Agri-Tech-Farm-Water-Management

This project is necessary to explore as India is an agriculture country that depends on farmlands run by small farmers. Irrigation and water management are the biggest issues for these farmers.
Farmers depend on monsoon rains mostly for their agricultural cycle. Along with it, ground water also plays a vital role when it comes to a farmer’s selection of a crop for a certain season.
This AgriTech project works with small farmers, and it needs to develop an automatic sprinkler system that can run on soil and air parameters coming from embedded sensors. The sensors will send data based on the ground temperature and humidity; in return, the water consumption can be controlled.
Farmers are exploring mechanical and digital ways to help them maximise their yields with a scientific, data-driven approach. The primary goal of this project is to use moisture sensors to monitor the soil's moisture and temperature content during dry and wet conditions, send the data to the IOT core, control the function of sprinklers, and display the relative output on a dashboard that a farmer can easily understand with necessary call-to-actions.

Solution Design

Local machine simulates 20 sensors mapped to 5 sprinklers.
Each sprinkler is connected to 4 sensors.
Sending Data through MQTT to IOT core.
Data further goes to DynamoDB as metadata and sensor data
Sensor data further goes through a kinesis funnel to AWS lambda function
Weather API data is being stored in DynamoDB through a lambda function
This is a basic pub-sub system – The simulators are publishers and the sprinklers are subscribers.
The test cases use the various temperature patterns to test the sprinklers.
The final data is being shown in a mobile dashboard for the farmer/user.

The Kinesis data stream is shown below; Stream data

The sensor data comes in Json format and gets stored in the table sensor-data. The sensor sends in data on deviceid, timestamp, moisture, temperature, respective sprinklerID and state of the sprinkler.

Visual Dashboard

Analytics