...Robotics and Artificial Intelligence can change the conversation about the interaction between farming and the environment...
On World Environment Day, I’d like to take a look at the future for farming and the environment. Last year I went to a very interesting public debate on this topic. As is typical with these things, the set up was a deliberately dramatised “farmer vs environmentalist” show.
The two speakers from the “environment” camp laid out articulate and it would seem well-evidenced arguments about ways in which they felt that farming, in its current form, was damaging the environment. To summarise their case: chemicals and fertiliser residues are appearing in places that they shouldn’t be; beneficial insect populations are declining, as are the populations of many other species which use farmland as their main habitats; huge monocultures created by intensive farming are having a negative impact on the biodiversity of our countryside; soil health is suffering.
The finger for all of these ills was pointed firmly at the farmer.
The response of the farmers was to say, to my mind quite reasonably, “what would you have us do?” We are already, the farmers said, doing everything we can to make our applications more accurate. We are using integrated pest management techniques. We are using natural fertilisers. We are building wildlife buffers on our farms and populating these strips with plants that encourage pollinators and act as corridors for animals. We are planting trees. We are reducing our tillage operations wherever possible - and improving soil health is a central part of our farm management approach. After all, the primary job of the farmer is to produce food, and it is our job to produce as much as we can. Populations are increasing but the amount of available farmland is not. Much of the damage is historical, environmental stewardship is taken very seriously and wildlife populations are returning, even sometimes abundant. In short, many of these issues, while still problematic globally, are already being addressed by farmers in the UK. British farming standards are already among the best in the world.
And so the debate went on, with neither side truly happy.
Sitting in the audience as someone who is part of a company developing a new farming system led by Robotics and Artificial Intelligence, I was wary of immediately saying that Robotics and AI are the answer to everything, that new technologies will save us.
Of course it is more complicated than that. Technology on its own will not save us or change the way that farming interacts with the environment. Only human beings thinking differently about how they want the future of food production to look will change this.
Robotics and Artificial Intelligence are tools, and they are only useful tools if we have a completely clear idea of what we want to do.
There are many instances in which we know that new technology is coming, but we don’t really know what we want to do with it. But this is why I am so excited about the potential for an Artificial Intelligence driven farming system, because I think that the way we want farming to interact with the environment is one area in which do know exactly what we want to do.
We want to produce more food, to feed a growing population, but we want to do so in a way that has the minimum possible impact on the environment.
Ultimately, we want the activity of food production to have a positive impact on the environment in which it is produced. And we are working closely with British farmers, on farm, to develop this technology.
An AI driven farming system, using robotics, will change the way that farms interact with the environment in three major ways; firstly an improvement in the way we gather and analyse data; secondly an improvement in the way we use that data to take action; finally a breakdown of the barriers between “food production” and “the environment” with the creation of a farming system that works both ways.
If we can do this, then we can get away from the boring “us vs them” debate that we are currently enduring, and move to a different and more productive conversation. Farmers have been unfairly demonised for years, but now there’s the chance to redress the balance with consumers.
The deployment of ground based data gathering robots in farming systems, such as the Tom robots currently being trialled on UK farms by Small Robot Company, will enable an exponential increase in the amount of data that is gathered on farms. The coupling of this data with an Artificial Intelligence tool, such as SRC’s Wilma system, will enable an exponential increase in the analysis and understanding we have of that data.
In time, Tom will gather data on every single plant growing in the field, with wheat being our first target crop, and he will also gather data on every single square metre of soil in the field. Wilma will analyse the data from that farm, and add onto it firstly the data gathered on other farms and then secondly all of the available public research anywhere in the world on that particular soil type, that particular variety of crop, the weather patterns, the market patterns and so on.
Where this is particularly relevant to the environmental question is that it will create a huge increase in our understanding of how the crop, the soil and the agricultural inputs are interacting, and this will bring a subsequent increase in the understanding of how farming is having an impact, which will drive improvements to best practice.
