Underwater mechanical arm shake compensation control method for suspension operation scene

Abstract

The invention discloses an underwater mechanical arm shake compensation control method for a suspension operation scene. The method comprises the steps that a sensing positioning system is installed at the rigid connecting position of an ROV platform and an underwater mechanical arm, and a positioning mark is arranged at a suspension operation target point; the tail end of the mechanical arm is controlled to move towards a target point, an initial homogeneous transformation matrix of the mechanical arm is established according to the initial frame sensing positioning data, and then the position point of the tail end of the mechanical arm is obtained to serve as an expected value; establishing a current homogeneous transformation matrix of the shaking base according to real-time sensing positioning data, establishing a homogeneous transformation matrix of the shaking base relative to a world coordinate system in combination with an expected value of a tail end position point, and obtaining a quasi-compensation angle value in real time by using a shaking compensation algorithm; and after preprocessing, inputting into a joint controller to carry out real-time shake compensation control on each joint. According to the method, all joints of the mechanical arm are optimized and compensated through the shake compensation algorithm, the stability and controllability of tail end operation of the mechanical arm can be ensured, and therefore the operation safety is improved.

Description

Technical Field

[0001] The invention relates to a sway compensation control method for an underwater manipulator arm, and relates to the technical field of underwater operations, and in particular to a sway compensation control method for an underwater manipulator arm used in a suspended operation scenario. Background Art

[0002] Traditional underwater facility operation and maintenance operations are often performed by divers, but due to the complexity of the underwater environment, the ability of divers to carry equipment, and the time window for diving operations, some operation and maintenance tasks cannot be carried out effectively. Faced with the challenges of traditional underwater operations, underwater remotely operated vehicles (ROVs), as representatives of underwater operation robots, have become an efficient alternative technology. Underwater remotely operated vehicles (ROVs) have the advantages of being unrestricted by water depth, not requiring personnel to be directly exposed to dangerous environments, and being able to carry more professional equipment, which greatly expands the scope and efficiency of underwater operations.

[0003] Precise motion control of underwater manipulators is essential for performing complex underwater operations. However, ROVs face a series of challenges when performing suspended operations. The ROV serves as the floating base of the manipulator, and the manipulator is rigidly connected to the ROV platform. The ROV's body control has control lag and compensation accuracy deficiencies, and does not always provide the required fine operation. Summary of the invention

[0004] In order to solve the problems existing in the background technology, the present invention provides an underwater robot arm sway compensation control method for a suspended operation scenario.

[0005] The technical solution adopted by the present invention is:

[0006] The underwater manipulator sway compensation control method for a suspended operation scenario of the present invention comprises:

[0007] Step S1: Install a sensor positioning system at the rigid connection between the ROV platform and the underwater mechanical arm, and set a mark for positioning the sensor positioning system near the target point of the suspended operation.

[0008] Step S2: Control the end of the underwater manipulator to move toward the target point of the suspended operation, establish the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator according to the first frame of sensor positioning data obtained by the sensor positioning system and the marker, and then obtain the position point of the end of the underwater manipulator as the expected value.

[0009] Step S3: Establish the current homogeneous transformation matrix of the shaking base of the underwater robotic arm based on the real-time sensor positioning data obtained by the sensor positioning system and the marker, and establish the homogeneous transformation matrix of the shaking base relative to the world coordinate system based on the current homogeneous transformation matrix of the shaking base of the underwater robotic arm and the expected value of the end position point.

[0010] Step S4: According to the shaking information in the homogeneous transformation matrix of the shaking base relative to the world coordinate system, a shaking compensation algorithm is used to obtain the quasi-compensation angle value of each joint of the underwater manipulator in real time, and the quasi-compensation angle value of each joint is pre-processed and input into the joint controller of the underwater manipulator to control each joint, thereby finally realizing real-time and precise shaking compensation control of the end position of the underwater manipulator.

