Media Controller: Weather Station

I worked with Christine Doempke and Matt Tennie.

Short description of our project: A weather station controlled by movement involving the body as a whole.

+ Ideas and Themes We Touched Upon +

• Awareness of breath. Where you’re breathing from (rib cage versus belly).
• Getting in sync with your environment through rhythm.
• Having a personal physical experience of a virtual world through personalized instructions that only the user can hear.

We gave up the idea of having explicit instructions pretty early on. But in retrospect, I think we kept the idea that there should be enough wiggle room in the interaction that each person’s experience of the weather station should be unique to their own way of exploring the piece.

+ The Media +

An animated grassland I created for my ICM Mid-term.

Below is a video of the application working through keyboard inputs:

+ Making it Physical: Three Inputs +

1. Accelerometer on the back of the head controls the x (horizontal) and z (back and forth) movement of the grass.
2. A breath band around the chest measures the extent of inhalation and exhalation which in turns controls the brightness of the sky.
3. A floor pad to measure shift in weight (e.g. running in place) which in turns controls the amount of rain falling from the sky.
4. Volume of music corresponds to the activity level on the platform and left-to-right panning is controlled by the x-axis readings from the accelerometer.

Below are early iterations of the breath band and head-accelerometer kit. In the final iteration, the accelerometer was pinned to a baseball cap and the breath band was sewn into a hoodie, which also housed all of the wiring down the back.

+ The Floor Pad +

I was responsible for constructing the floor pad.

I made a capacitive sensor for each foot. Basically two pieces of aluminum separated by foam bits. When you compress the foam, the current changes because the amount of resistance (air) between the two pieces of aluminum (capacitors) goes down. Got the idea from here.

I pulled my hair out trying to get the data to be clean. Spent a lot of time staring at graphs of the data in Processing, brought it into Excel, stared at a bunch of graphs.

In the end, I was able to massage four reliably distinct states out of the data:

1. Simply standing, no movement.
2. Mild shifting back and forth.
3. Medium shifting back and forth.
4. Really going for it. Fast and furious running.

Turns out capacitive sensors are just inherently noisy. Should have made a balance board our of four fsrs…which is what Christine and I are doing now for our final project.

+ Post-Mortem +

In the end, I’m not sure we ever reached clarity on the purpose of the weather station. What kind of physical experience did we want people to have? Was it dance? Exercise? Was there supposed to be a sense of competition? You versus the weather system? Was it intended to enhance awareness of how you move? how you breathe?

Watching others use the piece, people didn’t naturally move into a state where they moved as a whole, unconsciously responding to the media environment. Instead the many different sensors attached to different parts of the body emphasized the separate-ness of body parts and fragmented the way people moved. People took big breathes and then moved their head around and then stamped their feet, taking breaks to look at the screen to see the results.

I’d like to continue to develop this piece in the future and a few things I would do differently might be:

• Reduce the number of sensors
• Move away from a flat 2-D screen projection which seems to take people out of themselves and either create a more immersive arced floor-to-ceiling projection surface that wraps around you and/or control real-world mechanisms that create physical sensations such as fans (wind).
• Spend more time refining and calibrating the interaction between the user and the weather system so that subtler shifts in the quality of movement are expressed in the system.
• Create a clearer sense of competition between the user and the weather system to encourage sustained interaction.

What worked:

• Photorealism of the media. I simply liked, don’t have a well thought out reason to explain why.
• Multi-layered expression. The advantage of the 3 inputs (accelerometer, breathing band and feet) was that they enabled a richer, multi-textured expression of the body’s movement through the animated photograph. We had only a few short sessions to experiment with the completed system. Even so, I felt like there was a lot of room to play and experiment within the three simple controls we had created: changing light in the sky, the coming and going of the rain and the directional flow of the wind (ie. quick back and forths versus around in a continuous circle, etc).

Here is a poorly shot video. It’s not pretty, but it shows the system working.