Sintesis Dimensi Manipulator Paralel Bidang Dua Derajat Kebebasan Dengan Rantai Kinematik Paralelogram Simetris

Abstract

The paper discussed the process to find the optimum dimension for the kinematic constants of a two-degree of freedom planar parallel manipulator. This manipulator itself was constructed by symmetric three parallelogram chains. An optimization process using non-sorted dominated genetic algorithm II (NSGA-II) was carried out for maximization of ( i ) r MIC (the radius of the maximum inscribed circle) and GCI (global conditioning index), and (ii) r MIC and GTI (global transmission index). Here, GCI and GTI were evaluated on the useful workspace. Instead of using atlases of performance indices, a grid search evaluation was applied to obtain a region in PDS near the optimum values for both maximization cases. This region gave a small bound for NSGA-II to start searching the optimum values of the kinematic constants. For simplification, a python framework for the multi-objective optimization called pymoo was applied to solve the optimization problem. Henceforth, the maximization for two cases yielded an insignificant difference of results in terms of optimum kinematic constants, r MIC , GCI, GTI, area of useful workspace, area of good condition workspace (GCW), area of good transmission workspace (GTW), and the area ratio of GCW and GTW to the useful workspace.