We’ve been interested in the field of agricultural robots for some time now (see what we did there?), so when we brainstormed who best to consult on the matter, there was only one person for the job: Simon Blackmore. Professor Simon Blackmore of Harper Adams University is the utmost authority on the design and build of robots in agriculture.
This is part two of our interview with the expert on all things agriculture, Simon Blackmore. You can catch the first part here.
Tharsus: You’re currently heading up the Engineering Department at the UK’s leading academic institute for agriculture – Harper Adams University. What great robotic research and development projects can you talk about that are poised to enable the safe, efficient and economical production of food?
Simon: This has been my full-time task – my life’s work; to try and develop research projects that support the safe, efficient and economical production of food. It’s an exciting time here at the moment as my aim over the next couple of years is to commercialise this work. We need to stop talking about these systems and move them out of our labs and into the real world! We can definitely do it in this country, the work that I’ve done here at Harper Adams has made me one of the leaders in the world in terms of creating the vision, thought and development of a ubiquitous robotic agricultural system – one which I intend to bring to market in the very near future.
I’ve got a friend in CA that’s received $32M in joint funding from a VC and Sagenta to develop Blue River Technology. If you’ve got that sort of money, you can just make it happen – just push your product straight through development and onto the field, no mucking about with being drip fed funding, it’s very hard to make an impact that way. What I’m really looking for now is a group of people who really share our vision so that we can make these robotic systems. There’s no technological reason why not. We all know very well what the capabilities are in terms of the technology. We’ve just got to make sure we can get the investment to be able to come up with a machine that is economically viable.
Tharsus: So which sector is going to accept robots first?
Simon: This is an easy one for me to answer – it’s going to be row crops. So, high value produce. We’re currently working with a company called Geez who supply 60% of all fresh vegetables consumed by the general public in Great Britain. Automation is obviously big business for them, so we’re looking to come up with a machine that can help them be more efficient and sustainable.
Tharsus: Within the robotics industry, what specific state-of-the-art technologies or scientific underpinnings are going to be instrumental to the applied production of broadacre and horticultural crops in the future?
Simon: Well there’s two things: one is that it has to do what it says on the tin. So it has to do the job, but what is required is not the technology as such but the understanding of human behaviour. It’s usually the people that are the sticky parts of the system, you’ve got to get the robots to do what the people are expecting them to do. And that’s incredibly difficult. It’s because of this, I developed SAFAR – the System Architecture for Agricultural Robots. So far I’ve only been able to develop half of SAFAR but the fundamental premise of the system is to enable the base machines to work in a way that you and I would. So when you go to the garden and you’re weeding – you don’t plough up and cultivate the whole garden to kill one weed. You go in there and you recognise the anomaly and pull it out, or deal with it. It’s that same anthropometric approach that robots need to be able to take.
It’s this side of things that still needs to be developed. We’re only ever going to learn these lessons when we’re actually out in the field doing these tasks. So we’ve got to go and get these machines in the field now – working to be able to learn. But that’s only software, once the machines are out there running, the behaviour can be modified very easily in software.
Tharsus: The size of total UK labour force in agriculture has remained static for the past 5 years. Numbers of seasonal, casual and gang labour has remained fairly consistent at 67 thousand employees over this period – most of whom are migrant labour. A simple calculation suggests this is costing UK farmers approx. £231m if we assume a 3 month season. With the fallout from Brexit and the reverse osmosis of cheap migrant labour, do you think the government and private investors will have to support investment in robotic technology in response to this political move?
Simon: Yes, absolutely. I had Andrew Legland from VTB Capital here a couple of weeks ago to showcase the various technologies that we’ve got. You’re also probably aware that we’ve just set up the Agri-Epicenter now between SRUC in Scotland, Cranfield and ourselves. The Epicenter is a big new building here on campus which the government has put £18M into as capital. If we put forward a strong enough business case to Government they will continue to support this industry in developing exciting new projects and commercialising these high-technology products.
We were told that there probably wasn’t going to be anything in the Spring Statement but we’re certainly looking forward to the Autumn Statement where we’re expecting a new fund to be made available for these type of developments.
Tharsus: The UK has proven ability to develop, through commercial partnerships, new products and solutions for the food and farming market. For organisations that have something to offer in solving the challenges we’ve discussed, what mechanisms are in place that support the creation of commercial partnerships fostering new robotic services and business models?
