector of the cleaning drone in the carrier coordinate system can be obtained, and the lift force F of the cleaning robot in the navigation coordinate system is transformed E As shown in Equation (5):

[0041]

[0042] The cleaning drone is also affected by the gravity acting vertically downward along the OZ axis of the airframe. The gravity in the navigation coordinate system is expressed as: [00 - mg] T , considering the recoil force generated by water spraying, its magnitude is proportional to the water velocity. The recoil force coefficients along the X, Y, and Z axes of the navigation coordinate system are taken as K x , K y , K z , then the recoil force received by the drone can be expressed as shown in Equation (6):

[0043]

[0044] where v x , v y , v z are the components of the centroid velocity of the wall cleaning drone on the X, Y, and Z axes. The linear motion equation of the cleaning drone can be expressed as Equation (7):

[0045]

[0046] where x, y, and z represent the coordinates of the drone in the navigation coordinate system. The total moment of inertia of the cleaning drone consists of three parts, namely the moments of inertia about the X, Y, and Z axes. The moments of inertia about the X and Y axes are caused by the unequal lift forces of the four wings, while the moment of inertia about the Z axis is caused by the counter-torques generated by each wing. During flight, the rotation of each motor of the cleaning drone will generate a reaction force opposite to it. Let Q represent the counter-torque generated by the rotation of each motor of the wall cleaning drone, d represent the torque coefficient, and ω represent the angular velocity of each motor of the wall cleaning drone. Then the magnitude of the counter-torque can be expressed as Q = dω 2 . Define the distances from the motor to the X and Y axes in the carrier coordinate system as l x , l y , then the aerodynamic moment of the cleaning drone can be expressed by Equation (8)

[0047]

[0048] Define I x , I y , I zare the moments of inertia when the cleaning drone rotates around the three axes of the carrier coordinate system, and U2, U3, and U4 respectively represent the torques of the three axes x, y, and z in the carrier coordinate system. Then, the body gyroscopic effects generated by the pitching motion, rolling motion, and yawing motion during flight can be respectively expressed as Then, the angular motion equation of the cleaning drone can be obtained according to Euler's equation as shown in Equation (9):

[0049]

[0050] Combining the above formulas, the dynamic model of the cleaning drone established in step S1 is as shown in Equation (10):

[0051]

[0052] In step S2, the cleaning drone swarm consists of four quadrotor cleaning drones. One of the drones is the leader drone, which can directly receive the control command signal. The control command signal includes the desired roll angle and the speed or desired coordinate point in the navigation coordinate system. The other three drones are follower drones, which rely on the swarm cooperative control algorithm to perform flight movement and pose adjustment:

[0053]

[0054] where i is the number of the follower drone, j is the neighbor drone that establishes communication with the i-th drone, N i represents the set of neighbor drones of the i-th drone, n is the number of neighbor drones, v i , v j are respectively the speed magnitudes of the i-th drone and its neighbor robot in the xoy plane of the navigation coordinate system, ψ i , ψ j are respectively the rolling angle magnitudes of the i-th drone and its neighbor robot in the navigation coordinate system, z i , z j are respectively the heights of the i-th drone and its neighbor robot in the navigation coordinate system, v * and ψ * are the desired speed magnitude and the desired rolling angle, v i c , ψ i c , z i c are the command signals of the follower robot i, a ij is the weighted value of the communication topology graph, (x i ’, y i ’) are the horizontal and vertical coordinates in the carrier coordinate system, x ij ’r , y ij ’ris the predetermined desired distance between drones i and j, c i , b i , k i , γ are all control coefficients, k i v , k i ψ are the control gains for adjusting the drone spacing.

[0055] Controlling the following robots according to the above swarm control algorithm can enable each member to maintain the desired inter-drone distance from its neighbors and have the ability to autonomously generate formations. For the setting of the desired inter-drone distance in the insulator cleaning task, the present invention divides the cleaning task into three steps: converging for cleaning, ascending, and separating, and sets the corresponding desired inter-drone distances according to different steps. Figure 3 The top view of the drone cleaning the insulator is shown in the figure. 1 is the ascending task point of the leading drone, 2 is the cleaning task point of the leading drone, 3 is the outer diameter of the insulator, 4 is the following drone, d set is the desired distance between the drone and the insulator during the ascending process, d min is the safety distance between the drone and the center of the insulator, and Δs is the displacement traveled by the drone. The desired inter-drone distance can be expressed as Equation (12):

[0056]