[0011] In the step S1, the sensor positioning system includes a calibrated monocular vision sensor and an inertial sensor IMU (Inertial Measurement Unit) suitable for underwater environments. The monocular vision sensor obtains a two-dimensional image at the target point of the suspended operation in real time, and obtains the posture and position information of the base relative to the target point of the suspended operation according to the actual size and imaging size of the two-dimensional image at the target point of the suspended operation and the mark therein, and the inertial sensor IMU obtains independent posture information of the base as a supplement to the posture data in the subsequent algorithm; the posture and position information of the base of the monocular vision sensor relative to the target point of the suspended operation and the independent posture information of the base obtained by the inertial sensor IMU together constitute the sensor positioning data. The marker specifically adopts an Aruco two-dimensional code for the visual system to assist in positioning.

[0012] In the step S2, the first frame of the base attitude information is first obtained synchronously using the monocular vision sensor and the inertial sensor IMU to ensure the timeliness and calculation accuracy of the data when the base attitude is solved, and then the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator in the world coordinate system is established. Initial homogeneous transformation matrix It reflects the position and posture of the base in the world coordinate system before the shake compensation is turned on; the current joint angles of each joint of the underwater manipulator are obtained through the joint controller of the underwater manipulator, and finally the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator in the world coordinate system is calculated. The current joint angles of each joint are calculated to obtain the position of the end of the underwater manipulator in the world coordinate system. The number of joints of the underwater manipulator is i-1. The position of the end of the underwater manipulator in the world coordinate system is as expected value.

[0013] In the step S3, according to the camera intrinsic parameter matrix K and distortion parameter D of the monocular vision sensor and the actual size and imaging size of the two-dimensional image at the target point of the suspended operation and the mark therein, the mapping relationship between the two-dimensional image obtained by the monocular vision sensor and the actual three-dimensional point in space is obtained in real time, and then the external parameter information of the shaking base is obtained, including the rotation vector rvec and the translation vector tvec, rvec = (r x ,r y ,r z ), tvec=(tx,ty,tz), rx, r y and rz are the X-axis, Y-axis and Z-axis rotations of the shaking base in the world coordinate system, respectively. x ,t y and t z are the X-axis, Y-axis, and Z-axis translations of the shaking base in the world coordinate system, and the rotation vector rvec in the external parameter information of the shaking base is transformed using Rodrigues transformation to obtain the rotation matrix R rodrigues , according to the rotation matrix R rodrigues The posture information of the shaking base is obtained; the posture information of the shaking base obtained in real time by the monocular vision sensor and the inertial sensor IMU is filtered and fused by a particle filter algorithm to obtain the first posture information of the shaking base after fusion, and the accumulated error of the inertial sensor IMU and the measurement error of the monocular vision sensor are corrected to correct the error and improve the accuracy; then the particle filter algorithm is used to filter and fuse the first posture information of the shaking base after fusion and the translation vector tvec to obtain the second posture information of the shaking base after fusion, and the current homogeneous transformation matrix of the shaking base is established according to the second posture information of the shaking base after fusion Finally, the base's posture is converted from the original axis-angle information and translation information into the form of rotation matrix, Euler angle and homogeneous transformation matrix. Finally, according to the current homogeneous transformation matrix of the shaking base Combined with the expected value of the end position point, that is, the homogeneous transformation matrix of the base at the initial moment Create a homogeneous transformation matrix of the shaking base relative to the world coordinate system Perform pose calculation for relative positioning; homogeneous transformation matrix of the base at the initial moment Contains the position and attitude information of the base relative to the world coordinate system at the initial moment.

[0014] In step S4, according to the homogeneous transformation matrix of the shaking base relative to the world coordinate system The shaking information in the homogeneous transformation matrix of the shaking base relative to the world coordinate system is obtained, and then the quasi-compensation angle value of each joint of the underwater manipulator is obtained in real time using the shaking compensation algorithm, as follows:

