Generating innovative ideas

The use of Technology in Farming

By: Edson DaSilva
Project Engineer

11th June 2019

3 minute read

Home » Insights » EcoTech

The use of technology in farming is not a new subject, and over the years many have predicted that this technology or that gadget would revolutionise the industry, but that’s yet to happen. Make no mistake; farming now is significantly different from what it was like 20 years ago, and technology is widely used in the industry now. From the use of GPS Systems to automate machinery to the more recent adoption of drones for tasks such as crop surveying. However, the uptake has not been rapid and revolutionary, but rather a gradual and steady evolution. Some would argue that farmers are at fault, due to their resistance to progress and disinclination to try new technology. I struggle to buy into that. Whilst I can believe that this is true for a handful of individuals, innovators must take some responsibility too. I am persuaded that all too often those at the forefront of technological innovation fail to fully grasp the very real and relevant concerns expressed by prospective end-users. This in turn yields solutions that fail to meet the basic requirements needed for widespread uptake.

I recently attended two events which reinforced some those views, namely Drones for Farming Conference and Sensing for Economic & Production Gains in Agriculture. In particular, these two things that stayed with me looking back:

1. Data is only as valuable as the decisions it enables

In recent years we have become obsessed with data, the more the better some would argue. And with the spread of AI, this trend is set to continue. Whilst data can indeed be a very powerful tool, it needs to enable decision making for it to yield results. The following simple, but effective example illustrates this point. A drone that flies over a field and takes pictures of crops is of little use to a farmer. But if the pictures can be strategically taken such that an agronomist can inspect areas of a field that are not easily accessible it becomes more useful. Better still, if the software can use image recognition and data processing to advise the agronomist about potential diseases, at an early stage such that preventive action can be taken before it is too late, it becomes a very useful tool.

2. UAVs are not only for crops

One of the stories that really caught my attention was the use of a UAV to herd livestock. Wojtek Behnke, a farmer and tech enthusiast wanted to see if he could use UAVs to speed up the process of rounding up sheep. He started off by simply using the sound of the vehicle to steer the flock (similar to the fashion used with a sheepdog), and his initial results were impressive. But over time the flock became more and more accustomed to the sound, to the point where he could fly the vehicle within metres of them and they would not move. At this point, he switched his strategy and adopted a positive reinforcement behaviour approach. Instead of trying to scare the flock he started to persuade them to follow the UAV in the hope of gaining a reward, in this case, feed.
After a slow start, he managed to ‘teach’ the sheep to associate the sound of the UAV to food. The results were astonishing. He successfully ushered the flock through a number of difficult and tricky obstacles and into the desired location using only a UAV. Wojtek’s experience as a shepherd and his knowledge about the animal’s behaviour was vital in recognising the need to change the approaches.

As our population grows and our demand for food increases, there will have to be a paradigm shift in how food is produced. Affordability will continue to be a pressing demand, but sustainability is just as important, especially given the growing concerns about climate change. Technology will no doubt play a big part in it, but the size of its impact and the speed at which it happens will be dictated by how well technologist and producers collaborate.

The use of technology in farming is not a new subject, and over the years many have predicted that this technology or that gadget would revolutionise the industry, but that’s yet to happen. Make no mistake; farming now is significantly different from what it was like 20 years ago, and technology is widely used in the industry now. From the use of GPS Systems to automate machinery to the more recent adoption of drones for tasks such as crop surveying. However, the uptake has not been rapid and revolutionary, but rather a gradual and steady evolution. Some would argue that farmers are at fault, due to their resistance to progress and disinclination to try new technology. I struggle to buy into that. Whilst I can believe that this is true for a handful of individuals, innovators must take some responsibility too. I am persuaded that all too often those at the forefront of technological innovation fail to fully grasp the very real and relevant concerns expressed by prospective end-users. This in turn yields solutions that fail to meet the basic requirements needed for widespread uptake.

