Model-based design of a pneumatic actuator for a dynamically reconfigurable socket for transtibial amputees

Front Bioeng Biotechnol. 2024 Dec 23:12:1459056. doi: 10.3389/fbioe.2024.1459056. eCollection 2024.

Abstract

In this work, a cost-effective, scalable pneumatic silicone actuator array is introduced, designed to dynamically conform to the user's skin and thereby alleviate localised pressure within a prosthetic socket. The appropriate constitutive models for developing a finite element representation of these actuators are systematically identified, parametrised, and validated. Employing this computational framework, the surface deformation fields induced by 270 variations in soft actuator array design parameters under realistic load conditions are examined, achieving predictive accuracies within 70 µm. The results elucidate how individual design factors influence surface deformation and, consequently, pressure distribution. A novel speckle imaging technique is employed to address the complex non-linear deformations, enabling surface displacement measurements with an accuracy of approximately 40 µm. These measurements confirm that the Ogden N3 model can predict actuator deformation with an accuracy of 16%. These findings elucidate the relationships among actuator geometry, material behaviour, and surface deformation. Although demonstrated in a dynamically reconfigurable socket for transtibial amputees, these insights are readily transferable to other robotics applications that require soft, deformable, load-bearing interfaces. This validated modelling strategy and imaging technique provide a foundation for optimising soft actuator arrays, ultimately improving user comfort and enhancing the functionality of future prosthetic and robotic devices.

Keywords: adjustable prosthetic socket; computer vision; finite element method; hyperelastic modelling; image registration algorithm; pneumatic soft actuator array; reconfigurable socket.

Grants and funding

The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. The authors thank the support provided by Auckland Bioengineering Institute (ABI) and New Zealand Artificial Limb Service (NZALS) for this study. AT is supported by a James Cook Research Fellowship provided by the Royal Society of New Zealand.