IoT and ML Based Smart Irrigation System for Crop Prediction and Water Usage Optimization

The project proposes a Smart Crop Recommendation System implemented by the use of IoT that can help farmers choose the most appropriate crop based on the real-time analysis of the environmental data. Conventional forms of farming are usually on experience and few observations and this can make them take less than optimal decisions. To overcome this, the system will be combined with ESP32 microcontroller and a combination of sensors which will include DHT11 temperature and humidity sensor, soil moisture sensor, soil pH sensor and rain sensor to continuously measure the important parameters of agriculture. The gathered information is passed to the ThingSpeak cloud platform through Wi-Fi, which also gives a possibility to access the information remotely and$\log$the data. Through cloud storage, farmers are able to monitor the status of fields in the long run and have access to the past that can be used in longterm planning. The dataset on ThingSpeak is received with the help of MATLAB and a Decision Tree classifier is used to predict the crops. According to the environmental conditions such as temperature, humidity, soil moisture, pH, and rainfall, the system suggests the most appropriate crops such as rice, wheat, maize and chickpea based on hard-coded thresholds and trained data trends. The solution will improve precision farming by making decisions based on data and make them more resource-efficient and productive. The system does not only enable farmers to have the right information in a timely manner, but it also enhances sustainable agriculture through efficient use of water, fertilizers, and other agricultural inputs. Internet of Things with machine learning thereby offers a reliable, efficient and scalable instrument in contemporary farming.