Affine unfoldings of convex polyhedra (Q477459)

An interesting and well-known question in geometry, dating back to the Renaissance [\textit{A. Dürer}, The painter's manual. Abaris Books (1977)], deals with the question of whether a convex polyhedral surface may be cut along a tree spanning its edges so that the result may be unfolded into the plane as an embedding (e.g. [\textit{E.D. Demaine} and \textit{J. O'Rourke}, Geometric folding algorithms. Linkages, origami, polyhedra. Cambridge: Cambridge University Press (2007; Zbl 1135.52009)], [\textit{J. O'Rourke}, How to fold it. The mathematics of linkages, origami and polyhedra. Cambridge: Cambridge University Press (2011; Zbl 1234.00010)]). In this paper, the author proves that this is possible after applying an affine transformation to the polyhedral surface. This answers a question due to Croft, Falconer, and Guy [\textit{H.T. Croft} et al. Unsolved problems in geometry. New York etc.: Springer-Verlag (1991; Zbl 0748.52001)]. In particular, the main theorem of this paper gives the result in the case of a general cut tree \(T\) for a convex polyhedron \(P\): If \(u\) is a direction vector with respect to which \(P\) is in general position, and if \(T \subset P\) is a cut tree which is monotone with respect to the ``height'' direction \(u\), then a sufficiently large affine stretching of \(P\) in the direction of \(u\) is, when cut and unfolded along the corresponding affine stretching of \(T\), embedded. After observing that any such \(P\) is in general position with respect to an open and dense subset of directions in the two-sphere, and that monotone spanning edge trees may be constructed for each of these directions, the following corollary is obtained: An affine stretching of a convex polyhedron, in almost every direction, is unfoldable (admits an embedded unfolding along edges). Among the techniques utilized in the proof of the main theorem, the primary tool (which is used inductively on the number of leaves of a cut tree) provides a topological criterion for determining when a closed planar curve bounding an immersed disk is simple: If a disk \(D\) is immersed in \(\mathbb{R}^{2}\) with polygonal boundary, and if \(\partial D\) admits a decomposition into two arcs whose images are \textit{weakly monotone} under the immersion (that is, if each pair of arcs has an image that can be extended to an unbounded simple curve by attaching vertical rays pointing up and down from its initial and terminal points, respectively), then the immersion is an embedding.