I recently attended two events which reinforced some those views, namely Drones for Farming Conference and Sensing for Economic & Production Gains in Agriculture. In particular, these two things that stayed with me looking back:

1. Data is only as valuable as the decisions it enables

In recent years we have become obsessed with data, the more the better some would argue. And with the spread of AI, this trend is set to continue. Whilst data can indeed be a very powerful tool, it needs to enable decision making for it to yield results. The following simple, but effective example illustrates this point. A drone that flies over a field and takes pictures of crops is of little use to a farmer. But if the pictures can be strategically taken such that an agronomist can inspect areas of a field that are not easily accessible it becomes more useful. Better still, if the software can use image recognition and data processing to advise the agronomist about potential diseases, at an early stage such that preventive action can be taken before it is too late, it becomes a very useful tool.

2. UAVs are not only for crops

One of the stories that really caught my attention was the use of a UAV to herd livestock. Wojtek Behnke, a farmer and tech enthusiast wanted to see if he could use UAVs to speed up the process of rounding up sheep. He started off by simply using the sound of the vehicle to steer the flock (similar to the fashion used with a sheepdog), and his initial results were impressive. But over time the flock became more and more accustomed to the sound, to the point where he could fly the vehicle within metres of them and they would not move. At this point, he switched his strategy and adopted a positive reinforcement behaviour approach. Instead of trying to scare the flock he started to persuade them to follow the UAV in the hope of gaining a reward, in this case, feed.

After a slow start, he managed to ‘teach’ the sheep to associate the sound of the UAV to food. The results were astonishing. He successfully ushered the flock through a number of difficult and tricky obstacles and into the desired location using only a UAV. Wojtek’s experience as a shepherd and his knowledge about the animal’s behaviour was vital in recognising the need to change the approaches.

As our population grows and our demand for food increases, there will have to be a paradigm shift in how food is produced. Affordability will continue to be a pressing demand, but sustainability is just as important, especially given the growing concerns about climate change. Technology will no doubt play a big part in it, but the size of its impact and the speed at which it happens will be dictated by how well technologist and producers collaborate.

How to Harvest Infinite Power!

By: Henry Wadsworth
Project Engineer

9th May 2019

6 minute read

Home » Insights » EcoTech

Wouldn’t it be amazing if an electronic device could run forever, for free, on an unlimited supply of energy? No batteries to replace and no plug socket to be tethered to. This might seem like a flight of fancy, but with modern energy harvesting techniques, it is a much more realisable dream.

Energy harvesting is the capture and storage of energy from the environment which can be used to power an electronic device. This is not a new idea, we are all familiar with the idea of using solar panels to charge a battery, but there are many other energy sources to take advantage of such as kinetic, thermal, and electromagnetic energy. Kinetic energy such as that produced by wind can easily be harvested using a turbine, but even vibrations can be converted into a voltage using a piezo-electric transducer so any form of movement has the potential to power electronics. Thermal energy can be converted into electrical energy using a thermoelectric generator (TEG) which is able to produce energy from heat which would normally be lost to the environment. Electromagnetic energy is present all around us, for example, in the form of radio waves and microwaves which we rely on for wireless communications from WIFI to FM radio. In most circumstances this amount of energy is so small that it cannot readily be captured and stored, however, if a device is located within a reasonable distance of a powerful transmitter, it is possible to provide enough energy to keep a low-power device alive.

This is especially the case with modern integrated circuits designed for power harvesting which are enabling increasingly tiny amounts of energy to be gradually accumulated over time in a battery, or super-capacitor in order to power a low-power device almost indefinitely while the energy source is present. In many cases it is also possible to combine multiple energy harvesting sources, which can be useful in environments where the energy sources may be changing, such as if the device is moving through different environments. For example, solar could be combined with piezo-electric, so that the device can harvest sunlight when the sun is visible, and vibration energy when the device is being moved (perhaps the device is worn by a human, or transported on a vehicle).

Energy harvesting typically refers to small scale harvesting of energy as described above, however, larger power sources can also be harnessed. For example, the power flowing through a cable can be harnessed by placing a current transformer around the cable. We recently developed a system using this technology to supply energy in an underground environment where no other power supply was reliably available. This can be a useful way of retro-fitting a device where power cables already exist, but with no other convenient way to tap the energy from the cables. This method has the potential to supply large amounts of power to a device, depending on the amount of current flowing through the cables.

Energy harvesting is a particularly valuable technique for circumstances where it is not possible to carry out maintenance on a device to replace a battery. For example, the device may be inaccessible or just impractical and expensive to access if there are a large number of devices – which is likely to become an increasingly significant problem as inexpensive IOT devices are used for large monitoring networks.

