Roach Buster

Inspiration

Our inspiration was to create a product that could easily help people within the real world with a problem that can affect any demographic. We decided upon pest control, as it's often unfocused in engineering fields and could provide a more niche focus for product design.

What it does

The Roach Buster is a tool built to exhibit intuition and user friendliness, leaving minimal room for consumer error with added AI assistance. The agent will walk the user through the process of identifying high-risk areas step-by-step, providing real-time interpretation of the surrounding environment through visual and measured data.

How we built it

We built this using Arduino for the sensors, Raspberry pi for the Google Gemini model, Onshape for the 3D modeling, and ElevenLabs for audio synthesization.

Challenges we ran into

Our challenges were integrating multiple devices into one product, as they all held different information to be interpreted. The biggest challenge was finalizing our idea and coming up with a starting point and initial design, as it took a few hours to really get the process started. We also faced challenges in the 3D printing process, as our inital print had failed, leaving us with not much chance to revise a design and instead causing us to rush to a final print

Accomplishments that we're proud of

We're proud of fleshing out our initial idea into a proof of concept, as well as being able to demonstrate our general idea.

What we learned

We learned many things, such as implementing camera vision on Google Gemini on a raspberry pi, connecting to speakers, using an Arduino nano connected to several sensors.

What's next for Roach Buster

To further improve our design, we would finalize an exterior design to allow the device to be more user-friendly in terms of it's hand held capabilities, as in its current state, the device is more awkward to hold than we'd like, but serves well as a proof of concept. In the future we could implement better sensor technology, as the current one we have are not as sensitive to change, and need more time to calibrate than preferred.