Geometry in Action


From:           ins_acas@jhunix.UUCP (Catherine Schevon)
Newsgroups:     sci.math
Subject:        Unfolding Polyhedra
Date:           18 Feb 87 18:18:07 GMT
Organization:   Johns Hopkins Univ. Computing Ctr.

The problem of unfolding convex polyhedra, besides being part of
my thesis, has been open for many years. It was posed originally by 
G. C. Shephard, University of East Anglia, England.  A precise 
statement is:

Is it true that every convex 3-polytope has a simple unfolding?
An unfolding is produced by cutting a collection of the edges of a
polytope, imagining the polytope as a cardboard model, and flattening
the faces on a plane.  An unfolding is simple if there is no overlap,
i.e. the resulting shape on the plane is a simple polygon.

It is conjectured that the answer is "yes".  My advisor (Joseph O'Rourke) 
and I have proved the existence of simple unfoldings for
a few very restricted cases.

Observation: The collection of cut edges must form a spanning tree 
of the 1-skeleton.

Spanning Star Lemma:  If the 1-skeleton contains a spanning
star, i.e. one vertex is adjacent to all the others, then the unfolding
using the spanning star as the cut tree is always simple.

Hamiltonian Path Lemma:  If there is a Hamiltonian path spanning
the 1-skeleton such that all interior vertices of this path have an
exterior angle > pi when flattened, then the unfolding is simple.

We have an infinite class of polytopes for which some spanning trees
yield non-simple unfoldings.

We have also tried to find a polyhedron that cannot be unfolded.  It's
not hard if you allow non-convex faces (as in Karl Heuer's example) 
but surprisingly difficult for convex faces only.  Even a cube with 
nine spikes on each face turns out to have a simple unfolding.

A few related books that people might find interesting:

Cundy and Rollett, _Mathematical_Models_
Wenninger, _Polyhedral_Models_
Grunbaum and Shephard, _Tilings_and_Patterns_

It's a very interesting and difficult problem.  Any information or
references on it would be much appreciated!  Please respond to
schevon@hopkins and/or orourke@hopkins on Arpanet.

			-- Cathy

From:           highrisk@shellx.best.com (Michael Kelly)
Newsgroups:     comp.graphics.algorithms
Subject:        How Do I Flatten/Unwrap a 3D Object?
Date:           23 Apr 1996 10:14:33 -0700
Organization:   High Risk Ventures, Inc.

I need to be able to flatten/unwrap a 3D object.  That is, given a
relatively simple object (maybe 30 polys or so), I need to generate a 2D
image containing outlines of all of the polys laid out next to each other
(as much as possible).  If there is existing documentation on the algorithm,
please point me in the right direction.  If there is existing code for
the algorithm, please point me toward it or send it to me.

Thanks!

-- 
| Michael A. Kelly                  High Risk Ventures, Inc |
| President                                    PO Box 70690 |
| mkelly@hrvinc.com                        Eugene, OR 97401 |
| http://www.hrvinc.com                      (415) 359-4176 |

From:           eppstein@ics.uci.edu (David Eppstein)
Newsgroups:     comp.graphics.algorithms
Subject:        Re: How Do I Flatten/Unwrap a 3D Object?
Date:           23 Apr 1996 14:58:50 -0700
Organization:   UC Irvine Department of ICS

highrisk@shellx.best.com (Michael Kelly) writes:
: I need to be able to flatten/unwrap a 3D object.  That is, given a
: relatively simple object (maybe 30 polys or so), I need to generate a 2D
: image containing outlines of all of the polys laid out next to each other
: (as much as possible).  If there is existing documentation on the algorithm,
: please point me in the right direction.  If there is existing code for
: the algorithm, please point me toward it or send it to me.

As Joe O'Rourke will probably also tell you, it is open whether it
is always possible to do this (unfold the faces of a convex polyhedron into
a contiguous polygon on the plane).  If your object is non-convex it is
not always possible, but I don't recall seeing any discussion of the
problem's computational complexity in that case (I wouldn't be surprised
if it's NP-complete).

