The use of artificial intelligence (AI) in Irish agriculture provides a “huge level of opportunities”, and farmers should learn about the technologies and techniques that are available.
This is according to the assistant lecturer at the Department of Computing Science and Mathematics at Dundalk Institute of Technology (DkIT), Dr. Abhishek Kaushik.
Agriland spoke to Dr. Kaushik about the adoption of AI technologies on Irish farms, and why he believes “they will be the future” despite some “major challenges” ahead.
Artificial intelligence on a farm
The potential of AI in agriculture continues to grow as technology advances, providing valuable solutions and improving efficiency, the Department of Agriculture, Food and the Marine (DAFM) said.
“AI enables precision farming by analysing data from sensors, satellites, and drones, leading to optimised resource usage, increased crop yield, and reduced environmental impact.
“The use of AI in agriculture can also assist in automating tasks like harvesting and sorting, enhancing efficiency and reducing labour costs,” DAFM said.
Irish farmers
Dr. Kaushik’s current research focuses on the adaption of AI use in agriculture, having previously looked at the factors that influence Irish farmers who are not adopting smart technologies.
A lack of understanding and awareness of AI, as well as the absence of available funds, in particular, are reasons why farmers would be less likely to adopt AI technologies, he said.
However, a lot of Irish farmers, especially young farmers, are interested in adopting the technology for farm management, climate predictions, and disease management, he added.
While these farmers are more inclined to use smart technology software, which is cheaper, large farms that have available funds make use of hardware technologies, Dr. Kaushik said.
This hardware, he explained, includes drones, automatic milking, and radio frequency identifiers for maintaining animals and vegetables by using the internet of things (IoT) sensors.
Farmers who are unwilling to adopt technologies using AI on their farm always have concerns about the use of their data and data protection policy, Dr. Kaushik told Agriland.
There is simple software, for example farm management, which only needs basic information about a farmer’s land or crops to help in decision making, organisational or herd management.
However, software that incorporates machine learning and deep learning, which are both AI techniques, may need a “significant amount” of data, for example regarding crop behaviour.
Dialogue systems using a chatbot, such as ChatGPT, are a “great tool” to help farmers with their queries and are very easy to use, according to Dr. Kaushik.
Regarding any new technologies in agriculture, he said that farmers base their decision to adopt or not adopt on whether the tool is useful and whether the tool can be used easily.
Some tools can translate audio queries into text which then goes into the system and comes up with different answers. For example in relation to plant diseases or different crop types, he said.
Click on the video below to see a few examples of how ChatGPT could be used:
However, chatbots could also be used for more specific requests, Dr. Kaushik explained, as they can help automate processes in terms of reminders, or a farmer’s business schedule.
Discussion groups bringing together an expert and young and mature famers can provide a good platform for knowledge sharing of different technologies and the use of AI, he said.
Artificial intelligence and data protection
The assistant lecturer and researcher believes that a lot of farmers are not aware of General Data Protection Regulation (GDPR) policy and how it empowers them.
Under GDPR, the individual has the right of access, rectification, the right to object and “to be forgotten”, among others, according to the Department of Enterprise, Trade and Employment.
Researchers should work with farmers and build their confidence in and awareness of data mechanisms, data storage, and data policy. “Their concerns should be addressed,” he said.
“Unless farmers are confident, it is very difficult to get them to adopt new technologies.
“What are the positives and negatives of storing data, what is data, what are the concerns? That should be addressed in order for them to adopt new techniques,” he said.
“Ireland has a lot of good potential and if we apply those new techniques, then we can have better results, better agriculture and better production for the whole world,” Dr. Kaushik said.
In Ireland’s national AI strategy ‘AI – Here for Good’, the government states that the more people who understand and trust in the potential of AI, the more who will embrace it.
AI refers to machine-based systems, with varying levels of autonomy, that can, for a given set of human-defined objectives, make predictions, recommendations or decisions using data.
The use of AI brings risks such as opaque decision-making, discrimination, bias, privacy issues, or use for criminal purposes, and therefore requires a robust governance framework, the report states.
AI-based systems are highly data-dependent, thus a supportive and trustworthy data infrastructure is paramount to unlock AI’s full economic and societal potential, according to the government.
DAFM
The DAFM recognises that AI can be a “powerful tool” to improve sustainability and productivity, and is funding research that features the application of AI in agriculture and food.
A research project led by Munster Technological University (MTU), together with institutions in Italy and Germany aims to develop decision support systems for precision farming.
Under the “App4Farm” project, sensor data and machine learning models will be integrated to improve nitrogen management in agriculture systems, the DAFM said.
The European project “SpectroFood” plans to utilise AI and other technologies to develop digital solutions to accelerate the steps towards a resilient and sustainable food supply chain.
Through high-quality systems monitoring, the technology can lead to production increases, waste minimisation, and product labelling in terms of quality, safety, authenticity, and standards compliance.
While there are currently no specific Knowledge Transfer (KT) groups in AI, the DAFM told Agriland that AI could feature under ag-tech discussed in KT groups in the future.