Behavior Analysis of Soft Bending Actuators Equipped with Layer Jamming Mechanism

Soft robots are known to be versatile to the environment and superior in interaction with humans or delicate objects. However, in the case of interaction with heavier objects, it requires to have more stiffness. Jamming actuators are novel structures with the advantage of tunable stiffness. This research proposes a combination of these two actuators, which makes a more capable robot that can grip and actuate more securely. The development of a mathematical model to anticipate the behavior of hybrid actuators is essential in the field of soft robotics. Due to the nonlinear behavior of these robots, analyzing their behavior can be challenging, particularly when considering the additional nonlinearity of layer jamming. Regarding the hysteresis effect in soft robotic actuators in both active and nonactive jamming, Prandtl-ishlinskii hysteresis (PI) modeling has been selected for identification; however, the PI model is incapable of identifying the active jamming robot due to its complex behavior. To address this problem, a novel modeling technique has been proposed based on modified PI. Moreover, an experimental setup has been made to test and validate the model. The results demonstrate prediction with errors of about 9% for a case without jamming activated and 8.5% for a jamming-activated, which is an adequate prediction. Overall, it is claimable that the identification method is valid and could be used in applications and even controller design.