ROS Based Obstacle Avoidance Path Planning And Simulation Of Manipulator

Obstacle avoidance path planning technology for manipulator is one of the hot topics in the research of robotics technology.Scientific and reasonable obstacle avoidance path planning is a prerequisite for ensuring that the manipulator can complete a specific motion with a better trajectory in the working environment so as to ensure the normal operation.Therefore,the problem of obstacle avoidance motion planning has an important research value.Taking the obstacle avoidance motion planning algorithm as the main research starting point,the kinematics modeling and analysis of the manipulator is carried out at first.Then,some improvement strategies for path planning algorithm are proposed and a robotic motion simulation platform based on ROS is built.Finally,taking the robot model as the research object,the simulation of the algorithm was performed on the simulation platform.Firstly,this paper introduces and analyzes the basic theories of the motion planning problem and the kinematics of the manipulator.The theoretical basis of motion planning mainly includes the standardized description of the problem,as well as the common path planning methods and collision detection technology.The kinematics of the manipulator mainly introduce the robot's pose description,coordinate transformation and D-H parametric modeling,etc.Taking the UR5 robot as an example,the kinematics modeling,the inverse motion solution and the trajectory planning simulation are carried out to verify the correctness of the established model of the manipulator.Secondly,this paper briefly introduces the basic concept of ROS and its framework,and completes the modeling of the UR5 manipulator based on the URDF method.The construction of the motion simulation system of the manipulator is completed based on ROS,Moveit!(which is a motion planning package)and the Rviz(which is a 3D visualization platform),and the core function nodes of the simulation system are introduced.Finally,as the core content of this paper,the proposed improved RRT algorithm is discussed.This paper mainly introduces the target biased sampling strategy,the redundant point elimination strategy,the new metric function considering both distance and angle,and the path smoothing technology based on neural network.The proposed algorithm improves the searching efficiency and real-time performance considering the different performance requirements of path planning,and optimizes the length and curvature of the planned path.Finally,it takes the UR5 manipulator as an example and carries out the simulation of obstacle avoidance motion planning on the simulation platform based on ROS.The basic RRT algorithm and the improved one proposed in this paper are simulated under two different obstacle environments.The simulation results show that the proposed algorithm performs well in real-time performance and path optimization.It proves the effectiveness and the advantage of the proposed algorithm,and achieves the expected goal of this paper accordingly.