Design, Characterization, and Optimization of the Soft Bellows Pneumatic Actuator

Abstract

In recent years, the research progress of soft actuators has been rapid in robotics research community. To expand the application of soft bellows pneumatic actuators, we fabricated a bidirectionally bent soft bellows actuator. Five soft actuator samples were fabricated by changing the geometry and material of the actuator. Under the driving pressure of 70 kPa, the maximum bending angle and the maximum bending output force are 92.9° and 2.2 N, respectively. In addition, a bending angle numerical model and a bending force finite element model are established to further characterize the motion characteristics of the soft actuator. Experimental, numerical, and simulation results demonstrate the effectiveness of the model for characterizing the characteristics of soft actuators. A verified numerical model and a finite element model are used to construct the meta-model, and multi-objective optimization is performed to optimize the structural parameters of the soft bellows actuator. After executing the particle swarm optimization algorithm, Pareto optimal solution is obtained, indicating that the bending angle and the output force cannot be optimized at the same time, and the appropriate scheme needs to be selected according to the actual needs. The result shows that the initial design of the soft bellows actuator can be improved according to the Pareto optimal solution in terms of bending angle and output force.