A Multi-Objective Optimization Design Method of Shift Manipulator for Robot Driver Using SA-PSA

Chen Gang,Xu Xinyao,Wang Liangmo,Zhang Weigong

DOI: https://doi.org/10.1007/s00158-022-03301-1

IF: 4.279

2022-01-01

Structural and Multidisciplinary Optimization

Abstract:With the widespread use of robots in the world, including rescue operations and special surgeries, the precise design of robots is of particular importance. The shift manipulator is a part of the robot driver structure which is composed of a two-degree of freedom parallel mechanism with electric cylinders. The singularity analysis and inertia torque analysis are conducted based on the kinematic and dynamic models. The integrated design variables for multi-objective optimization are proposed to evaluate the kinematic and dynamic performance of the manipulator. Given dimensional parameters as design variables and geometric constraints in a moving area, the multi-objective integrated optimization model is established and solved by the Simulated Annealing Particle Swarm Algorithm (SA-PSA). Furthermore, the optimized performance is compared with different optimal methods, and the effectiveness of the presented method using SA-PSA is verified by performance simulations and test. The results demonstrate that the kinematic and dynamic performances are improved 22.22% through minimizing two integrated design variables.