[0015] Firstly, the DH (Denavit-Hartenberg) parameter table is established based on the configuration of the underwater manipulator, and the homogeneous transformation matrix of the base and each joint of the underwater manipulator relative to the world coordinate system is established. Then, the homogeneous transformation matrix of the base and each adjacent joint of the underwater manipulator relative to the world coordinate system is multiplied in sequence to establish the homogeneous transformation matrix of the end of the underwater manipulator on the base relative to the world coordinate system. Including the rotation and translation information of the base relative to the world coordinate system and the transformation relationship between the adjacent joints of the underwater manipulator; then the homogeneous transformation matrix of the shaking base relative to the world coordinate system Inverse and multiply the homogeneous transformation matrix of the end of the underwater manipulator on the base relative to the world coordinate system Obtain the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system Homogeneous transformation matrix of underwater manipulator relative to world coordinate system The first three elements of the fourth column are the positions to be reached by the underwater manipulator in the coordinate system of the shaking base; when the homogeneous transformation matrix of the shaking base relative to the world coordinate system When the posture position changes, the end of the underwater manipulator is always stable at the expected value of the position point of the end of the underwater manipulator in the world coordinate system. According to the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system And the DH parameter table obtains the optimal joint angle value of each joint as the quasi-compensation angle value q comp , the proposed compensation angle value q comp After Kalman filtering, the data is input into the joint controller of the underwater manipulator. Kalman filtering is performed to ensure the continuity and smoothness of the angle change, and then the joints of the underwater manipulator are controlled to move to the quasi-compensation angle value q after filtering. comp .

[0016] The electronic device of the present invention comprises: a memory and a processor coupled to each other, wherein the memory stores program data, and the processor calls the program data to execute the method as described above.

[0017] The computer-readable storage medium of the present invention stores program data thereon, and when the program data is executed by a processor, the method described above is implemented.

[0018] The beneficial effects of the present invention are:

[0019] The method of the present invention uses a monocular vision sensor and an inertial sensor IMU installed at the connection between the ROV body and the base, and an Aruco two-dimensional code set in the target operation area to measure and solve the base shaking information during the operation of the ROV-robotic arm suspension platform. Finally, the desired input value of each joint of the robotic arm is optimized and compensated through a shaking compensation control algorithm to ensure the stability and controllability of the operating gripper at the end of the robotic arm, thereby improving the safety of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the flow of a method for compensating for arm shaking according to an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of base trajectory positioning data for a shake compensation algorithm experiment in an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of base trajectory error data in a shake compensation algorithm experiment in an embodiment of the present invention;

[0023] Figure 4 Schematic diagram of the end error data of the robot arm working gripper after algorithm compensation and the end data without compensation in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] like Figure 1 As shown, the present invention takes an underwater three-degree-of-freedom electric manipulator as an example to implement a sway compensation control method for an underwater manipulator in a suspension operation scenario, which is specifically as follows:

[0026] Step S1: Install a sensor positioning system at the rigid connection between the ROV platform and the underwater manipulator, and set a mark for positioning the sensor positioning system near the target point of the suspended operation. The sensor positioning system includes a calibrated monocular vision sensor and an inertial sensor IMU suitable for underwater environment. The monocular vision sensor obtains a two-dimensional image at the target point of the suspended operation in real time, and obtains the posture and position information of the base relative to the target point of the suspended operation according to the actual size and imaging size of the two-dimensional image at the target point of the suspended operation and the mark therein, and the inertial sensor IMU obtains the independent posture information of the base as a supplement to the posture data in the subsequent algorithm; the posture and position information of the base of the monocular vision sensor relative to the target point of the suspended operation and the independent posture information of the base obtained by the inertial sensor IMU together constitute the sensor positioning data. The mark specifically adopts Aruco QR code for auxiliary positioning of the visual system.

[0027] The monocular vision sensor has been pre-calibrated, and reliable camera intrinsic parameter matrix K and distortion parameter D have been obtained. In order to solve the scale uncertainty problem of the monocular vision sensor, an Aruco QR code of known size is posted in the target area of ​​the suspended operation. By comparing the actual size of the Aruco QR code with the imaging size, the scale information and distance information of the monocular vision sensor relative to the suspended operation area are indirectly obtained.

