Abstract
Soft actuators, composed of pliable materials, are increasingly adopted in industrial grippers owing to their inherent flexibility, elasticity, and safety attributes, making them well-suited for anthropomorphic robotic applications. A significant gap in existing literature is the detailed exploration of hand abduction movements. Addressing this gap, the present study makes three principal contributions. First, it introduces the abduction soft-actuator (ASA), an innovative design tailored specifically for robotic hand abduction. Second, it establishes an analytical framework using the large deformation virtual beam (LDVB) theory for soft elastica, which enables a detailed analysis of the intrinsic physical properties of the actuator's internal membrane. Third, this study highlights the ASA's versatility, showcasing its ability to integrate membranes and springs seamlessly, thereby broadening its utility across diverse design paradigms. Empirical results underscore the ASA's capability to predict operational angles with varying spring stiffnesses, enhancing the precision of spring selection for a range of applications. This ASA exhibits an abduction angle ranging from 14.17 deg to 27.78 deg as the spring stiffness K varies from 200 N/m to 1600 N/m, with a root mean square error associated with these measurements ranging from 0.3321 deg to 1.2651 deg. Unlike traditional soft actuators that typically utilize a single material, the ASA demonstrates modularity, facilitating easy adjustments of springs to meet varied requirements. Contrasting with the typical case-by-case analytical approaches, the ASA significantly extends its applicability. Validation experiments using inflated silicone membranes corroborate the LDVB theoretical framework, suggesting that these empirically based estimations are conducive to analytical prediction. Collectively, this methodological advancement not only bridges the current technological divide but also enhances the understanding of soft actuator mechanics across a wide range of applications.