A Neural Network-Based Automatic Crop Monitoring Robot for Agriculture

A Neural Network-Based Automatic Crop Monitoring Robot for Agriculture

E. Udayakumar (Kalaignarkarunanidhi Institute of Technology, India), S. Balamurugan (QUANTS IS and Consultancy Services, India) and P. Vetrivelan (PSG Institute of Technology and Applied Research, India)
Copyright: © 2019 |Pages: 10
DOI: 10.4018/978-1-5225-9246-4.ch013
OnDemand PDF Download:
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

The economy, being highly based on agriculture, demands innovative and reliable methods of irrigation. In this paper, an idea of automatic irrigation method is proposed. Automatic irrigation is done using a soil moisture sensor. The manual method of irrigation is done by using automated process. In this proposed method, apart from a moisture sensor, other sensors like PIR sensor, ultrasonic sensor, humidity, temperature sensor, and water level sensors are used. This method has additional features like GSM. In wireless systems, electricity will be provided through solar panels. Whenever the moisture content of the soil reaches its maximum threshold value, the system sends a signal to the motor and it turns ON. The robot can do its work automatically through artificial neural network. Every time the motor starts or stops, the user will get the status of the motor's operation through SMS. The robot will continuously monitor the crop field using wireless camera. This provides security for the agriculture land. The main advantages of this system include minimization of water wastage, & error reduction
Chapter Preview
Top

System Design

The countries where agriculture has a big impact on economy demand a highly efficient way of irrigation. A timely and consistent irrigation is need of the hour in such countries. In places where lack of water is not tolerated by the soil during irrigation, the excess of water provision is also not recommended for crops’ flourishment. Hence, a feasible irrigation for any land requires suitable amount of water with minimum amount of delays. Today’s world demands improved methods as compared to older ones to carry out processes faster. The world is moving towards automation of every process. In this proposed system, automatic irrigation system has been suggested which detects the soil moisture level and is programmed in a way such that if the water level in soil goes below a particular threshold value, it automatically switches ON the pumps to supply water. In this way, maximum results are obtained out of the fields and water wastage is also reduced to significant level (Udayakumar et. al., 2016) (Nallani, Sandeep, & Hency, 2015) .

In traditional agriculture, monitoring was based on kit and wiring system. Humans were used for these processes which consumed a lot of time.

In the proposed approach, artificial neural networks (machine learning algorithms) are used to teach the robot to automatically work over the field. Wi-Fi cameras are used for consistent crop field surveillance (Agrawal, & Singhal, 2015).

The assembly of the robotic system is built using DC motors, solar panels, LCDs, Wi-Fi cameras and sensors. The direction of DC motor is controlled by using Atmega 2560 microcontroller.

The block diagram of robot is shown in Fig. It includes a solar Panel, DC motors, LCD, Wi-Fi camera, sensors and microcontroller assembly.

Complete Chapter List

Search this Book:
Reset