[0028] Step S2: Control the end of the underwater manipulator to move toward the target point of the suspended operation, establish the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator according to the first frame of sensor positioning data obtained by the sensor positioning system and the marker, and then obtain the position point of the end of the underwater manipulator as the expected value, as follows:

[0029] First, the first frame of the base's attitude information is obtained synchronously using the monocular vision sensor and the inertial sensor IMU to ensure the timeliness and calculation accuracy of the data when solving the base's attitude, and then establish the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator in the world coordinate system. Initial homogeneous transformation matrix It reflects the position and posture of the base in the world coordinate system before the shake compensation is turned on; the current joint angles of each joint of the underwater manipulator are obtained through the joint controller of the underwater manipulator, and finally the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator in the world coordinate system is calculated. The current joint angles of each joint are calculated to obtain the position of the end of the underwater manipulator in the world coordinate system. The end of the underwater manipulator is located at the point in the world coordinate system as expected value.

[0030] Step S3: Establish the current homogeneous transformation matrix of the shaking base of the underwater manipulator according to the real-time sensor positioning data obtained by the sensor positioning system and the marker, and establish the homogeneous transformation matrix of the shaking base relative to the world coordinate system according to the current homogeneous transformation matrix of the shaking base of the underwater manipulator and the expected value of the end position point, as follows:

[0031] Firstly, according to the camera intrinsic parameter matrix K and distortion parameter D of the monocular vision sensor and the actual size and imaging size of the two-dimensional image at the target point of the suspended operation and the mark therein, the mapping relationship between the two-dimensional image obtained by the monocular vision sensor and the actual three-dimensional point in space is obtained in real time, and then the external parameter information of the shaking base is obtained, including the rotation vector rvec and the translation vector tvec, rvec = (r x ,r y ,r z ), tvec=(t x ,t y ,t z ), rx 、r y and r z are the X-axis, Y-axis, and Z-axis rotations of the shaking base in the world coordinate system, t x ,t y and t z are the X-axis, Y-axis, and Z-axis translations of the shaking base in the world coordinate system, and the rotation vector rvec in the external parameter information of the shaking base is transformed using Rodrigues transformation to obtain the rotation matrix R rodrigues , according to the rotation matrix R rodrigues The posture information of the shaking base is obtained; the posture information of the shaking base obtained in real time by the monocular vision sensor and the inertial sensor IMU is filtered and fused by a particle filter algorithm to obtain the first posture information of the shaking base after fusion, and the accumulated error of the inertial sensor IMU and the measurement error of the monocular vision sensor are corrected to correct the error and improve the accuracy; then the particle filter algorithm is used to filter and fuse the first posture information of the shaking base after fusion and the translation vector tvec to obtain the second posture information of the shaking base after fusion, and the current homogeneous transformation matrix of the shaking base is established according to the second posture information of the shaking base after fusion Finally, the base's posture is converted from the original axis-angle information and translation information into the form of rotation matrix, Euler angle and homogeneous transformation matrix. Finally, according to the current homogeneous transformation matrix of the shaking base Combined with the expected value of the end position point, that is, the homogeneous transformation matrix of the base at the initial moment Create a homogeneous transformation matrix of the shaking base relative to the world coordinate system Perform pose calculation for relative positioning; homogeneous transformation matrix of the base at the initial moment Contains the position and attitude information of the base relative to the world coordinate system at the initial moment.

[0032] Step S4: According to the shaking information in the homogeneous transformation matrix of the shaking base relative to the world coordinate system, a shaking compensation algorithm is used to obtain the pseudo-compensation angle value of each joint of the underwater manipulator in real time, and the pseudo-compensation angle value of each joint is pre-processed and input into the joint controller of the underwater manipulator to control each joint, and finally the real-time and accurate shaking compensation control of the end position of the underwater manipulator is realized, as follows:

[0033] First, a DH parameter table is established based on the configuration of the underwater manipulator, and the homogeneous transformation matrix of the base and each joint of the underwater manipulator relative to the world coordinate system is established. Then, the homogeneous transformation matrix of the base and each adjacent joint of the underwater manipulator relative to the world coordinate system is multiplied in sequence to establish the homogeneous transformation matrix of the end of the underwater manipulator on the base relative to the world coordinate system. Including the rotation and translation information of the base relative to the world coordinate system and the transformation relationship between the adjacent joints of the underwater manipulator, as follows:

