Image Image Image Image Image Image Image Image Image Image

Bartlett School of Architecture, UCL

Scroll to top


No Comments

How do we make large scale performing inflatables that create immersive and collaborative experience? 

How do we make large scale performing inflatables that create immersive and collaborative experience? 

A Weekly report of the project E-COllaborative generator

Last week, we have a presentation on our preliminary idea of thinking one person’s immersive experience with the E-COllaborative generator. As one paddling, its energy is contributed to the architecture’s breathing body. An animated creature is activated, echos the person’s existence with organic motions.

Video.1 Prototype animation by Ting-yun Wu

In this stage, we focus on a ‘self’ interacting with the machine, and signaling out the rhythm to the external environment as a whole, as another ‘reflective self’. This is like a solo dance performance. As the choreographer Mathilde Monnier’s word goes ‘The concept of solo dance not only indicates one person on stage performing itself but also present the many other lonely people, to reflect the loneliness of others in crowd.’¹ Then in the next stage, we would like to know what will happen if we engage more people? Could multiple people construct a ‘self’?

We found that it may be the right time to think from a technical perspective, How do we make large scale performing inflatables that change their shape to involve multiple people interacting ? 

Fig.1 Input Mechanism Study Diagram in Fusion by Yuting Chen

1.1 Are there historical and current examples of large scale performing inflatables that change their shape? 

Most of the traditional large scale performing inflatables are dealing with air bags, air muscles or balloons. No matter what form, material it has, it owns the feature of responsive and dramatical, unpredictable changing. Projects such as ‘ birds’ by Chico MacMurtrie, ‘Fragile Occupancy Cloud ’ by Hidemi Nishida studio or even the air dancers before shopping malls are using air bags as medium, controlled joints by joints and change their outcome normally in three dimension, thus made themselves never doing the same thing twice. The performance is not random either. The machines encourage people to reinterpret technology—regain control of the air pressure and manually manipulate their behaviour. By giving people a dynamic preview of the real-time performance, it helps people to understand the air flow and pressure in aerodynamic  settings and feedback to it.However, it does not necessarily mean that it is better than deformable inflation with completely predictable behaviour.

Fig.2 Examples of early large scale performing inflatables controlled joints by joints
 by Chico MacMurtrie and Ichiro Kato
Fig. 3 Examples of large scale performing inflatables controlled by balloons and pressure sensors
air dancers: the unpredictable behaviour

 Actually, there is no hierarchy between the predictable behaviour and the unpredictable behaviour in these things. In the context of Stephen gage’s the trivial machine² , the trivial machines and non-trivial machines each has its own characteristics, we are able to get delight from a trivial machine. However, I caught a moment that projects involve inflatables has a statement that change between trivial machines and non-trivial machines.  As Jea-Luc Nancy ¹(a philosopher) said ‘The dancer takes the role of space, the space was bended surrounding the body ,at the same time, the body itself is the space which can be folded and opened, affected by some invisible mass, that is the black hole of ‘self’’. 

an idea of creating a body extension =  a method of signalling out from what we are doing into the environment around us

In our project, when the inflatables controlled by human body or body extension, being bend and fold in space, they become a visible signal of the subtle impact of the human body and its extension on the space. These are the results of effects inside effects to a point where we cannot predict outcomes visually with any accuracy beyond certain set time periods. But we may find a balance between our body and  whole environment it creates through physically controlling the input system and learn from the predictable and unpredictable outcome that the output feedback.

1.2  How do large scale performing inflatables work? 

a method of creating energy

Video.2 Green Inflatable mimic PET by Ting-yun Wu

Franco Mazzucchelli’s pneumatic sculptures³ transform the static interior with liveness. The rhythm of the sculptures’ inflated and deflated motions are in smooth speed, creating a contemplating mood in the environment. It inspires us to think of creating different moods and emotions with controlling the speed and volume of inflating and deflating.

In physical part, playing with plastic air bags and joints (plus a set of pipes, valves, hoses, manifolds, adapters, connectors, etc.), we began to set up a small scale pneumatic kit which efficiently allows the air inflate and deflate in seconds.

Fig.5 Airbag workflow by Yuting Chen

1.3 how do(and how could) large scale inflatables change their shape?

Fig: a, b, c Curvature control by size of the airbag
Fig.d, e Combination of the two different airbag sizes?

Learn from an experiment did in MIT Media lab, in this experiment showed above, the composite material consists of two layers: plain paper layer and three different sizes low-elasticity plastic bladder. The airbag is made by plastic welding and is bonded to the paper layer. When inflated, the airbag behaves like a bicep (muscle pulls the arm up) and compresses itself to bend the surface. Curvature control by different sizes the airbags( Fig a, b, c) have very subtle differences. While we combine the two different airbag sizes in one paper, the shape changes differently.

Fig.7 Workflow diagram by by Yuting Chen
using gauge to measure how much air need to inflate different size air bags
using hand valve to bypass the air between the air bag and the air tank
Fig.8 Air tanks moving vertically

This experiment using gauge and hand valves allows me to quantify the air speed and direction that make air bags shrinks and air tanks moving vertically. Through this workflow, the design has jump to another question: How can large scale performing inflatables that change their shape  be treated as whole body extensions?  Could a human operator or operators power the inflation of a large scale performing inflatable ?

As mentioned in the beginning, we are also looking at the context when more people involved and reflect themselves during the performance.

In the next step,  we will try to examine the idea of can the idea of creating a body extension, be a method of creating energy and be a method of signalling out from what we are doing into the environment around us simultaneously. Learn from the small scale experiments and develop components design and structure diagram according to different scenarios that give delight in our machine, make the statement flow between predicable and unpredictable. Still a long way to go!


  1. Monnier, M., & Nancy, J.-L. (2005). Allitérations: Conversations sur la danse. Paris: Galilée.
  2. A trivial machine which has a predictable output, such as a piece of music collected in the playlist that we can listen several times. The non-trivial machine which has an unpredictable output, such as a live performance where a singer improvised according to one’s mood. What’s more, non trivial machines do not have to involve humans. Probably our most usual experience of non trivial behaviour is the weather and the way that local airflows work    Stephen gage. the wonder of trivial machines(2006) and the extended conversation between him and the author
  3. Work details:“Untitled”, (Venice, Canale Gallery indoor), 1969PVC, air, timer, air pump (four compositions, each with four inflatable arches in orange)140 × 200 × 200 cm. Source website:
  4. The experiment came from the paper. Yao, L., Niiyama, R., Ou, J., Follmer, S., Silva, C.D., Ishii, H.: PneUI: pneumatically actuated soft composite materials for shape changing interfaces. In: Proceedings of the 26th Annual ACM Symposium on User Interface Software and Technology, pp. 13–22. ACM, New York City (2013)

Submit a Comment