Energy harvesting techniques are becoming increasingly sophisticated which, coupled with a low power microprocessor makes it possible to power devices with energy sources which previously would be considered impractical.

Wouldn’t it be amazing if an electronic device could run forever, for free, on an unlimited supply of energy? No batteries to replace and no plug socket to be tethered to. This might seem like a flight of fancy, but with modern energy harvesting techniques, it is a much more realisable dream.

Energy harvesting is the capture and storage of energy from the environment which can be used to power an electronic device. This is not a new idea, we are all familiar with the idea of using solar panels to charge a battery, but there are many other energy sources to take advantage of such as kinetic, thermal, and electromagnetic energy. Kinetic energy such as that produced by wind can easily be harvested using a turbine, but even vibrations can be converted into a voltage using a piezo-electric transducer so any form of movement has the potential to power electronics. Thermal energy can be converted into electrical energy using a thermoelectric generator (TEG) which is able to produce energy from heat which would normally be lost to the environment. Electromagnetic energy is present all around us, for example, in the form of radio waves and microwaves which we rely on for wireless communications from WIFI to FM radio. In most circumstances this amount of energy is so small that it cannot readily be captured and stored, however, if a device is located within a reasonable distance of a powerful transmitter, it is possible to provide enough energy to keep a low-power device alive.

This is especially the case with modern integrated circuits designed for power harvesting which are enabling increasingly tiny amounts of energy to be gradually accumulated over time in a battery, or super-capacitor in order to power a low-power device almost indefinitely while the energy source is present. In many cases it is also possible to combine multiple energy harvesting sources, which can be useful in environments where the energy sources may be changing, such as if the device is moving through different environments. For example, solar could be combined with piezo-electric, so that the device can harvest sunlight when the sun is visible, and vibration energy when the device is being moved (perhaps the device is worn by a human, or transported on a vehicle).

Energy harvesting typically refers to small scale harvesting of energy as described above, however, larger power sources can also be harnessed. For example, the power flowing through a cable can be harnessed by placing a current transformer around the cable. We recently developed a system using this technology to supply energy in an underground environment where no other power supply was reliably available. This can be a useful way of retro-fitting a device where power cables already exist, but with no other convenient way to tap the energy from the cables. This method has the potential to supply large amounts of power to a device, depending on the amount of current flowing through the cables.

Energy harvesting is a particularly valuable technique for circumstances where it is not possible to carry out maintenance on a device to replace a battery. For example, the device may be inaccessible or just impractical and expensive to access if there are a large number of devices – which is likely to become an increasingly significant problem as inexpensive IOT devices are used for large monitoring networks.

Energy harvesting techniques are becoming increasingly sophisticated which, coupled with a low power microprocessor makes it possible to power devices with energy sources which previously would be considered impractical.

Food for Thought: Food Industry Innovation 2019

Nicholas Hill, Plextek

By: Nicholas Hill
CEO

4th April 2019

6 minute read

Home » Insights » EcoTech

Food Industry Innovation 2019 was both stimulating and thought provoking. Organised by Innovate UK, it was a mixture of presentations, audience polls, pitch sessions and exhibitions by start-ups, and covered everything from novel food science and manufacturing engineering through to the future of supply and distribution. Some themes kept recurring throughout the day, both from presenters and audience polls, and I thought it would be interesting to look at some of these.

A poll covering top trends in food product development had ‘sustainable packaging’, ‘plant-based/vegan foods’, ‘free from foods’ and ‘personalised food/nutrition’ in top place, in that order.

Sustainable Packaging

It’s hardly surprising that sustainable packaging is top of the agenda, with a newly found mass consumer awareness of the environmental crisis that waste plastic has created. There’s a lot of interest in alternative (e.g. bio-degradable) plastics. In their favour they would permit a ‘business as usual’ approach, allowing manufacturers to continue to use the same volume of plastic packaging but making disposal more practical. However, plastics that retain all their functionality while storing food or drink while degrading rapidly and safely in the environment are still a good way off.

