Preface

Digital Agriculture (DA) refers to the practice of modern technologies such as sensors, robotics, and data analysis for improving the sustainability and profitability of farms, while at the same time increasing crops' yield and quality. Digitalization of agriculture is a technological response to climate change, global warming, and water scarcity that are affecting agricultural productivity and food security. Reports indicate that nearly one-quarter of the global greenhouse gas emissions comes from crop cultivation and livestock farming, which can significantly decrease crop yields. In modern farms, measurements from multiple in-situ sensors combined with images, maps, and data generated by satellites, drones, artificial intelligence, and prediction models are delivering detailed agronomic information on crop conditions and field variabilities to improve various aspects of farming practices whilst diminishing risks and uncertainties. The inputs and outputs of digital agriculture, as shown in **Figure 1**, have evolved based on data streams and flexible data-sharing services and contributed to mitigation strategies for climate change by providing a series of scientific solutions toward reducing pesticide usage, chemical fertilizers, and minimizing energy demands.

Until 2010, growers had to rely on Global Positioning System (GPS), ground-based sensing platforms, satellite maps, and local sensing devices such as data loggers to monitor their fields, identify deficiencies, and improve crop yield via better management of the resources (i.e., variable rate technology). These practices were referred to as Precision Agriculture (PA) and Smart Farming. With the rise of Unmanned Aerial

#### **Figure 1.**

*Schematic demonstration of the inputs and outputs of digital agriculture from a general perspective (Source: AdaptiveAgroTech).*

Vehicles (UAV), low-powered long-range wireless sensors, IoT gadgets, and advances in robotics, PA concepts and methods shifted toward digitization and contributed more to the economic development and sustainability of food production. By 2012, digital agriculture began to incorporate a wider variety of technological advances such as small-scale robots, swarm drone technology, distributed wireless networks, cloudbased automation, and mobile apps in order to continuously monitor, evaluate, and manage soil condition, water resources, and weather fluctuations on the farmlands to enhance field productivity and reduce operational costs. More recently, digital agriculture developed a series of methods based on artificial intelligence and machine learning to analyze and interpret high-resolution drone NDVI imagery and data (**Figure 2**) for monitoring crop water level and quality, determining soil moisture and soil salinity, creating yield maps, health assessment, and crop stress identification. On the automation side, wireless sensors and IoT devices have been used for smart irrigation, water loss management, and continuous identification of soil nutrient contents in remote areas.

With the introduction of the fifth-generation mobile network (5G), digital agriculture is redefining some of the concepts of the sense-think-act paradigm in the fields. One of the trending topics in this context is the deployment of distributed automation systems such as collaborative robots and a swarm of small-scale unmanned machinery that can autonomously execute various site-specific operations such as weeding and spraying via IoT-based cloud computing services. While similar solutions are being implemented as pilot plant projects or on commercial scales, connection stability and security between nodes have been always a concern. A review of the literature reveals that the use of robots in agriculture with modular electronic control units is growing

#### **Figure 2.**

*Illustration of UAV-based photogrammetry for estimation of crop parameters via nadir and oblique views (Source: AdaptiveAgroTech).*

rapidly and becoming an active field of research, drawing design attention to affordable components that can be easily replaced upon failure. These robots are expected to identify deficiencies and variations in large-scale cultivations and to respond to them with precision technology and site-specific management solutions. For this purpose, autonomous mobile robots that are equipped with various data acquisition devices, multi-spectral cameras, and Light Detection and Ranging (LiDAR) sensors provide a great opportunity for field scouting, health assessment, early disease detection, and yield estimation. In addition, these robots can be integrated with custom-built end-effectors and manipulators to perform specific tasks such as mowing, weeding, and spraying. In addition, mobile robots for digital agriculture are required to withstand harsh field conditions, have a flexible control design with interchangeable and compatible components, and benefit from a reliable navigation system with collision avoidance capabilities. In addition, farmers prefer that depending on the task requirements, different modules such as sensors, actuating devices, and manipulators can be easily swapped on a multi-purpose robot.

Digital agriculture is offering significant potential to replace conventional farming methods with cutting-edge technologies toward creating farms of the future that are expected to be connected and be zero CO2 emissions. If successfully integrated and implemented, digital agriculture can also play a key role in reducing agricultural production costs by decreasing the number of human workforces that are currently engaged in performing repetitive tasks. The presented book aims to expand and highlight these aspects from an academic perspective in separate chapters. Most of the solutions and strategies described in this book represent a valuable aspect of digital agriculture that is aiming at preserving natural resources and securing food production for the increasing world population.

> **Redmond R. Shamshiri** Technische Universität Berlin, Berlin, Germany

**Sanaz Shafian** Virginia Tech, Virginia, United States of America

**1**

Section 1

Perspective

Section 1 Perspective
