"Shape Optimization of Columns and Arches through Load-induced Curvature Adaptation"

Plants use light, contact, and gravity sensors to optimize their shapes in heterogeneous environments. The shapes of roots, tendrils, and shoots are varied by quasi-static modulation of curvature. In contrast, shape adaptation remains rare among engineering structures.  Inspired by the biological examples, we investigate simple models of slender structures with actuators enabling direct or indirect curvature modulation. Our goal is to identify control policies enabling successful shape optimization under slowly varying loads. The example of a soft arch with curvature generated by incompatible boundary conditions shows that shape optimization can be viewed as a navigation task on an equilibrium manifold with complex topology. We also investigate a discrete structural model, where curvature is tuned directly by rotary actuators, and the goal of shape optimization is to eliminate bending. It is found that shape optimization is achievable despite limited sensor data. Semi-active control strategies, preferable from the point of view of reliability are also identified.

Einladung als PDF