Research On Trajectory Planning Of Six-degree-of-freedom Manipulator Based On Neural Network

With the development of intelligent technology,industrial robot have become more and more widely used in modern manufacturing and have become one of the indispensable manufacturing equipment.However,in the actual operation of the robot arm will be restricted by some objective conditions,such as avoiding obstacles and running efficiently and smoothly.Therefore,it is necessary to plan the motion trajectory of the robot arm.Manipulator trajectory planning is a key link in the research of robot control system,and the result of trajectory planning will directly affect the quality of manipulator motion.This paper takes the XB7 h six-axis manipulator as the research object,and proposes a six-axis manipulator trajectory planning method based on RBF neural network by combining theoretical research and experimental verification.The simulation verification is carried out by MATLAB simulation software,and the experimental verification is carried out on the XB7 h six-axis manipulator platform.The results show that the efficiency of trajectory planning is improved under the premise of ensuring the accuracy of the manipulator's motion trajectory.The main research content and methods of this article include the following aspects:(1)Taking the XB7 h six-axis manipulator arm as the prototype,establish a six-axis manipulator linkage coordinate system by studying the pose description and homogeneous transformation of manipulator;use the D-H parameter method to establish the mathematical model of the manipulator.The forward kinematics equation of the manipulator is derived through homogeneous transformation,and the inverse solution of the kinematics of the manipulator is obtained by algebraic method.The robot arm model was established on the MATLAB simulation platform and the forward and inverse kinematics simulation verification was carried out.The coordinate of the robot arm was converted in Cartesian space and joint space.(2)Discuss how RBF neural network can be implemented in six-axis trajectory planning.This paper uses the newrb()function in MATALB software to design and train RBF neural network.The fifth-order polynomial interpolation algorithm is selected,and the RBF neural network adaptive control algorithm is written based on MATLAB software,and the angular displacement,angular velocity and angular velocity angular acceleration of each joint are simulated and analyzed.(3)Experiments verify whether this trajectory planning is feasible.The programmed six-axis manipulator trajectory interpolation program and control program are embedded into the XB7 h six-axis manipulator control system to verify the correctness of the trajectory planning method proposed in the previous article.