No abstract

Nowadays, soft actuator development has become a big trend due to higher safety and more complex movements. One of the most interesting types of these actuators is called the “soft fiber-reinforced bending actuator,” which has been chosen for investigation in this study. The aim is to provide a codified model to investigate the static behavior and deformation of the actuator in both free and positional constrained conditions. Indeed, modeling of these actuators in the presence of external factors and constraints is significantly challenging due to distortion of conventional constant curvature assumptions applied in free motion. In this paper, hyper-elastic theories have been utilized to model the free motion behavior of a soft-bending actuator made from elastomeric materials. Then, the Euler–Bernoulli beam theory for bending deformation is used to investigate the effect of positional constraints on the actuator’s deformation. In addition, an optimization method is utilized to estimate the required internal pressure, which in turn tunes the contact force between the actuator tip and desired objects/obstacles. The experimental results are presented to validate the proposed theories in both free and constrained conditions. In this regard, first, the actuator’s behavior in free motion is properly investigated and proven valid. Then, the effects of positional constraints on the actuator and their contact force tuning are analyzed.

Soft bending actuators show high adaptability for applications such as rehabilitation or grasping. Although constant curvature assumption has been extensively used for free motion modeling, these actuators do not bend circularly when interacting with the environment. In such situation, conventional bending sensors cannot provide useful information on their shape. In this paper, the Finite Rigid Elements approach is utilized to model the behavior of a soft pneumatic bending actuator in free motion and contact. With this method, the variable curvature configuration under different external loads can be modeled. Then, the contact force between the actuator and an object located in a specific position is estimated utilizing the Gradient Descent optimization method. Experimental results verified the combinatorial proposed approach for both force estimation and structure deformation.

No abstract