Carbonator | Devpost

Cactus with automatic watering system
Automatic light system at the top
Closer look at the sensors used
Closer look at the cactus with the soil moisture sensor and water tube
An image of the two main dev boards used to drive hardware
Wider picture of the automatic lighting system
Magtag eink dev board
Greenhouse enclosure
Domain

Inspiration

The motivation for our project is to reduce the impact of climate change, protect the environment, and increase yield for farmers. It boosts productivity of the farmers by monitoring planst using computer vision. In addition, it makes the farmer’s life easier to know when their plants are watered and monitored for weeds.

What it does

First of all, the Carbonator can help farmers increase the yield by creating a better environment for plants. It can track and manage the amount of gas (CO2, CO, Ethanol, NO2) and volatile organic compounds exist in the atmosphere. It can also give alerts to farmers whenever there is a sign of harmful plants by the weed notifier (using machine learning model). Moreover, the device can automatically provide supplemental lighting for the plant growth via the UV monitoring system. Secondly, our project has the ability to reduce the burden on the ecosystem or people’s everyday’s life with the carbon offset via smart contract carbon credits. The device is able to calculate and reduce carbon footprint through utilizing micro agriculture and carbon credits from the blockchain.

How we built it

The hardware that we used is as follows:

Capacitive Soil Moisture and Temperature Sensor
BME688 - Temperature and humidity
SCD40 - CO2
Multichannel gas sensor - CO, Ethanol, VOC, NO2
Arduino Portenta H7
Arduino Portenta Vision Shield
Submersible Pump
MagTag - ESP32S2 Eink Dev Board
Arduino Mega

The software tech stack is as follows:

Python
C++
Arduino
Twilio
Microsoft Azure
CosmosDB
Edge Impulse
AdaFruit.io

Challenges we ran into

The challenge that we ran into is the time constraint. We also learn the new tech stacks to ensure that our application works efficiently.

Accomplishments that we're proud of

We are proud that we were able to train and deploy a computer vision model on the edge with Arduino Portenta. We are also proud that we were able to implement the number of sensors that we did and have them all be connected to the cloud. We have 100% real time data.

What we learned

We learn how to work and collaborate in a short amount of time as team members. Moreover, communication is key to making sure that everybody on the team is on the same page and understands each other.

What's next for Carbonator

Increase the usage of green energy in solving/mining blockchain. Find ways to reduce the cost of materials so that we can further spread the device and let more farmers get in touch with it.