Optimal Design and Implementation of a Bioinspired Soft Supernumerary Robotic Limb

Inspired by the biological characteristics of the octopus tentacles with multiple muscle fibers, fiber-reinforced actuators (FRAs) are expected to be incorporated into supernumerary robotic limbs (SRLs) to achieve a wide range of motions. This paper firstly introduces the actuation mechanism of FRAs, which exhibit multiple deformation modes including extending, expanding, bending, and twisting motions. The fabrication process of the FRAs is then explained, and the design and implementation of the SRL driven by FRAs are proposed. Furthermore, analytic models of FRAs are established to elucidate the influence of the configuration parameters and the input air pressures on the deformation amplitude in different deformation modes. Subsequently, the analytic models are validated by finite element simulation. In addition, a trajectory optimization algorithm is developed for SRL, in which a multi-objective optimization function is formulated and the trust region method is employed to optimize the design parameters of each FRA. This algorithm effectively assists the wearer in accomplishing predetermined tasks. Finally, prototype experiments are conducted to validate the system performance of the proposed SRL. This SRL can effectively improve wearing comfort, and can realize flexible motion in three-dimensional space. The research results demonstrate that the SRL driven by FRAs is superior to the state-of-the-art soft SRLs in terms of providing a comfortable wearing experience, enabling various motions, and ensuring precise control.