Computer aided design of a single degree-of-freedom six-bar mechanism via Monte Carlo layered optimization method

Abstract

Planar single degree of freedom (DOF) 6-bar mechanisms, characterized by simple control and high stiffness, find extensive applications across various robotic systems. Nevertheless, the augmented quantity of design variables presents challenges in mechanism design, particularly when confronted with multiple constraints. This paper presents the implementation of computer-aided design for a single DOF 6-bar mechanism. Considering the characteristics of the 6-bar mechanism, a classified kinematics modeling method is proposed. Subsequently, a constraint index system is established, encompassing trajectory, posture, performance, and additional auxiliary variable constraints. A Monte Carlo layered optimization method is then proposed. The introduction of the constraint scaling coefficient divides the optimization process into multiple layers, wherein the constraint conditions are dynamically adjusted at each layer. The Monte Carlo method is integrated to screen initial values and determine the number of variables in each iteration, facilitating efficient optimization of the mechanism under multiple constraints. Building on this foundation, mechanism design software is developed to diminish reliance on experience. Numerous examples demonstrate the rapid acquisition of 6-bar mechanisms satisfying multiple constraints through the proposed method, showcasing exceptional computational efficiency. This study serves as a reference for different users seeking to accomplish the efficient design of 6-bar mechanisms.