Digital Farming and Robotics 2021:

Progress in R&D and commercialisation in the Crop Science sector

Access your free sample report

required fields

Our Digital Farming and Robotics 2021 report has now been published, read on for more information about the report, and fill out the form to access your free sample report.

Chapter 1: Introduction

This report provides an overview and update on the 'software' and 'hardware' aspects of precision agriculture, i.e. 'digital farming' and 'robotics' (and general automation) being researched, developed, commercialised and used in practice in the Crop Science sector.

Precision Agriculture is a management strategy that gathers, processes and analyses temporal, spatial and individual data and combines them with other information to support management decisions according to estimated variability for improved resource use efficiency, productivity, quality, profitability and sustainability of agricultural production.

Innovations have moved Precision Agriculture to Digital Farming, which involves technologies such as artificial intelligence, deep learning, drones and other robots, and has increasing emphasis on crop protection.

Although very challenging to estimate, the 2020 value of the global digital farming market has been put at $5-7 billion and with CAGRs reckoned to be 11-13%. It could be worth $15 billion by around 2027.

In the US, surveys show the rapid uptake of a wide variety of precision agriculture technologies and digital farming services.

In Europe, potential savings in pesticide use of up to 80% are being emphasised as a key benefit from adopting precision agriculture. In order to realise such benefits, farmers would need to make significant investments in training and equipment. Other issues relate to the ownership and protection of farm data that are shared in digital farming systems; and the public perception of farming becoming more industrialised.

Optimised crop protection requires detailed data on the occurrence and distribution of weeds, pests and diseases at the earliest possible stage of detection. Interest in the use of drones in scouting for weed, disease and pest problems and application of crop protection products where permitted is increasing rapidly.

Information is provided concerning associations and forums devoted to precision agriculture; scientific journals; and online magazines and other websites.

Chapter 2: Key enabling technologies

This chapter overviews the wide variety of technologies that enable the current state of development and application of digital farming and robotics in crop production.

Satellite technology has revolutionised navigation. Although GPS has become the generic term, there are, in fact, several global navigation satellite systems (GNSS).

Remote sensing is the detection and/or identification of an object(s) or landscape without direct contact. Proximal sensing occurs near ground level.

Imaging and spectrometric sensors capture light at various wavelengths (visible, near infra-red, UV, etc). Other sensors include LiDAR and RADAR, which are not impeded by clouds, and chlorophyll fluorescence.

Data acquired by remote sensing can be used to create imagery. Maps of features changing across a field can indicate aspects of the condition of soil or vegetation that can be used for site-specific management and variable rate application of inputs.

Big Data has been defined as data of greater variety arriving in increasing volumes and with ever-higher velocity, which has potential value if it can be verified.

Artificial intelligence is a branch of computer sciences that enables the development of intelligent machines, thinking and working like humans, e.g. speech recognition, problem-solving, learning and planning.

Machine learning and deep learning are routes to artificial intelligence for use in autonomous decision-making machines. Neural networks are often used to recognise underlying relationships or patterns in data sets like images through a process that mimics the way the human brain operates, i.e. by repeatedly clustering and classifying.

The Internet of Things (IoT) connects numerous diverse devices to collect and collate to ultimately guide decision-making.

Blockchain technology aims to achieve database consistency and integrity in the context of a distributed decentralised database. In practice, it allows transparency in complex supply chains to generate trust.

Chapter 3: Players in digital farming

This chapter concentrates on the activities in the digital farming space of the crop science majors and other agrochemical companies: their specialist collaborators; and other companies engaged in activities with some connection to crop protection and production.

BASF has stated that its strategy includes leveraging digital technologies to drive business growth, promoting innovations across chemicals, artificial intelligence, internet of things, robotics and other emerging technologies. Its digital farming platform Xarvio was acquired from Bayer CropScience in 2018. BASF has been collaborating with Bosch on a smart spraying project.

Bayer CropScience has described Digital Farming as bringing 'increased precision to crop production by supporting key farm management decisions with science-driven insights.' Bayer acquired The Climate Corporation subsidiary and its digital farming platform FieldView as part of the acquisition of Monsanto.

Corteva introduced a new digital agriculture platform, Granular Insights, in spring 2019. A 100 million acre (40 million ha) opportunity for the service was targeted by 2023. Starting in the US and Canada, the service has expanded to other countries including Australia and Brazil. Corteva has been particularly active in developing the use of drones for crop monitoring.

FMC launched its digital farming platform, Arc, in 2020. Arc is particularly focused on predictive analysis of potential future pest problems. The company is developing commercial drone spraying of its leading insecticide chlorantraniliprole in SE Asia.

