AgroBOT

The Problem

Modern technologies have given human society the ability to produce enough food to meet the demands of more than 7 billion people. However, food security remains threatened by a number of factors including climate change, the decline in pollinators, plant diseases, and others. Plant diseases are not only a threat to food security at a global scale but can also have disastrous consequences for smallholder farmers whose livelihoods depend on healthy crops. In the developing world, more than 80 percent of agricultural production is generated by smallholder farmers (UNEP, 2013), and reports of yield loss of more than 50% due to pests and diseases are common (Harvey et al., 2014). Furthermore, the largest fraction of hungry people (50%) live in smallholder farming households, making smallholder farmers a group that's particularly vulnerable to pathogen-derived disruptions in food supply.

The Solutions

In field of agriculture, the detection of disease in plants plays an important role. To detect a plant disease in very initial stage, the use of an automatic disease detection technique is beneficial.

The existing method for plant disease detection is simply naked eye observation by experts through which identification and detection of plant diseases is done. For doing so, a large team of experts as well as continuous monitoring of plants is required, which costs very high when we do with large farms. At the same time, in some countries, farmers do not have proper facilities or even idea that they can contact to experts. Due to this consulting experts even cost high as well as time consuming too. In such conditions, image processing based automatic technique proves to be beneficial in monitoring large fields of crops.

A lot of research has been done for disease identification of plants from images. Image-based disease management and surveys have a long history of more than 90 years when aerial images were used to study crop disease. Disease detection and identification have improved since then and informative and sophisticated analysis is being carried out. Image-based disease identification is under continuous development. Studies have shown that image processing methods are effective in identifying plant species or diseases from leaf images. Imaging devices have become cheaper and common with better quality images. Reliability, accuracy and precision of machine identification tasks have also continued to improve.

Different approaches to the identification and quantification of plant disease are in practice and leaf image-based identification of plant disease is one of them. It is by far the easiest way to automatically identify plant disease and can be used for the identification of various diseases. The occurrence of plant disease causes specific changes in the texture and color of the leaf and therefore leaf imagery can be used to extract color and texture-based features to train a classifier.

My Proposed Solution

I will make an autonomous robot that will take the plant image of an agricultural field and identify if there is any disease in the plant. If it detects any disease at any plant it will spray the pesticides automatically in that plant. At the same time, it will measure important soil parameters like NPK and based on the sensor value it will automatically apply the exact amount of liquid fertilizer required.

For driving the robot I will use Arduino Mega. Arduino will drive 4 motors, one stepper motor that will vertically move the sensor toward the soil, and liquid pumps for spraying fertilizer and pesticides. On the other side for capturing the plant image and processing the images I will use Raspberry Pi 4 and the official Pi camera module. These images will be processed on the Pi using Open CV and TensorFlow. I will also use Edge Impulse for doing some machine learning.

Power will be provided to the robot from a high-capacity 4-cell Li-po battery. Cytron 40A motor driver for driving the geared motors. I will use 4, 130mm X 60mm wheels with the gear motor. Raspberry Pi will get power from an external power bank.

Current Status of the Robot

The hardware part of the robot is already made as shown in the cover image. Now I will connect the Raspberry Pi and camera with the system and start working with image processing. This will be the most exciting and challenging part of the robot.