Research On Path Planning Method Of Manipulator Based On Improved RRT Algorithm

The mechanical arm has a wide range of applications in the fields of industrial manufacturing,medical services,and agricultural production,which can greatly improve production efficiency and reduce errors caused by human factors.At present,with the continuous development of the electronic and computer fields,people have put forward higher requirements on the robotic arm,especially in the autonomous planning of obstacle avoidance movement paths.Due to the complex and changeable working environment of the manipulator and its own complex structure,it has become a major difficulty in the application of path planning algorithms in the manipulator.RRT(Rapidly-exploring Random Tree)algorithm has a wide range of applications in the field of path planning.The algorithm is probabilistically complete and suitable for high-dimensional space.This feature can be used to study the application of RRT algorithm in robot arm path planning.This article takes six-degree-of-freedom industrial manipulator as the research object,mainly carries out the research on the improvement of RRT algorithm and its application in the path planning of the manipulator.The main contents are as follows:1.Kinematics analysis of the robotic arm.Introduce the coordinate transformation method in Cartesian space,introduce the DH method to establish a six-degree-of-freedom kinematics model of the robot arm,obtain the homogeneous transformation matrix of the six links,and solve the forward and inverse kinematics for it Planning research provides a theoretical basis.2.Study the improved method of mechanical arm collision detection technology.First,the characteristics of different bounding box algorithms are discussed,and the scheme suitable for this paper is selected.Then for the problem of low detection accuracy in traditional collision detection,the idea of rotating rays is used to improve the method of bounding box separation.Finally,a simulation experiment is carried out in the Matlab toolbox Robitics Toolbox.The results show that this method can improve the detection accuracy of collision detection.Provide security guarantee for subsequent path planning research.3.Research will use reinforcement learning to optimize the RRT algorithm for path planning strategies.The expansion method and path formation process of the RRT algorithm tree are studied,and the cause of the local minimum is analyzed.Based on the RRT algorithm,an improved algorithm Q-RRT is used,which is to integrate reinforcement learning on the original RRT algorithm.First,by constructing the Markov model of the robotic arm,the enhanced signal generated by the interaction between the search tree and the environment is used to continuously improve the expansion of the tree,and the node selection strategy is optimized to ensure the path quality.Then the B-spline function is used to optimize the resulting path,so that the robotic arm can complete the specified task stably and efficiently under the premise of satisfying its own kinematic constraints.Finally,algorithm experiments and simulation experiments are carried out separately.The results show that the path planning method not only shortens the movement time of the manipulator,but also takes into account the stability of operation and meets the movement requirements.The improved RRT algorithm has important application prospects for the research of robotic arm path planning with obstacle avoidance requirements.