Syngenta Group announced that it had achieved the milestone of 125 million acres (50.6 million ha) of farmland actively managed by growers using the company's AgriEdge and other digital services. These include advanced scouting, seeds selection, imagery decision support, weather risk management, financial planning, crop and farm operations management. The company is developing commercial drone spraying of rice in SE Asia.

The digital farming activities of other crop protection companies including Adama, Amvac, Land O'Lakes, Nufarm, Sumitomo and UPL are noted.

AGCO, CNH Industrial and Deere and Company are leading agricultural machinery companies with digital farming platforms and precision agriculture technologies.

Satellite and imagery companies working in crop protection are profiled along with their latest news.

Sensor and diagnostic specialist companies are also profiled along with their latest news. Digital farming and AI specialist companies are likewise profiled along with their latest news. Companies specialising in drones and ground robots are similarly profiled in those respective chapters.

Chapter 4: Drones

Drones (UAVs) can be used to map and monitor fields through remote sensing and to apply crop protection and other inputs. The value of global drone sales overall was forecast to be over $12 billion by 2021.

Most drones for agricultural use in Europe and the Americas are medium-sized multi-rotors used for surveillance and analysis. Compared to satellites they offer lower cost for smaller fields, more flexibility, more reliability in not being impeded by cloud cover and higher resolution imagery.

Larger drones are now in use, which carry a payload such as seeds or spray solution. In China, Japan and South Korea, spraying drones have been widely used for many years. Application of crop protection products by drone, albeit under strict regulations, is now permitted in the US, Australia and South Africa.

Regulations control the flying of drones per se and their use for crop spraying. Specific crop protection products must also be approved for aerial application. Spray drift and risks to bystanders and the environment are issues; also, who bears responsibility for any accidents?

The downdraught airstream of drone rotors pushes the crop canopy open and facilitates the distribution of spray droplets through the canopy to reach the base of the plant and underside of leaves. This should increase the efficacy of control and reduce spray drift.

There may be potential for low carrier volume electrostatic spraying to be done by drones and there is already some commercial use.

Profiles and activities of a selection of some of the most prominent drone companies operating in agriculture are described.

Chapter 5: Smart spraying

Sensors and deep learning are being used to enable 'Smart Spraying' as a progression from the concept of variable rate application of crop protection products.

Reducing the amount of pesticide applied should benefit profitability, timeliness of application and the environment.

Activities of the Crop Science Majors are described. These often involve collaborations with machinery and specialist technology companies to develop smart spraying systems.

Activities of the Machinery Majors in developing smart sprayers are covered.

Smaller companies with specialist technologies in sprayer design, sensors and imagery, and communications are profiled.

Chapter 6: Ground robots and other technologies

This chapter focuses on the technologies being deployed and the progress made by the industry from research to commercialisation of ground robots working in fields. The application of ground robots in crop protection is largely directed towards weed control by mechanical or other physical methods and spraying.

Key emerging technologies in robotics include robot vision and machine learning. Standardisation of data is needed to ease the exchange between devices, software systems and the various stakeholders.

There is a trend towards smaller robots. which could have benefits including the ability to quickly take advantage of windows of favourable weather; to work in soil conditions precluding the use of heavy machinery; and to work continuously in 'swarms' in a coordinated way, day and night.

A common approach to 'training' robots is by 'digitising' numerous crops under different circumstances, or analysing tens of thousands of images representing the targets, e.g. crops and weeds. The information is then automatically analysed by neural networks and then used to programme robots to deal with each plant and each patch of soil according to the 'deep learning'.

Some collaborative projects are described. Prominent companies involved in robotics and automation are also profiled.

About the Author

Alan Baylis is an independent consultant with more than 40 years' experience in world agriculture: from running a UK arable farm to international R&D management in Syngenta and previously in Zeneca Agrochemicals and ICI Agrochemicals. An agronomist and crop physiologist, he has specialised in many aspects of crop production and protection in global cropping systems. His career has covered the discovery process for all crop protection products, through glasshouse and worldwide field-testing to technical marketing. He was Chair of the Society of Chemical Industry (SCI) Board of Trustees (2015-2021) and is a past-Chair of SCI Agrisciences Group. He has BSc and PhD degrees from the University of Leeds and an MBA from Henley Business School, and is the author of many peer-reviewed publications, conference papers and other articles.

{"items" : [ {"name":"share","enabled":true,"desc":"<strong>Share</strong>","mobdesc":"Share","options":[ {"name":"facebook","url":"","enabled":true},{"name":"twitter","url":"","enabled":true},{"name":"linkedin","url":"","enabled":true},{"name":"email","url":"?subject=Digital Farming and Robotics 2021 | IHS Markit&","enabled":true},{"name":"whatsapp","url":"","enabled":true}]}, {"name":"rtt","enabled":true,"mobdesc":"Top"} ]}