Here is my entry on the subject from "The Geometry Junkyard"
(http://www.ics.uci.edu/~eppstein/junkyard/), lightly edited
from HTML back into text:

Unfolding convex polyhedra:
Catherine Schevon discusses whether it is always possible
to cut a convex polyhedron's edges so its boundary unfolds into a simple
planar polygon (http://www.ics.uci.edu/~eppstein/junkyard/unfold.html).
Dave Rusin's known math pages include
another article by J. O'Rourke on the same problem
(http://www.math.niu.edu/~rusin/known-math/polyhedral/unfolding).
-- 
David Eppstein		UC Irvine Dept. of Information & Computer Science
eppstein@ics.uci.edu	http://www.ics.uci.edu/~eppstein/

From:           orourke@grendel.csc.smith.edu (Joseph O'Rourke)
Newsgroups:     comp.graphics.algorithms
Subject:        Re: How Do I Flatten/Unwrap a 3D Object?
Date:           27 Apr 1996 04:43:54 GMT
Organization:   Smith College, Northampton, MA, US

In article <4ljjqq$1mt@wormwood.ics.uci.edu>,
David Eppstein <eppstein@ics.uci.edu> wrote:
>highrisk@shellx.best.com (Michael Kelly) writes:
>: I need to be able to flatten/unwrap a 3D object.  That is, given a
>: relatively simple object (maybe 30 polys or so), I need to generate a 2D
>: image containing outlines of all of the polys laid out next to each other
>: (as much as possible).  If there is existing documentation on the algorithm,
>: please point me in the right direction.  If there is existing code for
>: the algorithm, please point me toward it or send it to me.
>
>As Joe O'Rourke will probably also tell you,  [...]

See if you can get your hands on Fukuda's beautiful code for actually
performing the unfolding (despite the openeness of whether it is
always possible).

@unpublished{Fukuda
, author =      "M. Namiki and K. Fukuda"
, title =       "Unfolding 3-dimensional convex polytopes: a package for {Mathem
atica} 1.2 or 2.0"
, year =        1993
, note =        "Mathematica Notebook, Univ. of Tokyo"
}

From:           jeffe@ocarina.CS.Berkeley.EDU (Jeff Erickson)
Newsgroups:     comp.graphics.algorithms
Subject:        Re: How Do I Flatten/Unwrap a 3D Object?
Date:           29 Apr 96 22:11:58 GMT
Organization:   University of California, Berkeley

orourke@grendel.csc.smith.edu (Joseph O'Rourke) writes:

>See if you can get your hands on Fukuda's beautiful code for actually
>performing the unfolding (despite the openeness of whether it is
>always possible).

The package is available by anonymous FTP, either from ETH Zürich or
Tokyo, depending on whether you have a UNIX box or a Mac.

	ftp://ifor13.ethz.ch/pub/fukuda/mathematica/UnfoldPolytope.tar.Z
	ftp://waka.c.u-tokyo.ac.jp/pub/test/UnfoldPolytope.sea.hqx

Komei Fukuda's home page has a nice picture of what the package does:

	http://www.ifor.math.ethz.ch/staff/fukuda/fukuda.html

Here is the README file from the ETH ftp server.  The author's
addresses are out of date.  Fukuda's email address is now
"fukuda@ifor.math.ethz.ch".  According to hir home page, Namiki's
email address is now "namiki@waka.c.u-tokyo.ac.jp".

----------------------------------------------------------------------
               September, 1994
    
    Makoto Namiki*
    Department of Social Science
    College of Arts and Sciences, University of Tokyo
    3-8-1 Komaba, Meguro-ku, Tokyo, 153, Japan
    namki@is.titech.ac.jp
    +81-3-3467-1171 ext. 245

    Komei Fukuda**
    Graduate School of Systems Management
    University of Tsukuba, Tokyo
    3-29-1 Otsuka, Bunkyo-ku, Tokyo 112, Japan
    fukuda@gssm.otsuka.tsukuba.ac.jp
    +81-3-3942-6876

This package contains Mathematica functions to 
unfold general 3-dimensional convex polytopes.
The package consisits of four files:

   ReadMe
   UnfoldGallery-notebook.ma
   UnfoldPolytope-notebook.ma
   BadUnfolding-notebook.ma
   UnfoldPolytope.m

The UnfoldPolytope notebook explains how to use this package
through examples.  UnfoldGallery-notebook contains some
unfoldings of some interesting polytopes and Badfolding-notebook
contains several examples of unfoldings which are of theoretical
interests.

If your environment does not allow you to use notebooks, 
we suggest you to get it printed before using the package.  
For the moment, the notebook environment is supported
by the Macintosh, NeXT and Windows versions of 
Mathematica. 

This beta release is designed to run under 
Version 1.2/2.0./2.1/2.2 of Mathematica. 

Please send your comments, bug reports etc. to:

  Makoto Namiki or Komei Fukuda
-- 
Jeff Erickson
jeffe@cs.berkeley.edu
http://www.cs.berkeley.edu/~jeffe

Part of Geometry in Action, a collection of applications of computational geometry.
David Eppstein, Theory Group, ICS, UC Irvine.

Last update: .