Research On Key Technology Of Control Algorithm Of Underwater Manipulator

Underwater manipulators are playing an increasingly important role in underwater operations,and they also play an irreplaceable role in deep-sea resource survey projects.However,there are still many problems with underwater manipulators,such as poor submarine operations and underwater operations.Factors such as low operating accuracy and poor integration of the robotic arm will affect the operating accuracy of the robotic arm.Therefore,key technologies such as error analysis,motion control and path planning of the robotic arm have high practical significance.First,establish the kinematic model of the six-degree-of-freedom underwater manipulator,analyze the forward and inverse kinematics of the underwater manipulator,and verify the correctness of the established kinematic model;then,according to the actual rotation angle of each joint of the manipulator,Use the Monte Carlo method to find the motion space of the manipulator;use the Lagrangian method to build the dynamics model of the underwater manipulator,and simulate the motion curve of each joint of the manipulator;use the Morrison formula to calculate the water of the manipulator.The dynamics is modeled,and the water resistance torque and additional mass torque of each joint are calculated.Based on the above kinematic model of the robotic arm,the four-parameter matrix method is used to establish the error model of the robotic arm,and the geometric parameters of the robotic arm are analyzed to study the degree of influence of various geometric parameters on the pose error of the end effector of the robotic arm;At the same time,the idea of controlling variables is used to analyze the influence of each joint angle on the posture error of the manipulator.Finally,the perturbation compensation method is used to compensate the model error of the manipulator,and the effectiveness of the compensation algorithm is verified by simulation.After that,considering that the robotic arm system is a nonlinear system,there is an unknown gap lag problem.First,the basic knowledge of neural networks is introduced,and then an adaptive neural network algorithm is designed,which uses a radial basis function neural network to approximate the dynamics of the robotic arm Another neural network uses the hyperbolic tangent function to approach the backlash lag problem of the system,and uses Lyapunov theory to prove its stability.The designed controller is simulated under static water environment and ocean current environment to verify the proposed The effectiveness of the algorithm.Finally,study the path planning of the robotic arm,introduce the basic knowledge of reinforcement learning,analyze the collision between the robotic arm and obstacles in the path planning process,and calculate the spatial position of the obstacle model and the robotic arm through simulation.Two algorithms of strategy iteration and reinforcement learning are used to study the path planning problem of the manipulator,and the effectiveness of the proposed obstacle avoidance algorithm is verified through simulation.