Take cPPA for example, which has been known for decades to depolymerise rapidly on demand. The trick that is still being perfected is how to prevent it from degrading while still in use, under the full range of environmental conditions. Other work is looking at catalytic methods to rapidly break down plastics that are in common use already, such as PET. And even when the science has provided the techniques that we need, actually recycling or composting the current volumes of plastic packaging would require a massive change to the current localised, ad hoc approach to recycling that we have in the UK.

It was somewhat alarming to hear one supermarket representative make an appeal for other industries to take the hit on reducing packaging waste. The argument was that the hygiene requirements in the food industry gave them a greater justification to use packaging than other industries, and holding up the toy industry as an example of a less worthy case. I don’t know about you, but I’d estimate that food packaging makes up at least 90% of the waste in my recycle bin each week.

The two most obvious approaches that can be implemented in the short term are reducing the use of plastic packaging at source and encouraging greater return of plastic packaging by consumers for reuse or recycling. Both of these require government intervention to push the problem back on the companies that are putting the plastic on to the market, for example by taxing or regulating the use of disposable plastic, or by encouraging reuse and recycling by enforcing deposit-return schemes. It is good to see that the government is starting to take action on the latter.

Diet-Driven Foods

To see the development of plant-based foods rated by the audience as the second most important trend was music to my ears. Having lived a meat-free diet for almost thirty years, seeing the dramatic rise in interest in plant-based and vegan diets in the past year or so has been rather astonishing. Whereas the traditional motivation for non-meat diets came from animal welfare concerns, it seems that the current trend is being driven by awareness of the health benefits of a plant-based diet and of the environmental destruction caused by livestock farming. Whereas in the past the vegan consumer has been served exclusively by niche suppliers, the mainstream food industry has finally engaged with this market. I would expect this to create a welcome increase in innovation in the area, due to the size of research budgets at the disposal of the mainstream producers and the untapped potential in the market. If you’re reading this, someone please come up with a vegan cheese that really tastes like mature cheddar!

Somewhat related to the former topic is the rise in provision of ‘free from’ ranges of food. If vegetarians and vegans have had a hard time in the past finding acceptable food, people with dietary intolerances have had at least as great a struggle. It’s great news that ‘free from’ is also gradually moving into the mainstream, and likewise the engagement by the major manufacturers must aid rapid innovation leading to more appealing products and greater choice.

Personalised Nutrition

At fourth place in this list of trends was personalised food/nutrition. There was an interesting mix of ideas about what this might mean. From the online shopping perspective, this could be just increasing the intelligence of the virtual grocer that helps suggest foods that you might like; eventually, it might be a match for the real local grocers we used to have before the supermarkets took over. A proponent of wearables technology suggested that fitness trackers and smartphones could be making dietary suggestions based on activity or other deductions about lifestyle, and then ordering appropriate groceries or meals for you. Something like: “no pizza for you this evening as I see you skipped your scheduled run”. Into this mix was added the idea of using genetic profiling to identify foods that might be compatible, or not, with a particular individual.

Another section of the conference was looking at technologies that might have the biggest impact on manufacturing efficiency. While many of the identified technologies were predictable: automation, AI, big data, robotics, blockchain, some of the applications were interesting.

Supply Chain Traceability

A theme that caught my attention was traceability in the supply chain. I hadn’t realised what a huge issue tracking the provenance of food as it passes through the complex supply chain is. If I purchase some organic tomato soup, someone needs to be able to check that the original tomatoes were organic, that the same tomatoes made it to the soup factory, and through the factory’s many processes, that the resulting soup made it to the warehouse, and finally to the supermarket shelf. Blockchain technology was presented as the foundation for creating a distributed ledger that allowed multiple parties to track to provenance and progress of an item as it passed through the supply chain. In addition to the basic benefits of provenance checking, health and safety benefits due to the ability to organise swift and accurate product recalls were highlighted.

This is a really useful deployment of modern technology to an old problem. Its principal limitation is of course that you are only really following the provenance of the label that was attached to the tomatoes in the example used or the box they were in. You’d currently have no way of knowing if they were switched out for alternative products bearing the original labelling. Embedding RFID tags into our food is clearly a non-starter, so further technology would be needed to provide non-invasive scanning of food items to backup their provenance claims. There are plenty of chemical and optical detection solutions in existence to identify types of fruit (a Granny Smith from a Golden Delicious), or ripeness or damage to items. To detect pesticide residues on supposedly organic products, or prove that the country or region of origin is as claimed on the label, we’d need sophisticated sensing technology in small packages and at low cost – this appears to be coming, but we’re not there yet.

