Surrogate-based optimization algorithm for vibration trajectories of a translational flexible manipulator and experimental verification

Future advancements in robotics, aerospace, and precision equipment industries demand lightweight and flexible designs. However, structural properties of flexible structures result in undesirable dynamic behaviors such as residual vibration during rapid motion, which significantly affects the control accuracy and stability of the system. In light of this, this paper proposes a high-efficiency vibration suppression trajectory planning approach based on a surrogate-based optimization algorithm. Firstly, this paper establishes rigid-flexible coupled dynamic equations for a translational flexible manipulator system. Through the dynamic equation, we reveal the influence mechanism of translational acceleration and motion trajectory on the structural residual vibration. Then, this investigation formulates the motion planning problem for minimizing the residual vibration. Next, a high-efficiency planning algorithm is proposed to solve the issue based on the polynomial spline curves, the kriging model, and the expected improvement criterion. The efficiency of the algorithm was verified by comparison with the genetic algorithm. Finally, the experimental design and effect test are conducted. The experimental results demonstrate that the method can achieve the optimal vibration suppression motion trajectory with fewer than 200 samples. The optimized trajectory reduces the maximum residual vibration amplitude by over 79.14% and the residual vibration energy by 1-2 orders of magnitude, compared with the quintic polynomial trajectory.