BlauEdge

• 

  Logo

Inspiration

We were inspired by recent efforts in Barcelona showing that over 100 local marine species returned to the coastline in just a few months. That rapid recovery made us realize how resilient marine ecosystems can be and also how little real-time data we have to monitor and guide that recovery.

What it does

BlauEdge is a real-time marine ecosystem monitoring system that evaluates the health of coastal zones. It combines environmental sensors (temperature, turbidity, dissolved oxygen, salinity) with on-device AI species detection to generate a biodiversity score for each location. The system helps identify: Which areas are healthy or stressed Where restoration efforts (like seagrass repopulation) should happen Early warning signs of ecological decline All of this is visualized through an interactive dashboard with a live map and AI-powered insights.

How we built it

We used an Arduino to collect environmental data like temperature and water conditions (further improvements). We trained a simple AI model using Edge Impulse to detect marine species. We connected the Arduino to Python to read the data, and then sent it to our HTML website We built a dashboard using HTML to visualize the data on a map and in real time We added a simulation mode so the system still works even without real sensors connected

Challenges we ran into

We had a hard time connecting the LogiTech camera to the Arduino board due to an OS clash. Embedding live data collected by Arduino into our HTML website was way harder than we imagined. Training the AI model with google photos was harder than expected, as we found no dataset of local species.

Accomplishments that we're proud of

We discovered Arduino, Arduino App Lab, Edge Impulse and many HTML features, and learned so much through this experience.

What we learned

How to connect hardware (Arduino) with a website in real time How to use AI tools like Edge Impulse for simple detection We learned how combining different parts (hardware, backend, frontend) is harder than building each one separately How to turn raw data into something people can easily understand The importance of debugging and adapting quickly when things don’t work

What's next for BlauEdge

We would expand the system by adding more sensors (e.g., pH, light intensity) and testing it in real underwater conditions for higher accuracy. Future improvements include:

• Training models on a real dataset of local marine species in Barcelona
• Deploying more nodes across the coastline
• Improving the AI to better detect ecosystem changes over time
• Turning BlauEdge into a scalable solution for other coastal cities There is still a lot to build, but we see strong potential for BlauEdge to contribute to restoring and protecting marine ecosystems.

Built With

• arduino
• html
• python