+ installation design
+ microcontroller programming
My group and I decided we wanted to go big for our final project. With the prompt being simply "design a multimodal interaction," we wanted to create an environment for co-creation. In using their senses of touch and sight, three users must work together, mixing red, green, and blue light, to create a color to match the randomly generated goal. If they succeed, they are met with a light show that fills the room with every color. (Check out the bottom of the page for a demo video!)
Taking inspiration from the color knobs on old TVs, contemporary art installations, lantern festivals, album covers and everywhere in between, we sketched and mocked up ideas for our installation, and settled on an idea that would mix vision with tactility. We realized that people don't think of color in the same way that computers do: where someone sees a light pink, a computer would understand it as (243, 177, 241). We wanted to see how much of a challenge it would be for people to work together to bridge this gap.
The planned goal of our challenge was to have users be presented with a random color that lit up the central LEDs in the enclosure, and each user would use a simple, tactile interface (such as a knob) to alter their respective RGB value, which would apply to the surrounding lights in the enclosure. When the values were within a threshold of the central light, a light show would trigger. Along the way, tactile or auditory feedback could be given to each user to guide them closer to the goal, and allow them to inform their teammates that they were getting closer.
Development & Build
In the development process, I took charge of the coding and manufacture of the LED matrix and interface that would make up the interaction. A tough decision came in the construction of the LED network, where the addressable LEDs we were using were constantly breaking from their solder joints. We could choose to persevere and accept the unreliability of the system, but I suggested to make the hard decision to start over and use a different set of LEDs to both ultimately save the group's time and morale. By splitting a Neopixel strip into 68 individual addressable LEDs, we soldered each into a network of wires that spiraled outward from the center of our PVC frame.
I created a code that chooses a random value of color for the center lights and checks to make sure the color is not too close to white, as, through testing, we determined near-white colors were too difficult to match. The code then monitored the position of three potentiometers and mapped those values to an RGB spectrum which is sent to the outer LEDs.
To create a feeling of being surrounded by floating lights, like a lantern festival, we placed most of the paper lanterns just above eye level, to fill the users' field of view. We mounted the central, larger lantern lower to make it immediately visible as different from the surroundings.
Due to the initial ambition of the team at the beginning of the project, the work we allotted for ourselves turned out to be overwhelmingly large, so we had to make compromises throughout, the biggest being the removal of the sound portion of the challenge. However, we felt that the core of our project, which centered around the relation of vision to touch and being challenged to communicate with others about an abstract concept like color, was retained. We hope to add this feature back in future iterations of the installation.
Results & Future steps
Through the build, we learned that sometimes the most beneficial project decisions are initially the hardest to make. By pushing through this discomfort, we were able to make something we were proud to present.
The project debuted in a dark classroom to many smiles, shouts, and looks of surprise. HueCube was then transported to a foamcore cube (to block out ambient light) to be displayed at Northwestern's Fall Design Expo, where lines formed to go inside try it out.
One of my teammates and I will be submitting the project to ACTIVATE, to receive a larger budget to further develop HueCube. Primarily, we would like to make its electronics more reliable, the structure more stable, and, optimistically, develop the sound protion of the interaction. Hopefully, this will allow us to show HueCube to the public on the street in Chicago this summer!