Grains Research and Development

Date: 07.09.2012

Broadband sets up the wi-fi farm

Author: Sarah Clarry
Smart farming illustration.

When the northern NSW city of Armidale was chosen as a first release site for the National Broadband Network (NBN) there was much talk about how it would be utilised. Early public debate focused on entertainment, education and training for families, and even the convenience of online shopping, but there was no mention about the huge opportunities for farming.

However, over at the University of New England, Professor David Lamb and his colleagues at the Precision Agriculture Research Group (PARG) immediately grasped the significance. Professor Lamb has been driving precision agriculture technologies since the early 1990s. He recognised the NBN as an opportunity to plug-in precision agriculture directly to broadband communications.

With partners including CSIRO, the group set about transforming one of the university farms, Kirby Farm, into a SMART Farm (Sustainable, Manageable and Accessible Rural Technologies): a model that can showcase what is possible when existing precision agriculture technologies meet the heavy-lifting capabilities of the NBN.

Prof David Lamb from the University of New England with a soil moisture probe

Professor David Lamb from the University of New
England with a soil moisture probe.There are 100 of
these probes strategically located across Kirby Farm.
The data they generate builds a water profile of the
whole farm and assists with decisions such as when
to sow and topdress.

The SMART Farm is being developed as a network of soil, plant, weather, animal, machinery and other asset sensors, which will stream data into the command centre – the farmhouse – via information service providers that can be located anywhere in Australia.

Kirby Farm

Kirby is an 1800-hectare mixed farming operation that, at first glance, looks just like any other working property. The only hints that it may be different are antennae on the roof of the calf shed and a number of tall in-ground probes dotted around the property.

But when paired with the NBN, these unobtrusive add-ons create a management framework in which elements such as soil moisture can be measured in real-time, stock-tracking technologies can create SMS alerts about anything from dog attacks to a parasite burden, and the ability to talk to an agronomist, local veterinarian or heavy vehicle mechanic remotely and in real-time stand to revolutionise farm management.

“As soon as we got a whiff of the NBN coming we started thinking, ‘Now we’ll be able to start bringing together all these technologies’,” Professor Lamb says. “Things like yield monitors on harvesters, EM38s for measuring soil moisture and satellite remote sensing.

“We’ve always had the problem of getting the data off the sensors and to the service provider and then getting the service provider connected back to the farmer. It’s always meant time in the car, mailing DVDs, and so on. Now the NBN and SMART farming allow us to start stitching all this information together.”

The probes form a wireless sensor network for
monitoring of soil conditions. Every five minutes, each
wireless monitoring station samples soil moisture,
soil temperature, soil electrical conductivity and air
temperature, and this data is continually updated and
made available online. The project is a collaboration
between the University of New England and CSIRO
and can be viewed at
www.sensornets.csiro.au/deployments/684.

Farmhouse data hub

The NBN provides the opportunity for the farmhouse to become a data hub. The information needed for decisions ranging from crop choice, rotations and stocking rates to day-to-day tasks such as checking crops, stock and watering points can now be gathered by remote sensors.

The broadband capability of the NBN makes the farm digitally dynamic. Data can be streamed in via mobile or fixed-line networks to a website from which a farmer can share information with his or her advisers, service providers, or other growers.

Professor Lamb adds that a SMART Farm grower will be able to talk to their agronomist via a high-definition videoconference, with data able to be simultaneously shared and worked on together.

He says an agronomist will no longer need to spend half the day in a car driving to clients. There will be no need to lug around a laptop and a stack of client CDs.

Time freed up by eliminating long hours driving, says Professor Lamb, is time an agronomist can now use for creative thinking: for serious problem-solving rather than just ensuring routine tasks are met.

Smart searching

Although the technology for all this is still new, Professor Lamb says an intelligent search system will ultimately give farmers the information they need before they even ask for it.

“People using the internet are either in search mode or explore mode,” he explains. “Google works by looking at the historical pattern of things you’re after, then tries to match it with what it thinks you need. That’s how the semantic web works.

Prof David Lamb at the farmhouse hub

Professor David Lamb believes the National Broadband
Network will usher in a step change in precision
agriculture technologies.

“In other words, the farmer gives the computer a sense of what he or she is after. A search engine will look at pieces of information from what service providers are giving to that farmer, compile it and say, ‘Is this what you’re looking for?’ Yes or no. And every time the farmer goes through that inquiry process, the search engine gets smarter and smarter because it builds a profile of the farmer’s information needs.”

Remote diagnosis

The ability of the NBN to transfer vast quantities of data quickly and with high-definition imaging has led already to the development by the university team of a farm-wide wi-fi network to troubleshoot problems.

With a simple head-mounted camera a farmer can, for example, look into an engine bay and explore a problem with their equipment provider or mechanic, remotely.

The head cam could also be used to link the grower in a paddock with his or her agronomist to identify weeds, insects or diseases, to diagnose nutritional deficiencies, or just to get an opinion on how a crop is progressing.

