Research On Obstacle Avoidance Path Planning Of Six-degree-of-freedom Manipulator

With the proposal of the 14 th Five Year Plan,the level of robot technology in China has developed rapidly.In today’s society,robots are playing an increasingly important role in fields such as industrial production,medical services,warehousing and logistics,and national defense readiness.As an indispensable part of industrial robots,the robotic arm mainly completes a single assembly line operation in traditional processes based on the path set by manual teaching.This method not only requires certain technical requirements for operators,but also has low efficiency and poor adaptability.The increasing demand for intelligent development of robotic arms urgently requires them to independently plan task paths in different production environments.Therefore,designing an efficient obstacle avoidance path planning scheme for robotic arms to ensure their accuracy and safety is crucial.This thesis takes ROB3V6 six degrees of freedom manipulator as the research object,and studies the path planning of the manipulator in different workspace.The main research work of the thesis is as follows:(1)Study the kinematics modeling of the manipulator and design collision detection algorithm.Based on the modified D-H method,the parameter model of the ROB3V6 manipulator is established,the forward kinematics equation and the inverse kinematics equation of the manipulator are derived in detail,and MATLAB simulation software is used to verify the correctness of the forward and inverse kinematics equations of the manipulator.At the same time,the collision detection algorithm of the manipulator in the working environment is designed,and the collision situations that the six degrees of freedom manipulator may encounter in the working process are emphatically analyzed,And the space obstacle is enveloped by a sphere,and the robotic arm body is enveloped by a cylinder.The collision detection between the robotic arm body and the obstacle space,as well as the collision detection between the robotic arm body models,are completed through a combination of geometric and analytical methods.(2)Design a path planning algorithm for the robotic arm.In order to improve the quality and timeliness of path planning for robotic arms,the most widely used Rapidly-exploring Random Trees and their improved algorithms in three-dimensional space were studied,and new improved forms were proposed.To solve the problem of low search efficiency of RRT algorithm,a preset tree strategy is proposed,which distributes the exploration nodes in the whole space in advance to reduce the search time.For the problem of slow convergence,a bidirectional search structure is introduced and greedy algorithm,target bias strategy and variable step size strategy are used to speed up the acquisition of initial paths.For the problem of unsmooth paths planned by the algorithm,A circular pruning strategy was proposed and Nurbs curves were used for smoothing.Finally,use MATLAB simulation software to verify that the improved algorithm has more advantages in algorithm timeliness,path quality,and path smoothness compared to the Bias-RRT,RRT-star and RRT-connect algorithm.(3)Design path planning schemes for robotic arms in different workspaces.A detailed study was conducted on the path planning of the robotic arm in joint space and Cartesian space,and the differences in path planning between the two workspaces were compared and analyzed.A specific implementation plan and algorithm flowchart were designed,and the unique singularity problem of the robotic arm in Cartesian space was solved in the form of path planning.At the same time,the joint rotation problem that may occur during the algorithm’s pose planning in Cartesian space was avoided.Finally,based on the parameters of the robotic arm generated by Solidworks simulation software,the path planning algorithm proposed in this paper was implemented in the robot operating system ROS in Cartesian space,verifying the safety and effectiveness of the planned path.