Robots and food production


Let’s look in a bit more detail at food, a fundamental requirement for human life. The United Nations predicts that by the year 2050, we will need a 70% increase in food production. Not all of this is due to increase population. A lot of it is due to the requirements of many of the people on our planet to have more food than they’re currently getting and perhaps higher quality food, higher calorific value food. So, a big requirement to increase food production and 70% is a massive increase to be achieved in something like 30 years.

It’s pretty obvious that we need food for people to eat. We also need food for animals to eat. They are part of our food chain. So, a lot of the food that we produce is eaten by animals and in turn, many of us eat those animals.

A somewhat surprising use for food is putting it in cars. There’s quite a movement to turn food product, particularly corn into ethanol and we put the ethanol into the gas tanks of our cars. So, there’s a lot of competition for the food that we produce. Absolutely, we want food for people to eat but there’s also food that’s required to feed animals and there’s also food that’s required to feed cars. 

As our population grows, so too does the size of our cities. The cities get larger; they encroach on productive farmland. They take that land out of food production because we build houses and apartments on it, so that means we have less land on which to grow food. Now, this is a problem because overall we need to grow more food and we’re removing land which is the essential resource when it comes to food production.

Another issue is that farmers as a population are getting older. This statistic is from my country, Australia, and we look at this histogram here and it shows that the average age of farmers is well above 50. There's not young people moving into farming as a profession or as a lifestyle. A very scary statistic in my country, Australia, is that in the next decade, half of the farmers are going to retire. So, who’s going to do the work on the land to produce the food that we need? 

So, one of the opportunities for robotics is to help and solve this food production problem. One application is to convert these large machines which are used in industrial scale food production. Instead of having a human being on board the machine driving it, we turn it into a robotic vehicle so it can drive all by itself, 24 hours a day, seven days a week.

Another thing we can do is to replace human-driven machines which are used to spray crops in the field and we spray the crops to eliminate insect and weed pests. Rather than doing indiscriminate spraying as you see here, we can have much more intelligent, robotic systems which move perhaps more slowly through the crop and use vision sensors which identify insects pests or undesirable plants which we call weeds and apply the appropriate chemical to just where that occurrence is, not indiscriminately, not everywhere. That’s got huge environmental benefits and it’s also got huge cost benefits because we can reduce the amount of chemical that needs to be applied.

The use of chemical herbicides to control weeds has led to an unexpected outcome and that is that many weed species have now evolved a resistance to these standard herbicides. That presents a particular challenge to an agricultural system which has developed over decades and become incredibly reliant on the use of herbicides for weed control. So, perhaps smart robots could identify weeds in the soil and destroy them mechanically as shown here, actually disturb the soil around the weed to destroy the weed. 

Here’s a video of a robot we’ve developed in a research project in my lab here at QUT. This is an early prototype of our weed spraying robot. It’s a very small machine. It’s guided by cameras and other sorts of sensors. It can navigate along crop rows and apply herbicide only where it’s required.

Food. With more people we need to have more of it, but there are real challenges with producing what we will need. Can robots help?

Professor Peter Corke

Professor of Robotic Vision at QUT and Director of the Australian Centre for Robotic Vision (ACRV). Peter is also a Fellow of the IEEE, a senior Fellow of the Higher Education Academy, and on the editorial board of several robotics research journals.

Skill level

This content assumes only general knowledge.

More information...

Rate this lesson


Leave a comment