Visual Services-East Moline

Interview: How John Deere uses connectivity to make farms more efficient

After pioneering the use of automated vehicle technology 15 years ago, farm equipment supplier John Deere is now going further, with its sights set on efficiency and sustainability, says sales VP Christoph Wigger

With Google cars out and about on the streets of Mountain View, California, and driverless pods to be tested in Greenwich, London, the automated vehicle has rapidly become a poster child for the internet of things (IoT), as both a key early use case and a key concern around public safety.

But for Christoph Wigger, vice-president of sales and marketing for Europe, the CIS, North Africa and the Middle East in farming equipment supplier John Deere’s agriculture and turf division, driverless technology is merely the latest stage in a trend towards automation that has been going on for some time.

It was back in 2001 that John Deere brought its AutoTrac GPS-based automated guidance system to market, and the firm claims the technology is now in use in more than 200,000 vehicles in the field around the world, mostly in advanced markets in North America, Western Europe and Australia, but increasingly in other big agricultural countries such as Brazil, China, Russia and Ukraine.

“That is the ‘Aha!’ moment I want to create,” says Wigger, as he sits down with Computer Weekly at London’s IoT Tech Expo fair. “People love to talk about automation in vehicles, but automation is already a daily practice for us.”

In fact, the IoT is rapidly emerging as a huge area of interest in agriculture. Last year, a report from sector monitor Beecham Research said the IoT could be key to the 70% increase in food production that will be needed to feed a global population that could hit 10 billion within the next 35 years.

In its Towards Smart Farming report, Beecham’s Therese Cory said the farming industry would be quick to embrace precision agriculture with assistance from the IoT, using sensor networks and automation to improve productivity and reduce waste in both crop and livestock farming.

At the core of John Deere’s AutoTrac sits a satellite receiver, the StarFire 3000, which can acquire signals from 56 satellites – belonging to both the GPS and Glonass systems – around the world.

It offers a range of accuracy levels at different price points, calculating a machine’s position down to 23cm for the basic signal, down to 5cm for the mid-range option, and down to 2.5cm for those who choose to use the more expensive, but far more accurate real-time kinematic (RTK) navigation system, which comprises a base station on the ground using a combination of mobile or radio signals.

Exact position

But what are the immediate benefits for a farmer to know his or her exact position down to 2.5cm? For one thing, it makes farming cheaper, says Wigger – guidance systems can save farmers 10% of their costs straight away.

“When you drive a tractor towing an implement through a field, you can never guide it so precisely that you don’t overlap on each pass,” says Wigger. “For example, when you are seeding, you want to seed very precisely, but this is hard when you are towing an implement behind you that can be anywhere between 12 and 20 metres wide.

“That GPS device will allow you to do very precise, side-by-side seeding, tillage and all that. That’s a proven number in the field, you save 10% of your costs, and that’s a huge saving you have enabled through the guidance system, so it’s very much about efficiency gains.”

With mechanisation a key cost for farmers – a good tractor can cost upwards of £200,000, says Wigger – a 10% saving for a cereal farmer can easily add up to £50-100 per hectare.

Read more about IT in farming

  • .Vodafone’s Connected Farming report concludes that simple mobile services could boost farm-gate incomes by billions of dollars around the world.
  • The Flint River Partnership is testing technology that analyses a variety of data to generate localised weather forecasts for farmers in the US state of Georgia.

City-dwellers may like to think of the countryside as a bucolic idyll, but the realities of modern farming can be harsher than in any office. According to Farmers Weekly, suicide rates in farming may be three times as high as for other occupations – and as farms get bigger and margins on crops slimmer, farmers have to work longer hours under greater pressure just to keep food on their own table. AutoTrac can help them do this, says Wigger.

But will full automation come without supervision, with tractors and combine harvesters moving around the farm on their own? It may do eventually, says Wigger, but a lot will depend on regulation, and the drive towards connected vehicles for consumer use will go some way towards establishing policy in this area.

However, the farming industry cannot take all its cues from road-going vehicles: it also has some specific safety concerns. For example, because farm machines are generally moving around a wide area without roads, they may be a little more likely to encounter wandering humans or misplaced livestock than a Google car might come across on a California freeway.

