AI Powered Chatbot - Smart AgriTech iOT

WhatsApp Chatbot
Falcon LLM Chatbot
Live Deployment of iOT kit
Web Portal (sensor monitoring)
Water Motor Functioning

Inspiration

Our inspiration came from observing the struggles that modern farmers face with pests, diseases, and resource management.Additionally as I come from an agricultural background I have seen day-to-day struggles faced by farmers and experienced it. We wanted to create a solution that leverages latest AI and LLM technologies to make agriculture more sustainable, efficient, and eco-friendly. Our aim was to empower farmers with tools that simplify their work and maximize productivity while protecting the environment.

What it does

Smart AgriTech is an AI-powered IoT kit that transforms traditional farming. It integrates smart sensors and automated systems to manage pests, diseases, and irrigation in real-time.It facilitates bookkeeping task for farmers and also provide some valuable insights for the farmers in real time to increase their crop yield.It combines all functionalities into one simple WhatsApp chatbot powered with Falcon LLM 180-b which is very easy for any social level of farmers to use. The system uses infrared and ultrasonic sensors to detect pests and eco-friendly methods to repel them. It also monitors soil and environmental conditions to optimize irrigation and predict pest outbreaks.

ML and DL models to detect pests, disease using normal images taken using mobile phones and provide immediate remedy for the detected disease/pest.

MongoDB is used for book-keeping tasks. Farmers can just upload an image of the bill/invoice and it will stored in database for further accounting tasks. This prevents manually preserving these copies which is tedious.

RAG (Retrieval Augmented Generation) is used to help farmers gain knowledge about anything they are not able to understand. For example, a farmer is not able to understand what is given in an ploughing system catalogue, he/she can upload the pdf to WhatsApp chatbot. It will convert the pdf to vector embeddings and allow farmers to ask any question from it.

How we built it

Software Components - Falcon LLM 180-b, MongoDB, RAG (pinecone db), langchain, Twilio, Flask, Huggingface Spaces, Tensorflow, Scikit-learn.

Hardware Components - Ultrasonic Sensor, IR Sensor, PIR Sensor, DC Water motor, UV Light, HIW Battery, NodeMCU, Arduino UNO, ESP-32 Camera.

Challenges we ran into

Sensor Calibration - We spent lot of time debugging and rectifying the data transfer from NodeMCU to the web flask server which we solved finally after hours of work.

LLM Response - We faced token limit exceed issue while querying the LLM. Later we used Unsloth Optimisation with 4 bnb quantized models to overcome the issue.

Cost Effectiveness - Faced some difficulties in selecting the most cost effective hardware components for integration.

Accomplishments that we're proud of

Successfully reducing the cost of the solution by 65% compared to traditional methods. Developing a multilingual chatbot that supports farmers in their native languages, making technology more accessible.
Creating an eco-friendly pest management system that reduces the need for harmful chemicals. Winning multiple awards in hackathons, proving the effectiveness and innovation of our solution.
Participating and winning some awards at other hackathons gave us some idea on how to approach a problem and solve it in the real-time scenario.

What we learned

Of all we learnt, the most important is the hardworking nature of farmers and importance of farming in our country.
We also gained some crucial technical knowledge about how an LLM's response is being processed and how data from NodeMCU travels to web flask app using a local server network.