Worst case identification based topology optimization of a 2-DoF hybrid robotic arm

Abstract

In the design of robotic arms, structural topology optimization considering variable configurations with high computational efficiency is still a challenging issue. In this paper, the worst case identification based topology optimization of a 2-DoF hybrid robotic arm is accomplished, and the presented work mainly covers: (1) efficient worst case identification; (2) optimization problem construction and (3) iterative criterion and filtering method with fast convergence. The forward kinematics are investigated to identify the workspace. Thereafter, the equivalent external load is proposed to unify the effect of axial load and shear by force analysis and compliance calculation. The worst case is the load case with maximum compliance and can be located efficiently by searching for the maximum equivalent external load. The optimization problem is constructed based on the solid isotropic material with penalization (SIMP) interpolation scheme. For links with multiple worst cases, the objective function is constructed as the weighted sum of compliance under each worst case. For better computational efficiency, the modified guide-weight method is used to solve the optimization problem. To eliminate the mesh dependence and checkerboard problem, a guide weight filtering method is proposed. Under the guidance of derived optimal topology, the CAD model of the hybrid robotic arm is presented. The effect of the optimization is testified through performance comparison in finite element analysis. The optimization method can derive the optimal topology with global validity within allowable computational time and the optimization approach can be applied to other hybrid robotic arms as well.