The safety mechanisms built into fully autonomous farm vehicles will therefore need to be of a different nature to those in autonomous cars, adding a new dimension to future regulatory frameworks.

Optimised for efficiency

Guiding farm vehicles autonomously through fields would be a big deal for John Deere and its customers if that was as far as it went, but the agricultural industry is also realising a number of other benefits around process optimisation.

For example, on an average commercial farm, a combine harvester is generally used for only two to three weeks a year, which leaves a lot of time for its driver to unlearn how to operate it optimally. Assuming the size and shape of the field has not changed since the previous harvest, GPS data can be used to fine-tune the harvester so it can follow the same paths that it did a year ago.

“Because you can connect the fleet, you can optimise speed, you can optimise fuel consumption, you can literally make an opaque thing transparent”

Christoph Wigger, John Deere

This also enables the combine to run more efficiently. “A combine is typically driven at five kilometres per hour through the field, but it is designed for seven or eight,” says Wigger. “It’s very human not to want to push the vehicle too hard, but it means the average combine driver maybe uses 60% of the installed capacity. Machine optimisation allows them to move from 60% to 80%.

“The second thing you see is that big machines often have a system of machines around them. Think of a combine and 10 tractor-trailers that the harvester is loading one after the other. When the vehicles are all connected, the combine can communicate with them to optimise their movements.

“Sometimes the tractor driver may want to push harder, to drive faster to get back to the field, but now he doesn’t have to, because the system knows there are still other tractors waiting.

“Because you can connect the fleet, you can optimise speed, you can optimise fuel consumption, you can literally make an opaque thing transparent.”

The internet of manure

The IoT is already bringing new efficiencies to the farming industry through fleet management, but it also has a role to play in sustainability. By its very nature, commercial farming of vast fields of single-strain crops, sprayed with large amounts of chemicals, is neither environmentally friendly nor sustainable. Something needs to change, says Wigger, and the European Union is taking a lead here.

“Under the Common Agricultural Policy (CAP), farmers are already paid more and more to farm sustainably, so that, for example, you only fertilise as much, or give as much nutrition back to the land as you have taken off it,” says Wigger.

“You can use sensors to, first, measure how much you did take off the land, and then the computer can compute how much nitrogen, phosphate and potassium you have to give back to the land to have equilibrium.”

John Deere is now also offering manure constituent sensing. Manure is highly in demand and well-used in farming as a natural, organic fertiliser, but variations in its make-up and consistency mean it can deliver highly variable results.

Building big vehicles is yesterday’s news; building intelligent vehicles is tomorrow’s
Christoph Wigger, John Deere

By mounting sensors on the applicator that can deliver 4,000 readings per second, farmers can now measure nitrogen, phosphorous, potassium, ammonium, dry matter and volume accurately, while spraying their fields.

“When the manure is very light, the sensors will guide the speed of the tractor and decelerate it so that it gives more. When it is heavier, it speeds up,” says Wigger.

This means users can achieve site-specific nutrient target levels on the go, while saving money on bought-in mineral fertilisers.

The other key use case for sensor networks in farming is to detect the presence of environmentally-protected areas on a farmer’s land, such as watercourses that must not be polluted with chemicals.

“When you have a spray boom, the GPS picture shows the sprayer exactly where it is and automatically shuts off some of the nozzles so that you stay back to protect the water,” says Wigger.

Huge potential for IoT

Looking further into the future, Wigger says the agricultural industry is still only beginning to explore using the IoT in precision farming, but in Europe, there is huge potential to embrace and utilise this technology, and to drive its future development.

He also calls for the industry to do more to motivate young people to move into connected vehicle and device engineering.

“Building big vehicles is yesterday’s news; building intelligent vehicles is tomorrow’s,” he says. “Because we are limited by vehicle length, height, width and weight on the roads, you can only innovate by making vehicles more intelligent.”

Given the long-held fascination and affection that many small children hold for huge tractors, encouraging them to take an interest may not be the toughest ask in the world.

Read more on Internet of Things (IoT)