[0034] The underwater three-degree-of-freedom electric manipulator carried by the ROV platform is modeled using the DH representation method, and the corresponding posture modeling is performed for the shaking base. The homogeneous transformation matrix It is expressed as follows:

[0035]

[0036] in, Indicates the rotation information of the base shaking, n x 、n y and n z Respectively represent the projection of the x-axis on the original x, y, and z-axis after rotation, o x , o y and z Respectively represent the projection of the y-axis on the original x, y, and z-axis after rotation, a x 、a y and a z Respectively represent the projection of the z-axis on the original x, y, and z-axes after rotation; Indicates the translation information of the base shaking, p x 、p y and p z Respectively represent the X-axis, Y-axis, and Z-axis translation of the base shaking.

[0037] The shoulder joint, upper arm joint and lower arm joint of the underwater manipulator are respectively and Indicates that the transformation matrix of each joint is given by:

[0038]

[0039]

[0040] Among them, T joint Represents the transformation matrix of each joint, R joint Represents the rotation matrix generated by the rotation of each joint, t joint represents the translation vector under the rotation of each joint; q1, q2 and q3 represent the joint angle values ​​of the three joints respectively, d1, d2 and d3 represent the first, second and third DH parameters respectively, d i Represents Z i X axis direction i-1 Axis and X i The distance between the axes; a i Represents X i Axis direction Z i-1 Axis and Z iThe distance between the axes.

[0041] The transformation matrix of the end gripper of the robot arm on the shaking base relative to the world coordinate system can be obtained by multiplying the transformation matrices of adjacent joints, as follows:

[0042]

[0043] in, Represents the transformation matrix of the end gripper of the robot arm on the shaking base relative to the world coordinate system. The first three elements of the fourth column Used to indicate the position of the end of the robotic arm under the shaking base relative to the world coordinate system.

[0044] For a multi-link robot, the Jacobian matrix method is used to solve the inverse kinematics of the robot according to each DH parameter, and the optimal solution is selected from multiple solutions according to the current joint angle values ​​of each joint as the output of the inverse kinematics solution. The optimal solution is the joint angle value with the smallest change from the previous moment. Through the above forward and inverse kinematics modeling and solution of the robot, the mathematical model and connection between the end gripper posture of the robot and the base posture of the robot and the joint angles of the robot are established.

[0045] The base shake information is saved in the homogeneous transformation matrix of the shaken base relative to the world coordinate system In the code, the homogeneous transformation matrix of the shaking base relative to the world coordinate system is Inverse And multiply the homogeneous transformation matrix of the end of the underwater manipulator on the base relative to the world coordinate system by Obtain the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system Homogeneous transformation matrix of underwater manipulator relative to world coordinate system The first three elements of the fourth column in are the positions to be reached by the end of the underwater manipulator in the coordinate system of the shaking base; the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system is obtained The details are as follows:

[0046]

[0047] Take the first three elements of the fourth column It is used to indicate the position that the end gripper of the robot arm needs to reach in the shaking base coordinate system. At this time, in the world coordinate system, there is a transformation matrix when the base shakes. When the posture position changes, the end of the robotic arm gripper will stabilize at the expected value of the position point of the end of the underwater robotic arm in the world coordinate system The above-mentioned geometric method robot inverse kinematics solver is used to solve the problem.

[0048] When the shaking base is relative to the world coordinate system, the homogeneous transformation matrix When the posture position changes, the end of the underwater manipulator is always stable at the expected value of the position point of the end of the underwater manipulator in the world coordinate system. According to the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system And the DH parameter table obtains the optimal joint angle value of each joint as the quasi-compensation angle value q comp , the joint compensation angle value q is obtained comp Kalman filtering is performed to ensure the continuity and smoothness of the angle change. The filtered joint compensation angle value q comp The input is given to the joint controller of the robotic arm, which then controls the joints of the underwater robotic arm to move to the pseudo-compensation angle value q after filtering. comp , to achieve real-time and accurate compensation of the end gripper position.

