Toward Actively Morphing Lattice Structures
Computational Design and Demonstration
F. Bressan (TU Delft - Industrial Design Engineering)
J. Wu – Mentor (TU Delft - Industrial Design Engineering)
M. Özdemir – Mentor (TU Delft - Industrial Design Engineering)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Lattice structures are traditionally designed to be rigid, holding a single shape. Adaptive lattices move between shapes, letting one structure serve changing needs, such as those of a morphing wing rib. Existing actuated lattices follow transformations defined at design, and methods for placing actuators keep all other members at fixed length. None of them releases a member to change length freely, although mixing fixed, actuated and released members could bring a lattice to targets chosen after its design. This thesis pursues that possibility as an inverse design research, for planar pin-jointed lattices of given topology. Given a target curve, it asks which members to fix, actuate or release, how far each actuator must travel, and under what conditions the target can be reached. A computational framework searches the role assignments and computes the actuator travels that bring a set of target nodes onto the curve. Four structures were studied; in the three of these that carry redundant members, no assignment moved the target nodes without a release. These members bind the frame into groups whose lengths cannot change independently, so an actuator stays blocked until releases break every group it belongs to. Which members are released therefore decides whether such a lattice can adapt at all. Tolerance and design preferences, in contrast, change which assignment is selected, not whether one exists. On physical demonstrators built for two of the structures, four computed assignments, with three to seven actuators, ran without adjustment of the computed travels. Taken together, these results state four conditions for adaptation: the roles form a mechanism its actuators control; that motion moves the target nodes toward the curve; an exact fit needs at least one actuator per placement and spacing requirement; and the path avoids folds and stays within travel limits. The first and third can be checked before any search is run. Actuation can thus be computed rather than chosen by hand, a step toward lattices that reconfigure after deployment.