Image Image Image Image Image Image Image Image Image Image

Being Brunel |

Scroll to top


No Comments

Adaptive Structures

Adaptive Structures

I love my indie games; there’s something about the smaller studios that gives their creations so much more life than their AAA counterparts. My latest fancy has been the superb Transistor, which you should probably go and play right now (work be damned). As I made my way though Supergiant’s wonderfully crafted world, however, I found myself thinking about the structural engineering behind it.

First: our structures need to become intelligent, and second: they must be capable of physically manipulating themselves.

You see, one of the core motifs of Transistor is the notion of self adapting structures: Buildings and infrastructure that alter their states, with apparent sentience, to cope, survive and provide for scenarios beyond their original usage and design. Is it a glimpse of the future, a pipe dream, or maybe even a reflection of something we have already?

Adaptive Engineering Today

Backing away from the hypothetical; the works of civil engineers are already regularly adapted. They are retro-fitted to comply with new standards, provide greater capacity, or to extend their working lives. In fact, this ‘manual’ adaptation forms a major part of the global engineering market. Similarly, in the UK at least, Managed Motorways are becoming the norm; where traffic can be quickly controlled and guided to maintain a more constant flow (which is nothing new- look at Sydney Harbour Bridge’s tidal-flow running.)

One could argue that many of our structures already ‘adapt’ to ‘unusual’ scenarios- through a combination of redundancy (those partial factors) and vigorous loading envelopes. In a similar vein they are protected from disproportionate collapse when under attack by ‘adapting’ to use alternative load paths. These examples, however, aren’t really adaptation in the 7-of-9 ‘sci-fi’ sense; they are more mundane- statistical fail-safes to make the unexpected expected.

Adapting For Tomorrow

To go beyond manual/redundant adaptation we need to solve two problems, first: our structures need to become intelligent, and second: they must be capable of physically manipulating themselves. Have hope, however, as work already exists in these fields, which might just pave the way to a Transistor like existence; although probably not for a while yet…

Intelligence in structures is already appearing in the digital world (where the issue of physical manipulation is inherently solved). On a day to day basis programs like Revit are making our structures parametric– building in the knowledge of how to develop and (to a limited extent) extend themselves. Researchers are already pushing this further with fabrication based geometric constraints and rudimentary form-finding algorithms that will one day be capable of establishing these parametric formulations themselves…

These developments, however, will have to make the leap from the digital world into ours before self-adaptation becomes a reality. This challenge is, arguably, a much harder one than finding intelligence. In the real world entropy rules, and you cannot create something from nothing. For the majority of contemporary engineering materials, it is not possible to perform in-situ reversions from their working state (steel will not flow into new shapes, and concrete can’t re-cast).

There are exceptional efficiencies to be made by being able to manipulate support conditions, and therefore internal stresses, on-the-fly.

Perhaps the closest thing we have is 3D printing. It’s already being used to construct houses, and in combination with the growing wave of construction-bots it is feasible to consider a slow adaptation system; where the structural intelligence commands these bots. Until a material is found, however, that can be freely replicated and manipulated this process will be slow: ‘adaptation’ through demolition and retrofitting- arguably worse-off then the current situation (as we’ll all be out of a job!).

The closest I’ve found to a truly adaptive structure (rotating floors, and moving façades aside) is the SmartShell; a lightweight timber shell designed by the University of Stuttgart. Sure, it cannot adapt beyond its nature; but it has reached exceptional efficiencies by being able to manipulate its support conditions, and therefore internal stresses, on-the-fly. The ability to adapt means that a 10x10m area can be roofed by only 40mm of material. Pretty impressive…

SmartShell Adaptive Roof

It’s not quite SkyNet, but we’re getting there… (Photo: Bosch Rexroth)

So who knows? Maybe Supergiant really do know what the world will be like in 40 years time

Submit a Comment

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.