[0049] Finally, in order to verify the control effect of the sway compensation control method proposed in the present invention, a physical experiment was carried out on the robot arm and the sway platform:

[0050] First, the robotic arm base equipped with a monocular vision sensor and an inertial sensor IMU is installed on the shaking platform. During the free movement of the platform, the data solved by the base shaking compensation is measured and recorded. The attitude shaking amplitude of the base in the bow degree of freedom is kept greater than 0.5 radians, and the displacement shaking amplitude in the front and back, up and down, left and right degrees of freedom is greater than 300 mm. The actual shaking data of the base is collected by a motion capture device and compared with the data collected by the sensor. Figure 2 what Figure 3 As shown, the base shaking posture acquisition positioning data and error data of the shaking compensation algorithm experiment can be obtained, and the average error between the collected shaking trajectory and the real trajectory is calculated to be 19.1 mm, and the maximum error is 35.0 mm, which proves that the relative positioning accuracy of the algorithm of the present invention for the base shaking meets the experimental requirements.

[0051] At the same time, the robot arm and the base are installed on the shaking platform, and the position data of the end of the robot arm gripper are collected to ensure that the shaking amplitude of the posture in each direction is greater than 0.5 radians, and the shaking amplitude of the displacement in each direction is greater than 300 mm. In the case of horizontal shaking, front and back shaking, up and down shaking, bow shaking and mixed shaking, the above algorithm is applied to compensate the position of the end of the robot arm gripper, and the motion capture device is used to collect the compensated working gripper end data, and the final trajectory effect and error are displayed, such as Figure 4 As shown, the average operating error achieved at the end is 19.1 mm, and the maximum operating error is 52.5 mm, which proves that the method of the present invention can ensure the stability and controllability of the operating gripper at the end of the robot arm, thereby improving the safety of the operation.

[0052] The above is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for compensating and controlling the sway of an underwater manipulator for a suspension operation scenario, characterized in that: include: Step S1: installing a sensor positioning system at the rigid connection between the ROV platform and the underwater manipulator, and setting a mark for positioning the sensor positioning system at the target point of the suspended operation; Step S2: Control the end of the underwater manipulator to move toward the target point of the suspended operation, establish the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator according to the first frame of sensor positioning data acquired by the sensor positioning system and the marker, and then obtain the position point of the end of the underwater manipulator as the expected value; Step S3: establishing a current homogeneous transformation matrix of the shaking base of the underwater manipulator according to the real-time sensor positioning data obtained by the sensor positioning system and the marker, and establishing a homogeneous transformation matrix of the shaking base relative to the world coordinate system according to the current homogeneous transformation matrix of the shaking base of the underwater manipulator and the expected value of the end position point; Step S4: According to the shaking information in the homogeneous transformation matrix of the shaking base relative to the world coordinate system, a shaking compensation algorithm is used to obtain the quasi-compensation angle value of each joint of the underwater manipulator in real time, and the quasi-compensation angle value of each joint is pre-processed and input into the joint controller of the underwater manipulator to control each joint, thereby finally realizing real-time shaking compensation control of the end position of the underwater manipulator.

2. The underwater manipulator sway compensation control method for a suspended operation scene according to claim 1 is characterized in that: In the step S1, the sensing and positioning system includes a calibrated monocular vision sensor and an inertial sensor IMU. The monocular vision sensor acquires a two-dimensional image at a target point of the suspended operation in real time, acquires the posture and position information of the base relative to the target point of the suspended operation according to the actual size and imaging size of the two-dimensional image at the target point of the suspended operation and the mark therein, and acquires independent posture information of the base according to the inertial sensor IMU; the posture and position information of the base relative to the target point of the suspended operation obtained by the monocular vision sensor and the independent posture information of the base obtained by the inertial sensor IMU together constitute the sensing and positioning data.