[0057] So far, the entire cleaning process designed in S2 is as shown in Figure 4 the figure. After the drone arrives at the designated cleaning location, the leading drone receives sensing signals (such as ascending height, degree of cleanliness of the insulator surface, etc.) to judge the working mode. If it reaches the height of the insulator skirt, the drone enters the converging cleaning mode. The drone swarm will calculate the centroid control signal of the drones forming the cleaning mode formation according to Equations (11) and (12), and then calculate the control inputs of the four motors of the quadrotor drone according to Equation (10). Finally, it starts the cleaning task at a spacing of d min from the insulator support. After cleaning for a specified time, it starts the separation mode. The drone swarm repeats the above swarm formation process to expand the formation, and starts to hover at a spacing of d set from the insulator support. After the hovering formation is completed, the swarm starts to vertically ascend to the next insulator skirt and repeats the steps of the converging cleaning mode until the entire insulator is cleaned, that is, the drone ascends to the highest point, and the drone swarm ends the cleaning work.

[0058] In addition to the above embodiments, the present invention may also have other implementation manners. All technical solutions formed by equivalent replacement or equivalent transformation fall within the protection scope required by the present invention.

Claims

1. An insulator outer diameter adaptive cleaning method based on drone swarm control, characterized in that Step S1: Form a cleaning formation with N quadrotor cleaning drones. Based on the flight principle of the quadrotor cleaning drones, analyze their dynamic characteristics on the basis of establishing a coordinate system, and establish a dynamic model of the cleaning drones using the Newton-Euler equations. Step S2: Based on the drone dynamic model, with a cluster cooperative control algorithm, adaptively adjust the formation of the drone swarm and the distance from the insulators according to the measurement data of the sensors and the cooperative information mutually sent between the drones, so as to realize the cleaning work of insulators with different outer diameters.

2. The method for adaptively cleaning the outer diameter of an insulator based on drone swarm control according to claim 1, wherein: In Step S1, establish two basic coordinate systems, namely the navigation coordinate system and the body coordinate system, and deduce the transformation matrix for the body coordinate system to be rotated three times to transform to the navigation coordinate system: where θ is the pitch angle of the UAV and ψ is the yaw angle of the UAV, and is the roll angle; Analyze the torque and lift generated by the four rotors in the body coordinate system, comprehensively consider gravity, resistance, water flow reaction force, and the gyroscopic effect of the fuselage, and establish a dynamic model using Newton's second law: Among them, I x , I y , I z are the moments of inertia when the cleaning drone rotates around the three axes of the body coordinate system respectively. U1 represents the resultant force of the lift generated by the quadrotor. U2, U3, and U4 represent the torques on the x, y, and z axes respectively in the body coordinate system. x, y, and z represent the coordinates of the drone in the navigation coordinate system. The recoil force coefficients of the X, Y, and Z axes of the navigation coordinate system are taken as K x , K y , K z respectively. m is the total mass of the cleaning drone body plus the cleaning structure. v x , v y , v z are the components of the centroid velocity of the wall cleaning drone on the X, Y, and Z axes.

3. The method for controlling a cleaning UAV cluster with an adaptive insulator outer diameter according to claim 1, wherein: In step S2, the cleaning UAV swarm consists of N quadrotor cleaning UAVs, where one UAV is the leading UAV that directly receives the control instruction signal, and the control instruction signal includes the desired roll angle and the velocity or the desired coordinate point (x d , y d , z d ) in the navigation coordinate system; the other UAVs are following UAVs that rely on the cluster cooperative control algorithm for flight movement and pose adjustment: where i is the number of the drone being followed, j is the neighbor drone that has established a communication link with the drone numbered i, and N i represents the set of neighbor drones of the drone numbered i, n is the number of neighbor drones, and v i , v j are the magnitudes of the velocities of the drone numbered i and its neighbor robot in the xoy plane of the navigation coordinate system, respectively. ψ i , ψ j are the magnitudes of the roll angles of the drone numbered i and its neighbor robot in the navigation coordinate system, respectively. z i , z j are the heights of the drone numbered i and its neighbor robot in the navigation coordinate system, respectively. v * and ψ * are the desired velocity magnitude and the desired roll angle, respectively. v i c , ψ i c , z i c are the command signals for the following robot i. a ij is the weighted value of the communication topology graph. (x i ’ , y i ’ ) are the horizontal and vertical coordinates in the vehicle coordinate system. x ij ’r , y ij ’r are the predetermined desired distances between the drone i and j. c i , b i , k i , and γ are all control coefficients. k i v , k i ψ are the control gains for adjusting the drone interval.

Images