Position Algorithm Study On Steel Joint Assembly System

As the current steel structures become more beautiful and apply to more occasions,the steel joints are also more complex.While the traditional assembly with long processing cycle and big assembly errors,is difficult to follow structure development.Then we present a 3P3 R system to assemble complex steel joints,where the position algorithm is the core technology.This paper studies the assemble position algorithm,the work is as follows:The kinds of joints are discussed under the 3R3 P system,and the most extensive rectangular-bracket joint is to be the object.In the joint model,the initial datum and reference coordinate system is established,and each mathematical model of bracket outlet is respectively set up,then the model is applied to the special joints from circular to triangular.Based on the mathematical model of complex joints,the position algorithm solution including forward and inverse kinematic for 3P3 R assembly system is studied.Due to D-H method’s hard to understand and uneasy to convey,the relative coordinate system method is given out to get forward kinematic relations,which obtains kinematical relations by setting up a local coordinate system in global coordinate system.The equations are found at the kinematic relations,form which the two inverse kinematics solutions are obtained by considering mutual relationship between equations.Through a case study and MATLAB toolbox simulation,the forward and inverse kinematics is analysis and an optimal solution is determined according to the distance.In the end,considering the error of the assembly system in the actual engineering,the Jacobi iterative algorithm is proposed to reduce the errors.To test the algorithm,the experimental platform is developed on the basis of original machine,which includes hardware and software.The hardware uses delta EH3 PLC,B2 controller and 6 servo motors;the software function origin back to zero and positioning through PLC and text display programming.At last,joint model assemble experiment is carried out on the platform.The results show that the algorithm has a high accuracy with model errors between 1.2mm to 3mm and has a good reliability.The errors are analysis and solutions to reduce errors are provided,and the algorithm set a basic foundation for developing an auto assembly machine.