3. The underwater manipulator sway compensation control method for a suspension operation scenario according to claim 2 is characterized in that: In the step S2, the first frame of the base posture information is first obtained synchronously using the monocular vision sensor and the inertial sensor IMU, and then the initial homogeneous transformation matrix of the base and each joint of the underwater manipulator in the world coordinate system is established. The current joint angles of each joint of the underwater manipulator are obtained through the joint controller of the underwater manipulator, and finally the initial homogeneous transformation matrix of the base of the underwater manipulator and each joint in the world coordinate system is calculated. The current joint angles of each joint are calculated to obtain the position of the end of the underwater manipulator in the world coordinate system. The number of joints of the underwater manipulator is i-1. The position of the end of the underwater manipulator in the world coordinate system is as expected value.

4. The underwater manipulator sway compensation control method for a suspension operation scenario according to claim 2 is characterized in that: In the step S3, according to the camera intrinsic parameter matrix K and distortion parameter D of the monocular vision sensor and the actual size and imaging size of the two-dimensional image at the target point of the suspended operation and the mark therein, the mapping relationship between the two-dimensional image obtained by the monocular vision sensor and the actual three-dimensional point in space is obtained in real time, and then the external parameter information of the shaking base is obtained, including the rotation vector rvec and the translation vector tvec, rvec = (rx, ry, rz), tvec = (tx, ty, tz), rx, r y and rz are the rotation amounts of the shaking base in the world coordinate system, respectively, x ,t y and t z are the X-axis, Y-axis and Z-axis translations of the shaking base in the world coordinate system respectively. The rotation vector rvec in the external parameter information of the shaking base is transformed by Rodriguez transformation to obtain the rotation matrix R rodrigues , according to the rotation matrix R rodrigues Obtaining the posture information of the shaking base; using the particle filter algorithm to filter and fuse the posture information of the shaking base obtained in real time by the monocular vision sensor and the inertial sensor IMU, to obtain the first posture information of the shaking base after fusion, and then using the particle filter algorithm to filter and fuse according to the first posture information of the shaking base after fusion and the translation vector tvec, to obtain the second posture information of the shaking base after fusion, and establishing the current homogeneous transformation matrix of the shaking base according to the second posture information of the shaking base after fusion Finally, according to the current homogeneous transformation matrix of the shaking base Combined with the expected value of the end position point, establish the homogeneous transformation matrix of the shaking base relative to the world coordinate system 5. The underwater manipulator sway compensation control method for suspension operation scenarios according to claim 2 is characterized in that: In step S4, according to the homogeneous transformation matrix of the shaking base relative to the world coordinate system The shaking information in the homogeneous transformation matrix of the shaking base relative to the world coordinate system is obtained, and then the quasi-compensation angle value of each joint of the underwater manipulator is obtained in real time using the shaking compensation algorithm, as follows: First, a DH parameter table is established based on the configuration of the underwater manipulator, and the homogeneous transformation matrix of the base and each joint of the underwater manipulator relative to the world coordinate system is established. Then, the homogeneous transformation matrix of the base and each adjacent joint of the underwater manipulator relative to the world coordinate system is multiplied in sequence to establish the homogeneous transformation matrix of the end of the underwater manipulator on the base relative to the world coordinate system. Then the homogeneous transformation matrix of the shaking base relative to the world coordinate system Inverse and multiply the homogeneous transformation matrix of the end of the underwater manipulator on the base relative to the world coordinate system Obtain the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system When the shaking base is relative to the world coordinate system, the homogeneous transformation matrix When the posture position changes, the end of the underwater manipulator is always stable at the expected value of the position point of the end of the underwater manipulator in the world coordinate system. According to the homogeneous transformation matrix of the underwater manipulator relative to the world coordinate system And the DH parameter table obtains the optimal joint angle value of each joint as the quasi-compensation angle value q comp , the proposed compensation angle value q comp After Kalman filtering, the Kalman filter is input into the joint controller of the underwater manipulator, and then the joints of the underwater manipulator are controlled to move to the pseudo-compensation angle value q after filtering. comp .

6. An electronic device, characterized in that: include: A memory and a processor coupled to each other, wherein the memory stores program data, and the processor calls the program data to execute the method according to any one of claims 1 to 5.

7. A computer-readable storage medium having program data stored thereon, characterized in that: When the program data is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images