Dimension optimisation of a dual-arm robot for enhanced stiffness in task-dependent polishing operations

Abstract

To address the issue of poor absolute positioning accuracy caused by the low stiffness of serial robot arms, we previously proposed a novel dual-arm robot (DAR) by optimising the configuration of a traditional robot. Building upon this foundation, the present study proposes a dimension optimisation method aimed at further enhancing the stiffness performance of the DAR in task-dependent polishing operations. A new stiffness index, the Polishing Deformation Index (PDI), is proposed to quantitatively evaluate the robot's stiffness under constant-force polishing conditions. The PDI considers both contact force and gravitational effects exerted by the tool and the workpiece. Furthermore, the maximum PDI (MPDI) is employed to characterise the stiffness along the complete polishing trajectory. A dimension optimisation model targeting the minimisation of MPDI and the singularity index is established, incorporating constraints such as joint range limits, kinematic feasibility and trajectory smoothness. Both simulation and experimental results demonstrate the effectiveness of the proposed index and optimisation strategy. Specifically, the surface roughness of the polished blades is reduced by over 25 %.