Dynamics and Control of Soft Robots With Implicit Strain Parametrization

Efficient modeling of soft robots for design optimization and control is an active area of research. When representing soft robots as an assembly of rods and rigid bodies, strain-based parametrization has proved to be able to reduce the required number of degrees of freedom drastically. However, this reduction ability strongly depends on the choice of strain basis employed to describe the system. In this letter, we proposed a new implicit strain parametrization, and we showed its use to represent the system's configuration manifold with no more degrees of freedom than the number of actuators. Coupled with standard strain bases, this parametrization is applied to the dynamic simulation and control of soft robots employing a handful of generalized coordinates, drastically simplifying the control design problem. The approach is validated against a high-order model and exploited for shape (configuration space) and tip-pose (task-space) control of a complex tendon-driven soft robot.