Similarly, a vet would be able to make a remote animal health diagnosis, potentially saving the farmer the cost of a call-out fee.

Professor Lamb says this type of technology will not replace farm visits, but confine face-to-face meetings to essential matters, making such service delivery more cost-effective.

A tiny head-mounted camera

A tiny head-mounted camera such as this one allows
the farmer to relay high-definition video from remote
locations back to a service provider. It could assist with
animal health diagnosis, machinery troubleshooting or
agronomic advice. PHOTO: Precision Agriculture
Research Group

Remote vision systems

To further make use of the opportunities offered by a wi-fi network, Professor Lamb and his team are also working on remote vision systems.

“You can already buy technology that works via mobile modems, where you can set up a tilt-pan-zoom camera at a point such as a dam, shed or cattle crush. The trick is to trigger an alarm when something needs attention and for the farmer to be able to remotely control the camera to quickly assess the situation. This could be a particularly valuable safety measure around farm dams or silos.

“Even in its most rudimentary form, if the sensor, for example, couldn’t tell the difference between a child and a mob of animals, it would at least alert you to use the remote camera to check.

“And if you establish a farm-wide wi-fi, as we’ve started doing on the SMART Farm, you can set up wirelessly connected cameras wherever you like.”

New way to weigh

This remote vision system will also be used for weighing stock.

Professor Lamb’s team is working with CSIRO researchers to develop an alternative to walkover weighing: a three-dimensional vision based on images gathered by an array of cameras around a watering point. “These cameras continually capture images of animals that visit the watering point and we plan to use those images to effectively create a three-dimensional animal and make a volumetric analysis to determine their weight.”

Remote camera setup on a farm

Cameras such as this one can be set up around the
farm to feed back into a farm-wide wi-fi network. This
one will be used to capture images of stock entering a
watering point. These images will be used to make a
volumetric analysis of the animal's weight, removing
the need for a walkover weighing system.

The livestock on Kirby Farm wear eartags that allow them to be tracked remotely in real time. Radio signals from the tags are captured by antennae located around the farm. University mathematicians have been able to examine data associated with normal stock behaviour and write algorithms that trigger an alarm when an animal is behaving abnormally.

For example, if a cow is ill and has not moved for some hours, the system can send the farmer an SMS alert. It triggers similar warnings when stock move quickly (such as if they’re being chased by a predator or stolen) and can even alert the farmer when the herd has a parasite burden, because the pattern of a sick animal taking water differs from that of healthy animals.

Mapping crop vigour

The PARG team has also been trialling a plane-mounted sensor that measures crop vigour. The small sensor bolts underneath a crop duster, allowing it to complement fertiliser application with crop-vigour mapping.

Professor Lamb says the technology is already moving beyond creating a static map to creating a real-time map able to control how much fertiliser the aircraft puts out – in other words measuring and meeting the crop’s fertiliser requirement simultaneously.

Because of the speed of an aircraft, the technology creates a prescription map of the paddock ahead of the aircraft’s nose. Once the plane has completed a few laps, the sensor uses geostatistics to map where the high, medium and low fertiliser requirement zones are ahead of the aircraft and then calculates the required application.

Professor Lamb says that it is the speed of data transfer now that makes this technology possible. “There is no new science in this sensor that wasn’t around 10 years ago. What is new is the grunt – the computing power and speed.”

The Raptor sensor attached to an aircraft

Precision Agriculture Research Group senior technical
officer Derek Schneider (left) and Professor David
Lamb pictured with the wing-mounted Raptor sensor.
The speed of data transfer allows the sensor to build
a prescription map for fertiliser application and then
control how much fertiliser is applied based on high,
medium and low-requirement zones.
PHOTO: David Elkins

Precision agriculture

Professor Lamb sees the introduction of broadband into farming as the kick along that precision agriculture (PA) needs to be more widely adopted.

“The adoption rate for PA technologies is around 20 per cent, yet we’ve been doing it for 15 to 18 years,” he says. “Why? Because we haven’t been able to find a way to make the data, or information, available to farmers in a ‘ready-to-wear’ way.

“PA technologies that have been developed to date have been limited by the lack of access to broadband. Now that is changing there will be an explosion of ideas around what is possible.”

What is the NBN?

The National Broadband Network (NBN) refers to the broadband (bandwidth), but the physical construction of the network will depend on how different users will actually receive that broadband. This will be a combination of optical fibre, point-to-point wireless and satellite.

The interim satellite product that is available to rural Australia uses the existing mobile satellite network. The limited high bandwidth available through this offers transmission rates of about 100 kilobytes per second upload and one megabyte (MB) per second download. However, when the new NBN dedicated satellites come online, the speed can increase up to a factor of 10, to around 1MB per second upload and 10MB per second download and even better. Some growers will not have access to this technology for some time.

More information:
Professor David Lamb,
02 6773 2844, 0428 886 088, dlamb@une.edu.au,

www.une.edu.au/smart, www.une.edu.au/smart/nbnsmart.php

Region South, North