Visions of the Future

A lot of the drive for automation is coming from the desire for increased productivity, and the grandest vision presented painted a picture of a field-to-table manufacturing and supply chain that had no human involvement at all, with robotic harvesting, shipping, sorting, warehousing and delivery, all driven by vast amounts of AI. On the plus side, this would certainly be great for productivity. It would also allow for a great degree of customisation and tailoring to each end customer, with the economies of scale and sophistication needed to deliver bespoke products on demand. And perhaps this technology would be an enabler for another trend – the growth in desire for artisan foods. There’s a move to the simplicity of ‘homemade’ values and away from mass-market, highly processed foods. What’s needed to support this are ways of making artisan goods without the labour intensive processes traditionally required. Proponents of robotics and AI would claim to have the answer.

On the other hand, food isn’t just another consumer product. We have a much more basic, emotional connection with food than anything else we buy. Millions of years of evolution have given us a sensory system that allow us to assess, judge and select the food we eat. Fresh food is a complete multi-sensory experience – we can see it, smell it, touch it, feel it, taste it, and sometimes even hear it. Boxing in a person behind a computer screen so that purchasing decisions are made using only one sense – our eyesight – surely sanitises and diminishes the experience. Can technology ever replace the human experience of the classic fresh food market? There’s a challenge.

Food Industry Innovation 2019 was both stimulating and thought provoking. Organised by Innovate UK, it was a mixture of presentations, audience polls, pitch sessions and exhibitions by start-ups, and covered everything from novel food science and manufacturing engineering through to the future of supply and distribution. Some themes kept recurring throughout the day, both from presenters and audience polls, and I thought it would be interesting to look at some of these.

A poll covering top trends in food product development had ‘sustainable packaging’, ‘plant-based/vegan foods’, ‘free from foods’ and ‘personalised food/nutrition’ in top place, in that order.

Sustainable Packaging

It’s hardly surprising that sustainable packaging is top of the agenda, with a newly found mass consumer awareness of the environmental crisis that waste plastic has created. There’s a lot of interest in alternative (e.g. bio-degradable) plastics. In their favour they would permit a ‘business as usual’ approach, allowing manufacturers to continue to use the same volume of plastic packaging but making disposal more practical. However, plastics that retain all their functionality while storing food or drink while degrading rapidly and safely in the environment are still a good way off.

Take cPPA for example, which has been known for decades to depolymerise rapidly on demand. The trick that is still being perfected is how to prevent it from degrading while still in use, under the full range of environmental conditions. Other work is looking at catalytic methods to rapidly break down plastics that are in common use already, such as PET. And even when the science has provided the techniques that we need, actually recycling or composting the current volumes of plastic packaging would require a massive change to the current localised, ad hoc approach to recycling that we have in the UK.

It was somewhat alarming to hear one supermarket representative make an appeal for other industries to take the hit on reducing packaging waste. The argument was that the hygiene requirements in the food industry gave them a greater justification to use packaging than other industries, and holding up the toy industry as an example of a less worthy case. I don’t know about you, but I’d estimate that food packaging makes up at least 90% of the waste in my recycle bin each week.

The two most obvious approaches that can be implemented in the short term are reducing the use of plastic packaging at source and encouraging greater return of plastic packaging by consumers for reuse or recycling. Both of these require government intervention to push the problem back on the companies that are putting the plastic on to the market, for example by taxing or regulating the use of disposable plastic, or by encouraging reuse and recycling by enforcing deposit-return schemes. It is good to see that the government is starting to take action on the latter.

Diet-Driven Foods

To see the development of plant-based foods rated by the audience as the second most important trend was music to my ears. Having lived a meat-free diet for almost thirty years, seeing the dramatic rise in interest in plant-based and vegan diets in the past year or so has been rather astonishing. Whereas the traditional motivation for non-meat diets came from animal welfare concerns, it seems that the current trend is being driven by awareness of the health benefits of a plant-based diet and of the environmental destruction caused by livestock farming. Whereas in the past the vegan consumer has been served exclusively by niche suppliers, the mainstream food industry has finally engaged with this market. I would expect this to create a welcome increase in innovation in the area, due to the size of research budgets at the disposal of the mainstream producers and the untapped potential in the market. If you’re reading this, someone please come up with a vegan cheese that really tastes like mature cheddar!

