Inspiration
When researching ocean problems and brainstorming ideas for what project to work on, we found that there was a significant lack of data about organisms in the deep sea. Because of the lack of light at low depths, sound is the ideal way to identify various organisms and how they move. This inspired us to explore using techniques such as beamforming and fourier transforms to identify both the type of organism and the location using an array of 4 microphones at different locations.
What it does
The project uses a simulated environment to display organisms moving around in three dimensions. The simulation produces sample data which is fed to a beamforming algorithm which first decomposes the frequencies of each organism, and then uses the phase differences of each of the 4 sensors to locate the source of the sound for each frequency (the organism emitting that frequency).
How we built it
We used Python for web development and for our algorithms.
Challenges we ran into
Implementing the website was challenging for it to function.
Accomplishments that we're proud of
We achieved a 95% accuracy in our predictions of locations.
What we learned
We learned more about website development and different triangulation algorithms.
What's next for DeepTrace
Integrate hardware, using more species, and real world testing.
Log in or sign up for Devpost to join the conversation.