Performance improvement in soft robots by exploiting viscoelastic non-integer-order dynamical systems

Soft robots comprised of flexible materials, such as inflatable actuators and capacitive touch sensors, have recently gained attention for their capability to produce complex and adoptable motions through nonlinear deformation, simplistic synthesis, ease of fabrication, and low cost. A continuum soft body can be modeled as a high-dimensional, locally coupled nonlinear system. Soft robots are defined by their nonlinearity and the viscoelasticity of their soft components; these properties cannot be fully captured by ordinary linear differential equations without using many model parameters. The complex behavior of viscoelastic materials can be approximated with the classical Maxwell and Kelvin–Voigt models via many parameters, which dramatically limits model estimations and optimal design mechanisms. In this study, we propose a fractional model that captures complex viscoelastic behaviors with fewer parameters. We will present new insights for analyzing and designing complex soft robots by combining control theory, fluid dynamics, network science, and fractional-order calculus tools.