<p>Soft-material actuation (SMA) design is based, for the most part, on building physical prototypes; a process, which is not only costly and time consuming, but also imposes stringent constraints on efficiently experimenting with new design configurations. This paper discusses challenges in developing <i>inverse-dynamics computational algorithms</i> for the SMA design considering large displacements and significant changes in shape. The <i>role of geometry</i> and need for general continuum mechanics and discretization procedures in the SMA design are demonstrated. A general continuum-mechanics procedure for predicting feedforward actuation forces for the interrelated <i>shape, strain, and spillover controls</i> (SSSC) is developed, and <i>geometric simplifications</i> that directly influence the SMA designs are discussed. In particular, using angle representation and rotation-based shear-strain measures, such as <i>Timoshenko beam theory</i>, and their relation to the SMA forces is explained. A new <i>control spillover problem</i> attributed to the soft-material under-actuation is identified, and a procedure for determining the control forces associated with the <i>actuation-placement points</i> is proposed. As discussed in the paper, some shape and motion configurations can be achieved with low <i>power consumption</i>, while others require extreme power that is impractical despite the softness of the material. Large-deformation examples are presented to demonstrate the limitations of conventional strain measures and challenges in the soft-material motion and shape control.</p>