Last month, Land Mobile ran an interview with the founder and CEO of Hiber, a company which is in the process of deploying a constellation of nano satellites for the purpose of providing IoT coverage for the most isolated parts of the world. Stimulating as that was, we are subsequently following up this month with a feature taking a more in-depth look at specific use-cases relating to the provision of this kind of remote area coverage.
With that in mind, arguably the most intuitive environment to help explore this topic is farming, an activity that not only consists of any number of invariably complex individual processes but also, very often, takes place in the absolute middle of nowhere. It also just happens to be one of the core activities on which depends the continued survival of the human race on planet Earth.
The first technology we shall be focusing on is Thingstream, a low-power IoT solution that has recently been rolled out on a macadamia nut farm in South Africa. We will also be touching on the work being carried out by Cambridge Consultants within this field, looking in particular at irrigation.
The Internet of Nuts
Thingstream describes itself – according to its website – as: “The only provider of truly global, low-power IoT connectivity.” It accomplishes this through enabling devices to connect to “the strongest GSM [2G, 3G or LTE] network available, almost anywhere in the world, and communicate using IoT Messaging, a unique combination of MQTT and USSD/UDP messaging protocols. Thingstream enables IoT connectivity in over 190 countries across 600-plus networks.”
Going into greater detail about why this functionality might be necessary, particularly when it comes to work within a farming environment, the company’s sales director for South Africa, Tony Doubell, said: “When you’re talking about wide open spaces, typically situated on massive areas of farmland, obviously cellular data connections can be a real issue. There will be little or no traditional cellular data in those types of rural areas.
“That’s very much the case in Africa, which is obviously a challenge for us, for instance when trying to collect information from the fields. Looking at it from the point of view of the farmer, if you’re trying to monitor something, you could also be in a situation where you have a number of IoT sensors, all of which are connected, but operating on different GSM networks. You’re then getting in a situation where you’re operating on multiple platforms, which again complicates things further.”
According to Doubell, the story of Thingstream begins with its original owner Myriad Group AG, a company which develops and markets messaging gateways to mobile telecoms operators around the world.
Picking up the narrative, Trevor Hart-Jones, MD of Thingstream partner company Pylot, says: “We basically went right back to the drawing board, and began to focus on the networking side of things. We built our first IoT unit in collaboration with Thingstream, which we called the P3000, and we’re currently building the next version – the P7000 – which will have an increased number of input and output options.
“We can connect to most analogue or digital-type interfaces, soil moisture probes and so on, and over the past few months we’ve built a more generic gateway which also helps us to connect to more devices in the field. Clearly, we’ve also now converted our hardware component in order to support the Thingstream platform, so it’s really just a matter of inserting our sims.”
He continues: “We basically utilise the operator network coverage, which is far greater than Sigfox or LoRa. The other nice thing about the network is that it also allows users to roam ubiquitously across other mobile operators, so I can take the same device, deploy the system in something like 190 countries, and not have to change anything.
“The crucial point is that rather than using the data network, it’s effectively using the low-level voice network to send a few data packets. In terms of security, because there’s no IP connection it’s also very difficult to eavesdrop or to hack.”
Going back to the subject of farming, particularly in remote regions of a country such as South Africa, there are an apparent myriad of use-cases, ranging from the aforementioned soil monitoring to functions related to the processing of product itself.
Speaking of Pylot’s current relationship with the farm in question, Hart-Jones says: “The initial use-case was built around automating their drying process, which essentially means pulling the data from their plant into our portal every five minutes or so. After we’ve accrued enough information, we can start to look at trends and analysis.
“We also have a project to manage their assets on the farm itself, which is something we’re in the process of deploying at the moment. That includes things such as tractors, trailers, harvesting equipment and so on, and again, we’re reporting back with the data every five minutes. It’s an incredibly useful way – for instance – of discovering why one tractor in particular might be constantly going into the maintenance yard. Is it the route it takes across the field, a particular driver, or something different entirely?”
He continues: “In terms of the environment itself, a lot of farming contexts are very similar to one another. The only difference really is that other types of farming don’t typically need drying or curing facilities, which is one of the requirements with macadamia nuts.”
As is probably apparent from the above comments, the Thingstream solution consists of not just network access and sensors, but also the means to analyse the information itself. “What the farmer needs to do, fundamentally, is to be able to look back at his yield over the course of the year and work out why some weeks were better than others,” says Hart-Jones. “From there, you cross-reference the information in terms of the processes themselves.
“Another common challenge for farmers is around the management of water and electricity – both of which are absolutely key – and again we have various proposals out in relation to this.
“We’re monitoring the flow of water through the pumps and pipes, as well as levels of irrigation, which a large number of farms in the country rely on, particularly when there hasn’t been much rain. Fertigation management is an issue as well, which essentially involves monitoring the level of potassium when running fertiliser through the water system.”
‘Smart irrigation’ is one area where technology is aiding crop farmers
Precision agriculture
Staying on the subject of water, the aforementioned Cambridge Consultants is also helping to develop a variety of solutions in relation to the efficient use of this precious resource. A key area of work for them in this regard is around irrigation.
“We’ve been involved in a number of projects in the field of what we call smart irrigation,” says Dr Derek Long, the company’s head of telecoms and mobile. “The objective of these is basically to help increase crop yield and minimise water consumption – or rather, water wastage – through it not being used appropriately.
“Farmers need to provide their plants with just the right amount of water in relation to what it is they’re growing. A part of that is reducing evaporation, waste water run-off and so on.”
According to Long, there are a variety of technological solutions being rolled out in this area, such as the kind of sensors discussed above. These can either be deployed on a plant-by-plant basis, or in the vicinity of several plants at a time, depending on the potential value of the yield. This is something which Long refers to as “precision agriculture”.
“In terms of what kind of solution is deployed, the choice is less dependent on the environment and more on the kind of plant that is being monitored,” he says. “If you have a high-value plant – grapes used in wine-making, for instance – then putting a sensor in each small group of plants is not an issue at all. If it’s something like carrots, then that might be considered too costly a solution.”
In that case, a more appropriate technology might be “machine vision”, which essentially facilitates visual analysis of the crops in question. Long continues: “Machine vision is a way to survey large groups of plants, in order to be able to make a judgement on their health and what they need in order to thrive.
“You could use drones for that, or alternately position the solution on top of a tractor or other type of farm machinery to give you a view of the field. Another potential application is in relation to harvesting, again, at the end of the scale, containing more valuable crops such as fruit.
“There’s a substantial amount of interest at the moment in incorporating ‘robots’, which would perform some kind of machine vision to discern whether something is ripe. They would then pick it themselves, something which involves a tactile feedback mechanism making sure they do it with the appropriate amount of force.”
Examples of these robotic capabilities are ‘Mamut’ and ‘Hank’, both developed by Cambridge Consultants. Mamut, which is an AI-powered autonomous unit, prowls the fields equipped with an “array of sensors to map and navigate its surroundings” and capture “detailed crop data”. Hank is a robot with a sensory system embedded in flexible pneumatic ‘fingers’ inspired by the human hand. According to the organisation, this is designed to provide “a sophisticated sense of touch and slip”, meaning it can grip delicate objects using the right amount of pressure.
Farming has always been a crucial activity through which human society maintains its own existence. That being the case, and with scientists now warning us of the impact of global warming on the environment (and therefore food production), it makes sense to make use of any advantage that we can find.
While the technology mentioned above probably won’t save the human race, if nothing else it will likely herald the dawn of increasingly efficient systems. That can’t help but be good for both consumers and the companies that are supplying them.
