Inspiration

I was inspired by the everyday challenges farmers face every day of their work, especially the time, effort, and knowledge needed to make the right decisions and plant the right crop. I wanted to use my love of technology to create a solution that would address a real-world challenge instead of just creating a screen-only project that looks good on the computer.

As a self-taught developer passionate about robotics, IoT, and AI, I asked myself: What if I combine artificial intelligence and a physical robot to make agriculture smarter? That is what led me to create Agri Seed Rover and Agri Seed AI – two connected projects driven by a single idea: using tech to help farmers.

What I Built

Agri Seed Rover – The Physical Robot

Agri Seed Rover is an agricultural robot designed to take care of planting for you. It can move around, plant your seeds, cover the seeds with soil, and water the planted seed.

I built the Agri Seed Rover with Arduino boards, C++, motor shields, DC motors, a servo motor, an ultrasonic sensor, a DHT11 sensor, an LCD, and a water pump controlled by a relay.

My main contribution was programming the code, connecting the hardware components, and allowing all components to work as one. I developed the movement, planting, obstacle detection, and watering features of the Agri Seed Rover.

This was developed as a team project and was awarded the first-place trophy at an embedded systems competition.

Agri Seed AI – The Intelligence

Agri Seed AI is the brain behind the smart farming solution I built. I wanted to go beyond automation of planting and wanted also to make it easier for farmers to understand their crops and make better decisions.

I created a web-application agricultural assistant with the following features:

  • An AI Chat Assistant that helps farmers and users ask questions related to agriculture
  • An AI Vision-powered Plant Doctor for image analysis and assistance on the identification of potential plant diseases
  • An Agriculture Crop Advisor for crop advice
  • An AI-powered Smart Irrigation Advisor for irrigation advice
  • Location-aware weather information and live weather updates from the location where the user is
  • Multi-language support for English and Kinyarwanda (Rwandan language)

For Agri Seed AI, I used HTML, CSS, JavaScript, PHP, and Gemini API to connect the frontend to a PHP backend that would make the AI requests.

Together the two projects Agri Seed Rover and Agri Seed AI are my answer to how I think technology can complement agriculture and farming.

How I Built It

I started from the problem I wanted to solve and then broke down the idea into smaller chunks that could be managed easily.

For Agri Seed Rover, I started by connecting the electronic components and then worked on the movement and planting system. I programmed Arduino with C++ to control the motors, servo, sensors, and water pump. I then tested the different features and combined them to have a planted seed.

For Agri Seed AI, I started by creating a working web application. I designed the user interface, integrated the PHP backend with the Gemini API, and created the different farming assistants available to farmers. I then integrated weather and location information for more contextual assistance.

Throughout, I used testing and debugging to improve each project. It became clear that a working solution involves more than just writing code and including feature. It involves patience, trying out multiple ideas, learning from other team members, and finding the best fit for the problem.

Challenges I Faced

One of the biggest challenges I faced was to get the Agri Seed Rover's hardware components to work together, communicate smoothly, and get the robot to perform well. I spent a lot of time testing, adjusting, and finding solutions to have multiple motors working at the same time.

Another challenge I faced was only a week before the competition, when the water pump stopped working. The team had to work together to find solutions to ensure we could continue moving forward.

Lastly, I faced technical issues connecting the frontend, the PHP backend, and the Gemini API on Agri Seed AI. I worked with limited access to components and resources to realize my plans.

This experience has taught me that to be innovative, all you need is resourcefulness and the right mindset. Innovation is not based on having all resources, but what you can come up with what is available.

What I Learned

Creating Agri Seed Rover and Agri Seed AI has helped me learn a lot of technical skills like Arduino and C++, PHP, JavaScript, AI APIs, PHP backend & frontend development, electronics hardware connection, and hardware programming. But it has also strengthened my skills on the softer side, like problem-solving skills, working under pressure, teamwork, debugging, and getting something working as a prototype.

Moreover, thanks to Agri Seed rover I learned that building on my ideas and having patience pays. When I got stuck, I found that with a little bit of perseverance, what started as a very small idea or project can grow into a successful one.

What's next for Agri Seed ROVER And AI

My plan is to keep developing Agri Seed ROVER and Agri Seed AI, linking the two ideas more closely and making the two projects grow.

For Agri Seed ROVER, I will work on making it more stable, more accurate, and easier to use for planting seeds on farms. I will also explore more ways of using the ROVER to make farming easier, better, and more efficient. For Agri Seed AI, I will extend the digital assistant and build on its features so farmers in low-resource settings can ask questions and get guidance in the language they understand.

Finally, I want Agri Seed to transition from being just a prototype project into a practical solution for farmers that can help save them time and support them with knowledge to make better decisions.

Agri Seed is more than a project; it is my message that using passion to contribute to real problems is possible. Innovation begins with a problem, begins with Patience and determination, and continues with people's help.

Built With

Share this project:

Updates

Submission history