The hardware for our project
Illustrations of how analog to digital conversion works
Our circuit diagram for the sensors
Visualization of how the triangulation math works when sensors are placed in a rectangle
Some of our work on the triangulation math
A moment that inspired this project was when one of our school's Smartboards was acting up. Our teacher remarked about how the school rarely replaced the outdated device because of how expensive Smartboards are. We realized that creating a cheaper version of a Smartboard could bring this really efficient technology to more schools. This also inspired the idea of a compact device that could turn anything into a touchscreen. We realized that we could find the position of a tap on a surface using multilateration. Piezo sensors seemed the natural choice to pick up vibration through a surface.
As a result, we created Tap. Tap's primary purpose is to be used as a smartboard-type device, but it has other uses (discussed below) as well. Our product was created for $20 - a negligible cost compared to the $3500 that a smartboard can go for.
What it does
Tap is a compact device that allows the user to convert any surface into a touchscreen. The user plugs Tap into their computer and runs a simple configuration to map the computer screen to the surface (in our case, a table). After that, the surface can be used as a fully functional touchscreen with relatively low latency - perfect as a replacement for a smartboard, a compact and elegant computer keyboard, or even a virtual piano. Tap can be used without a projector in order to be fully portable.
How we built it
Tap uses four piezo sensors to calculate the position of a tap through a surface using multilateration. The piezos are connected to an op-amp circuit that leads into a comparator and a 555-timer IC. The comparator acts to convert the analog signal to an easier-to-read digital signal, while the 555-timer IC makes the signal predictable and consistent by outputting a signal for a given amount of time after a tap is detected. The arduino is connected to the circuits and notes the time in microseconds when the signal is received. We use a number of derived equations to convert the relative times of arrival to the position of the tap. Finally, the arduino sends a mouse move command based on the position to the computer and clicks.
Challenges we ran into
Overall, it was difficult to assemble and code the entire project over the course of the 36 hour competition. This was all the time we were given. It was also difficult to get the math behind 2-dimensional triangulation working, as we do not actually know the time of the original tap. All we know are the relative times that the sensors receive the sound waves of the tap through the surface. It took a while to derive an equation that would give us the relative time of the original tap so we could then perform traditional multilateration.
Accomplishments that we're proud of
We are proud of getting the piezos, which are incredibly sensitive and difficult to work with, to work together accurately.
What we learned
1) Teamwork 2) Simplicity 3) Improved coding/circuitry skills
What's next for Tap
The accuracy of the position of the tap could be improved, as well as the simplicity of the product until it is something people could use comfortably. We could also investigate other uses for Tap.