Krishi
Bringing science to your soil
The problem Krishi solves
Krishi addresses a common yet critical problem faced by farmers—making decisions without truly understanding the condition of their soil. Traditional soil testing is either expensive, slow, or simply unavailable in rural areas, which leads to improper use of water and fertilizers, ultimately affecting crop yield. Our solution is a smart, affordable, and scalable IoT-based system that continuously monitors soil parameters such as moisture, pH, temperature, and nutrient levels using sensors. What sets Krishi apart is its ability to function even in offline areas using a GSM module (SIM800L), which sends instant SMS alerts to farmers, eliminating the need for a smartphone or internet access. By applying moving averages and machine learning, we reduce data noise and improve prediction accuracy, offering farmers meaningful insights that lead to better resource use, healthier soil, and improved harvests.
Challenges we ran into
One of the main challenges we faced was ensuring reliable communication between the Arduino and the GSM module, as the SIM800L is highly sensitive to voltage fluctuations and power drops. Another challenge was calibrating the sensors to provide accurate soil readings, especially since different soil types and environmental conditions can affect the values. Integrating multiple sensors while maintaining power efficiency and data consistency was also tricky. Finally, finding authentic datasets to train our machine learning model and minimize prediction errors posed a hurdle, as publicly available soil data is often limited or inconsistent.
Tracks Applied (2)