Analysis And Design Of A Quotient Kinematics Machine

With the rapid development of modern industrial manufacturing, therequirement of industrial equipment is also growing. From the perspective ofstructure and mechanism design, it is impossible to meet the increasingperformance of kinematics and dynamics with single multiple degrees of freedom(DoF) mechanism. Thus, a quotient kinematics machine (QKMs), which consistof two modules comes forth. It combines the advantages of serial and parallelkinematics machine, realizing one motion task with the cooperation of multiplemechanisms with lower DoF. Its lower DoF mechanism modules leading to betterrigidity, lower cost, and higher reliability of the system. In the current study,there already have systematic investigation of their mechanism synthesis, eachmechanism module can be either serial kinematics machine, parallel kinematicsmachine, or even hybrid kinematics machine. Position error is a very importantperformance index. Based on the analysis of the sources of position error, thejoint clearance has more significant impact on the position error than the othererror sources, such as assembly and manufacturing errors and motor actuationerrors, etc. Contrary to the assembly and manufacturing errors, joint clearanceleads to uncertain error motions at an arbitrary pose of the mechanism. Its effectsare highly non-repeatable and cannot be rectified with any kind of calibration.Therefore, this paper reports a comprehensive research effort on the positionerror that caused by the joint clearance.Firstly, based on the definition of configuration error and pose error, thisthesis established the pose error model of the clearance-affected revolute joint,prismatic joint and spherical joint, respectively. Then, based on the internal errorand motion type error, this thesis established the relationship between the overallinternal error and motion type error of the QKMs and its modules. After that, thisthesis take a QKM, which consist of a five-bar mechanism and a prismatic jointto realize the planar motion type PL (z), as an example to show how to establishthe QKMs' pose error model. And by the CVX toolbox of Matlab, obtained themaximum pose error in any position included in the workspace of QKM. And on this basis, this thesis analysis the internal and motion type error's distribution ineach direction in QKM's objective workspace. Then proposed an integrated indexthat considers both the internal and motion type error. And set this index asobjective function of QKM's dimensional optimization problem, by using thegenetic algorithm, the optimal bar length of the five-bar mechanism has beenfound. By analyzing the optimal result, it shows that with the optimization modelproposed in this thesis, the internal and motion type error can be reducedeffectively, and due to the choice of index's weight, the effect of suppressingmotion type error is more obvious.