Growing foods with AI – An Overview

Photo by Pixabay on

Artificial Intelligence(AI) is one of the terms frequently heard these days. It is the ability of computers to perform tasks that generally require human intervention, like driving a car. Anything from self-driving cars and smartphones to most electronic devices nowadays is equipped with some form of AI. Nowadays, AI is being applied in the field of agriculture, which will only seem obvious in the future.

Growing food requires experience in the field, human intuition, complex analysis, and most of all, environment variables like water, temperature, and light in the right amount. This can be achieved efficiently with appropriate technology. Also with technology comes the concept of indoor agriculture and in other words Controlled Environment Agriculture (CEA). Here, factors related to the growth of food can be controlled within a closed environment, so that it mimics the natural environment, just without soil. If a tomato plant grows at say, 23°C, the same temperature is maintained in a controlled environment, along with other variables. However, it is widely accepted that the quality of food grown in such farms is not like the one grown in an open field, in terms of taste. Here, AI comes in.

In an open field or indoor farms, Artificial Intelligence can be used to analyze data provided by sensors and other IoT devices, to derive the real-time conditions of the crops. These conditions include physiological and pathological conditions, which are most important while producing food. With the development of IoT (Internet of Things), where electronic devices are interconnected through the internet, data regarding the environmental variables required to grow the best quality of food will be obtained from different parts of the world. Along with these, AI can be used to figure out and relate these conditions with the production data. These data can be processed with AI models trained to figure out the combination of the environment variables best for the food production, dubbed by many as “climate recipe”.

Suppose Japanese strawberries taste different and are tastier than strawberries grown elsewhere. But it requires a set of environmental variables peculiar to the climate of Japan, call them climate recipe for Japanese Strawberries. Now, you want to grow them in another country but want them to taste like those of Japanese ones. What you do here is grow strawberries in a confined space, control the environment around it exactly as per the climate recipe, and make those strawberries feel like they’re in Japan. Voila, you’ve grown strawberries that taste like they’re straight from Japan.

Controlling the environment directly affects the physiology of plants in many ways, from the production of chemicals relating to the taste and aroma, color, and in some cases, chemicals as a part of a defense mechanism. Plants in some cases, require certain stress for the release of phytochemicals, which can be induced using climate recipes and the control of the environment.

The MIT Media Lab Open Agriculture Initiative (OpenAg) has developed “Personal Food Computers”. These computers use open-source data including climate recipes for different plants and combine it with onboard sensors, and grow plants within confined space. Likewise, another company Iron Ox uses robots, equipped with computer vision, a form of AI, to grow foods hydroponically.

Machine Learning, Computer Vision, and other forms of AI can be used in indoor farms, to monitor and grow food with the minimum human intervention. The concept of AI might seem far-fetched at times, but with the increasing population, urbanization, and lesser lands, we need an efficient food production system. This can be obtained only with the intervention of Artificial Intelligence.

We’re far off the future as predicted by Sci-Fi, where robots take over the world, but it’s likely that AI will take over the food production system to feed the global population in near future.

Leave a Reply

Your email address will not be published. Required fields are marked *