Simon: It’s probably important to make a clear distinction to our readers who are not currently servicing the Agricultural market, but are instead maybe looking to support in the future. Precision Agriculture is the cultivation of broadacre arable crops with 19% of UK land used for this type of farming – typically the produce is low in value and processed using large format, heavy and costly tractors and peripheral machinery. Contrary to popular belief these machines are already fairly autonomous in capability, equipped with auto-steer and capable of automated function – variable rate dispensing being one.
Alongside Precision Agriculture, livestock farming, and horticulture crops pose two other areas that could be made more sustainable through the adoption of robotic technology.
Horticultural crops take up 1% of the UK’s green and pleasant land – and are typically high-value crops such as fruit, vegetables and glasshouse produce. It’s the delicate fruit and vegetables and those produced on perennial trees or shrubs with variable ripening times which require human manual harvesting. The specially constructed environments that house these produce lend themselves exceptionally well to the introduction of robotics opposed to arable land cultivation which has a multitude of variables.
Tharsus: Groups of robots cooperating with each other to achieve a well-defined objective is an emerging concept for autonomous systems in daily agricultural tasks. Kit Franklin – one of the researchers working on the ‘Hands-free Hectare’ project- recently commented in an interview with Farmers Weekly that: “Automation is the future of farming. We’re currently at a stage where farm machinery has got to unsustainable sizes”. Has there been any work done to illustrate that the deployment of smaller autonomous vehicles will reduce the carbon footprint left by farm machinery? And what economic benefits will these types of machines provide?
Simon: CO2 and sustainability is not an area of expertise for me. But as an engineer I do understand efficiency. When we look at the crop production system. It’s very inefficient. We currently use a lot of energy, and it’s not green. The energy that is consumed costs an incredible amount of money, that, if applied off-target is not only a waste of resources, but is also a harsh pollutant to the atmosphere. These are the factors that affect CO2 and drive sustainability.
If we can then come up with a new system that minimises the amount of energy that we use, in all of its different forms, then the efficiency gains become very clear. Therefore, as a second order, sustainability and carbon footprint will be drastically reduced. That’s how I look at it.
Tharsus: According to the “Agriculture in the UK” report, total agricultural land usage in the UK hasn’t grown in the past 5 years (18.5M Hectares). What does this mean for the role of automated vertical and hydroponic farms to increase potential output?
Simon: In my opinion, this is a great example of technology looking for an opportunity. So, vertical farming – they’ve deployed it effectively in Japan, in the middle of Tokyo. It works there because the cost of transporting lettuce from farms into central Tokyo takes at least 3 hours. If you’ve got a 3-hour drive carrying a truck full of lettuces it makes economical and commercial sense to grow the lettuces in hydroponic tanks in the back of supermarkets. So there are some reasons for doing it.
Coming back to your first question, how are we going to feed the planet, where are we going to get this extra yield from? It’s going to come from robots. It’s not going to come from the big farms and fields – these environments are already very efficient because of their economies of scale, but we’ve got thousands of hectares in the UK made up of small fields and farms that cannot take advantage of these same cost advantages. If robots can go into the small fields and get to the same level of efficiency as the big farms – not through economies of scale but instead through what I call “intelligently targeted implements” – we can then make a lot of this marginal land sustainably intensified into producing more.
I think small machines running around on family farms and smaller fields will be able to increase production and yield. We will not bring any new land into production in the UK, we don’t want to do that now, that’s the natural area, but lower productive lands can be made more productive by the use of simple technologies.
Tharsus: Finally, Simon, over the following three phases what robotic applications do you think we’ll start to see emerge in the agricultural sector? Is it going to be autonomous tractors equipped with robotic implements farming broad-acre farms or swarms of smaller autonomous machines supporting the cultivation of horticultural produce that we will see widespread adoption in first?
Simon: Figuratively speaking, if we were to look across the hedgerow and into the next field – in the livestock sector – robot milking is now mainstream and it has been for a while. It’s way more difficult to get a machine to interact with a cow than it is to get it to interact with a crop. But this is an area that has fully accepted high technology. People are still amazed by robots milking cows, but it is very much a mainstream activity that has been happening for years!
We do not have an equivalent in the cropping area. The longer this goes on the more intense the vacuum becomes, increasing the need for these types of robots. I think that if robots meet the criteria that we’ve already discussed and we can build something that can deliver on those three factors of embedded economics, embedded knowledge, and embedded psychology, we can make farming sustainable, economical and commercially prosperous in the UK. I don’t see any reason why we can’t just get on with it and build these things now! It feels like we’ve spent too many years talking about it.