 |
My brother was creating artifacts from recycled materials as part of his final year work at Central St. Martins, London. I found a lot of his work and research very inspiring. |  |
Part of his final project was a large nest on a disused railway viaduct in Shoreditch, near Liverpool Street Station, London. While helping him build the structure i started thinking about woven, flexible structures and how this could be user in an architectural setting. |
 |
Another of my brother's projects; a nest in a hotel lobby in stockholm. Not really woven but bloody cool. |  |
In the background the City district can be seen. A big sign says 'Welcome to the financial capital of europe' |
 |
World Expo Pavilion by Shigeru Ban |  |
Bamboo and paper roof. |
 |
Contemporary artist and basket maker Masao Ueno makes some objects and installations that i find very interesting. How can I simulate this? How can I bring this into an architecture environment and make it work with modern manufacturing techniques? |  |
Masao Ueno. Many Living things in nature are flexible and can adapt to local environmental changes. Wouldn't it be great to develop structures that can take local constraints as inputs and grow within a given space? |
 |
Masao Ueno |  |
I love the old methods, used in basket making, to vary the weave to achieve different characteristics and functions. There is a very profound beauty in making something so complex and infinitely variable from something so simple. |
 |
Volume-weave: if we could make machines to do this kind of weave we could build some very interesting characteristics into a material. Variable elasticity and flexibility to suit a purpose. Like bone tissue. |  |
Very simple geometries underlying a complex end-result. |
 |
I started experimenting with control-rigs using bones in XSI. One to control the crossings and one to control overlap. In the foreground the virtual controls. |  |
Nodes attached to the vertices and edges of a mesh. |
 |
A lattice controlled deformation of the mesh. here the mesh is visible as a surface. |  |
|
 |
Weave with the rigs visible. |  |
The experiment didn't take into account the behavior of the material. |
 |
Although interesting this approach was hardly practical. in retrospect I'm sure it would have been possible to script parts of the process, even so the system quickly becomes very heavy and XSI can't really be used for anything other then making things look good and move. Too bad Catia doesn't have bones . |  |
Changing the local controls altered the characteristic of the weave |
 |
Same thing applied to a polyhedron. Here with 3 threads crossing rather than 2. |  |
Solid model of a bracket to hold the threads at each node. |
 |
CAM nesting. |  |
CNC programming. |
 |
Prototype. Flat weave with 3 threads/node. |  |
Prototype. We had a lot of fun with this kit, many different shapes can be built |
 |
Knitting in its simplest form. Knitting is perhaps even more interesting in that it uses only one thread. |  |
What if we could build a machine that would knit 3-dimensional components? |
 |
By varying the length of thread fed through each loop the global shape could be varied. Add to this a resin injection method and we have a machine to build composite elements without forms. Nice. More work to be done. |  |
L-system basket. |