Somewhat related to the former topic is the rise in provision of ‘free from’ ranges of food. If vegetarians and vegans have had a hard time in the past finding acceptable food, people with dietary intolerances have had at least as great a struggle. It’s great news that ‘free from’ is also gradually moving into the mainstream, and likewise the engagement by the major manufacturers must aid rapid innovation leading to more appealing products and greater choice.

Personalised Nutrition

At fourth place in this list of trends was personalised food/nutrition. There was an interesting mix of ideas about what this might mean. From the online shopping perspective, this could be just increasing the intelligence of the virtual grocer that helps suggest foods that you might like; eventually, it might be a match for the real local grocers we used to have before the supermarkets took over. A proponent of wearables technology suggested that fitness trackers and smartphones could be making dietary suggestions based on activity or other deductions about lifestyle, and then ordering appropriate groceries or meals for you. Something like: “no pizza for you this evening as I see you skipped your scheduled run”. Into this mix was added the idea of using genetic profiling to identify foods that might be compatible, or not, with a particular individual.

Another section of the conference was looking at technologies that might have the biggest impact on manufacturing efficiency. While many of the identified technologies were predictable: automation, AI, big data, robotics, blockchain, some of the applications were interesting.

Supply Chain Traceability

A theme that caught my attention was traceability in the supply chain. I hadn’t realised what a huge issue tracking the provenance of food as it passes through the complex supply chain is. If I purchase some organic tomato soup, someone needs to be able to check that the original tomatoes were organic, that the same tomatoes made it to the soup factory, and through the factory’s many processes, that the resulting soup made it to the warehouse, and finally to the supermarket shelf. Blockchain technology was presented as the foundation for creating a distributed ledger that allowed multiple parties to track to provenance and progress of an item as it passed through the supply chain. In addition to the basic benefits of provenance checking, health and safety benefits due to the ability to organise swift and accurate product recalls were highlighted.

This is a really useful deployment of modern technology to an old problem. Its principal limitation is of course that you are only really following the provenance of the label that was attached to the tomatoes in the example used or the box they were in. You’d currently have no way of knowing if they were switched out for alternative products bearing the original labelling. Embedding RFID tags into our food is clearly a non-starter, so further technology would be needed to provide non-invasive scanning of food items to backup their provenance claims. There are plenty of chemical and optical detection solutions in existence to identify types of fruit (a Granny Smith from a Golden Delicious), or ripeness or damage to items. To detect pesticide residues on supposedly organic products, or prove that the country or region of origin is as claimed on the label, we’d need sophisticated sensing technology in small packages and at low cost – this appears to be coming, but we’re not there yet.

Visions of the Future

A lot of the drive for automation is coming from the desire for increased productivity, and the grandest vision presented painted a picture of a field-to-table manufacturing and supply chain that had no human involvement at all, with robotic harvesting, shipping, sorting, warehousing and delivery, all driven by vast amounts of AI. On the plus side, this would certainly be great for productivity. It would also allow for a great degree of customisation and tailoring to each end customer, with the economies of scale and sophistication needed to deliver bespoke products on demand. And perhaps this technology would be an enabler for another trend – the growth in desire for artisan foods. There’s a move to the simplicity of ‘home made’ values and away from mass-market, highly processed foods. What’s needed to support this are ways of making artisan goods without the labour intensive processes traditionally required. Proponents of robotics and AI would claim to have the answer.

On the other hand, food isn’t just another consumer product. We have a much more basic, emotional connection with food than anything else we buy. Millions of years of evolution have given us a sensory system that allow us to assess, judge and select the food we eat. Fresh food is a complete multi-sensory experience – we can see it, smell it, touch it, feel it, taste it, and sometimes even hear it. Boxing in a person behind a computer screen so that purchasing decisions are made using only one sense – our eyesight – surely sanitises and diminishes the experience. Can technology ever replace the human experience of the classic fresh food market? There’s a challenge.