Research And Development Of Trajectory Planning And Optimization Of Robots

Robot is regarded as one of the advanced manufacturing equipment, and has beenextensively used in every area and plays an important part in the sustained and rapiddevelopment of modern industry. An industrial robot is the highlight in the robot family,because it can not only improve labor efficiency effectively, make the entire productionprocess automation and enhance the quality of products, but also can work in hostileenvironment that human beings cannot do. Research for industrial robot trajectory planningand optimization is always hotspot in the robot field.This paper firstly summarizes background, significance, and the research and applicationstatus both from domestic and overseas, and analyzes the development trend of industrialrobots.6DOF serial robot kinematics which is widely used is analyzed. Using the6DOF seriesrobot in the laboratory as an object, and adopting D-H method to build link coordinate systemand get the homogeneous transformation matrix of adjacent rod, then forward kinematicalequation can be deduced. Closed-form solution is used to solve inverse kinematics solution,because DENSO VP6242G is fulfilled with Pieper standard in the struction. And this lays thetheoretical foundation of trajectory planning research.Secondly, task space and joint space smooth track construction method are established.Equations that describe linear and circular track of the task space are derivated, and cubicspline is adopted to construct smooth trajectory in task space. Based on the joint movement onrequirements of smoothness, B-spline is used to generate joint trajectory, so joint velocity,acceleration and jerk are continuous, and the velocity, acceleration and jerk at the starting andending points can be configured. In order to improve the production efficiency of the robot,the time optimal trajectory planning problems is researched. By transforming the kinematicconstraints to constraints on control points of B-spline, the optimal time node vector isobtained with trust region algorithm, and then time optimal trajectory is generated. The taskspace and joint space trajectory planning and optimization algorithm laid a foundation for robot’s fast andsmooth movement.Finally, do experiment with a two DOF parallel robot. Using machine vision&PC&NIRIO7842board to set up the control system, to implement the independent teach mode ofrobot path planning by adding vision sensor. First analysis kinematics of the two DOF parallelrobot, then aim at the characteristics of real-time visual guide and joint space trajectory ofsmoothness requirements, a joint space dynamic B spline trajectory planning method isproposed, so that joint space B spline trajectory with constant change over visual guide can beachieved. The proposed method fit the characteristics of real-time guidance, and ensures thesmoothness of the joint path. Experimental results verify the validity.