Small Robot Company is aiming to create a learning network of robots, so that when something is learnt on one farm, all of the farms in the network get the benefit of that learning.
As an example of direct data gathering on environmental factors, Small Robot Company has already been able to detect and count individual entrances to the nests of ground nesting bees on one of our trial farms. Ultimately our robots will be able to count and map this data across all of our farms. That data, coupled with other similar observations could see dramatic improvements in our understanding of the activity of different species on our farms.
Although Small Robot Company has ambitions to go much, much further than just deploying Tom and Wilma, the collection and analysis of this data on its own has the potential to create significant improvements in the way that existing farm machinery is used, lessening the impact of farming on the environment.
Where things start to get really interesting, however, is when we see the deployment of new hardware on farms, such as Dick and Harry.
Gathering data at the individual plant level is one thing, but being able to take action at the same level across whole farms is when we will truly be able to claim that we have created the Digital Farm. This is what SRC is working towards with the development of Dick, our non-chemical weeding and micro-spraying robot, and Harry, our digital planting robot.
One of the reasons that the “Environmentalist vs Farmer” debate is unfair on the farmer is that the machinery that farmers currently use to take action in field is an enormous constraint on their ability to be accurate.
Farm machinery in its current form has not been designed to be as accurate as possible; it has been designed to be as fast as possible, to limit the amount of time that a human being is operating it. While farmers take into account a wide range of factors to ensure sprays are applied as accurately as possible (including wind to prevent drift, and dry conditions to ensure the chemicals stay where they should), and tractor-based precision systems are developing, there’s a limit to what can be achieved when you are treating the whole field, rather than the individual plant.
Automated, smart machines will be designed with accuracy as the uppermost engineering consideration.
Farming needs new hardware in order for it to move to the next level, increasing productivity whilst reducing inputs. Using robotics and artificial intelligence brings a new level of precision which is beyond the hard limitations of tractors. We will be able to treat each plant individually, giving them exactly what they need, with minimal waste. This can reduce chemical usage in arable farming by up to 95%.
3. BREAKING DOWN THE BARRIERS
Ultimately, the Digital Farm is about more than just increasing the productivity and profitability of farming whilst reducing its impact on the environment. It is about redesigning our countryside so that it truly does meet the needs of everyone.
At the moment, the food production happens in one area of the farm, in which we have big, square fields containing very clean monocultures. Very few pests or weeds, highly efficient production facilities.I’m not talking here about farming on the prairies of North America, which is on another level again; however, even in the UK the current farming system dictates that the crop remains in the centre of the field, with environmental features on the margins. We do this because it is the most efficient way to use the machinery.
So off to the side of the field we have areas which are more diverse such as field margins and woodlands. There is lots of good work happening in these areas, and some farmers are fantastic managers of richly biodiverse farmland.
But why do these two areas have to be separate?
The final evaluation of the success of the Digital Farm will be in our ability to move away from a monoculture system towards a system in which we have multiple different species of commercial crop growing in the same field, possibly alongside other plants which are not commercial crops but which are having a positive impact in some way. Ultimately, we are talking about employing permaculture techniques at scale.
Artificial Intelligence systems such as Wilma will help us to rethink our crop rotations by creating a plan for the best range of species to optimise the balance between commercial productivity, soil health and biodiversity. Not choosing one over the other as we often forced to do now, but optimising the balance.
With digital direct drilling, we can more easily practice strip tilling. We can plant thin strips of wildlife-friendly species at regular intervals, such as wildflowers for beneficial pollinators, but at no detriment to production.
Ultimately we will be able to harvest a crop on a plant by plant basis which will enable us to move away from the combine harvester. This accomplishment is still some years away, but this is the bold vision which farming should not shy away from embracing.
Ecologically harmonious, efficient and profitable farming. The best of both worlds. A fresh start for our relationship with consumers. Redressing the balance, and paving the way for a new future.