No escape even for agrochemicals!

28/09/2017
In this article key points that are covered in depth in the IDTtechEX published report “Agricultural Robots and Drones 2017-2027: Technologies, Markets, Players” by Dr Khasha Ghaffarzadeh and Dr Harry Zervos are discussed. 

New robotics is already quietly transforming many aspects of agriculture, and the agrochemicals business is no exception. Here, intelligent and autonomous robots can enable ultraprecision agriculture, potentially changing the nature of the agrochemicals business. In this process, bulk commodity chemical suppliers will be transformed into speciality chemical companies, whilst many will have to reinvent themselves, learning to view data and artificial intelligence (AI) as a strategic part of their overall crop protection offerings.

Computer vision
Computer vision is already commercially used in agriculture. In one use case, simple row-following algorithms are employed, enabling a tractor-pulled implement to automatically adjust its position. This relieves the pressure on the driver to maintain an ultra-accurate driving path when weeding to avoid inadvertent damage to the crops.

The computer vision technology is however already evolving past this primitive stage. Now, implements are being equipped with full computer systems, enabling them to image small areas, to detect the presence of plants, and to distinguish between crop and weed. The system can then instruct the implement to take a site-specific precision action to, for example, eliminate the weed. In the future, the system has the potential to recognize different crop and weed types, enabling it to take further targeted precision action.

This technology is already commercial, although at a small scale and only for specific crops. The implements are still very much custom built, assembled and ruggedized for agriculture by the start-ups themselves. This situation will continue until the market is proven, forcing the developers to be both hardware and software specialists. Furthermore, the implements are not yet fully reliable and easy to operate, and the upfront machine costs are high, leading the developers to favour a robotic-as-a-service business model.

Nonetheless, the direction of travel is clear: data will increasingly take on a more prominent (strategic) role in agriculture. This is because the latest image processing techniques, based on deep learning, feed on large datasets to train themselves. Indeed, a time-consuming challenge in applying deep learning techniques to agriculture is in assembling large-scale sets of tagged data as training fodder for the algorithms. The industry needs its equivalents of image databases used for facial recognition and developed with the help of internet images and crowd-sourced manual labelling.

In not too distant a future, a series of image processing algorithms will emerge, each focused on some set of crop or weed type. In time, these capabilities will inevitably expand, allowing the algorithms to become applicable to a wider set of circumstances. In parallel, and in tandem with more accumulated data (not just images but other indicators such NDVA too), algorithms will offer more insight into the status of different plants, laying the foundation of ultra-precision farming on an individual plant basis.

Agriculture is a challenging environment for image processing. Seasons, light, and soil conditions change, whilst the plant themselves transform shape as they progress through their different stages of growth. Nonetheless, the accuracy threshold that the algorithms in agriculture must meet are lower than those found in many other applications such as autonomous general driving. This is because an erroneous recognition will, at worse, result in elimination of a few healthy crops, and not in fatalities. This, of course, matters economically but is a not safety critical issue and is thus not a showstopper.

This lower threshold is important because achieving higher levels of accuracy becomes increasingly challenging. This is because after an initial substantial gain in accuracy improvement the algorithms enter the diminishing returns phase where lots more data will be needed for small accuracy gains. Consequently, algorithms can be commercially rolled out in agriculture far sooner, and based on orders of magnitude lower data sizes and with less accuracy, than in many other applications.

Navigational autonomy
Agriculture is already a leading adapter of autonomous mobility technology. Here, the autosteer and autoguide technology, based on outdoor RTK GPS localization, are already well-established. The technology is however already moving towards full level-5 autonomy. The initial versions are likely to retain the cab, enabling the farmer/driver to stay in charge, ready to intervene, during critical tasks such as harvesting. Unmanned cable versions will also emerge when technology reliability is proven and when users begin to define staying in charge as remote fleet supervision.

The evolution towards full unmanned autonomy has major implications. As we have discussed in previous articles, it may give rise to fleets of small, slow, lightweight agricultural robots (agrobots). These fleets today have limited autonomous navigational capability and suffer from limited productivity, both in individual and fleet forms. This will however ultimately change as designs/components become standardized and as the cost of autonomous mobility hardware inevitably goes down a steep learning curve.

Agrobots of the future
Now the silhouette of the agrobots of the future may be seen: small intelligent autonomous mobile robots taking precise action on an individual plant basis. These robots can be connected to the cloud to share learning and data, and to receive updates en mass. These robots can be modular, enabling the introduction of different sensor/actuator units as required. These robots will never be individually as productive as today’s powerful farm vehicles, but can be in fleet form if hardware costs are lowered and the fleet size-to-supervisor ratio is increased.

What this may mean for the agrochemicals business is also emerging. First, data and AI will become an indispensable part of the general field of crop protection, of which agrochemical supply will become only a subset, albeit still a major one. This will mandate a major rethinking of the chemical companies’ business model and skillsets. Second, non-selective blockbuster agrochemicals (together with engineered herbicide resistant seeds) may lose their total dominance. This is because the robots will apply a custom action for each plant, potentially requiring many specialized selective chemicals.

These will not happen overnight. The current approach is highly productive, particularly over large areas, and off-patent generic chemicals will further drive costs down. The robots are low-lying today, constricting them to short crops. Achieving precision spraying using high boys will be a mechanical and control engineering challenge. But these changes will come, diffusing into general use step by step and plant by plant. True, this is a long term game, but playing it cannot be kicked into the long grass for long.

@IDTechEx #Robotics #Agriculture #PAuto
Advertisements

Is AI all it is cracked up to be?

28/03/2017
In this article, Stephen Parker, CEO of Parker Software, examines whether artificial intelligence is all it’s cracked up to be.

If planet Earth had been created one year ago, the human species would be just ten minutes old. Putting this into context, the industrial era would have kick-started a mere two seconds ago. Thanks to human influence, the pace of technological advancement on Earth is astonishing. However, we are already on the verge of the next change. The potential of artificial intelligence has been discussed by scientists since the 1950s and modern technological advances are finally bringing this technology to the masses. 

Research suggests that artificial intelligence could be as ‘smart’ as human beings within the next century. Originally, human programmers were required to handcraft knowledge items painstakingly. Today, however, one-off algorithms can teach machines to take on and develop knowledge automatically, in the same way a human infant would. Artificial intelligence has reached a critical tipping point and its power is set to impact every business, in every industry sector.

Already, 38 per cent of enterprises are using artificial intelligence in their business operations and this figure is set to grow to 62 per cent by 2018. In fact, according to predictions by Forrester, investments in artificial intelligence technology will increase three-fold in 2017. These figures mean that the market could be worth an estimated $47 billion by 2020. 

Intelligent assistance
One of the most notable applications of AI from the past few years is the creation of intelligent assistants. Intelligent assistants are interactive systems that can communicate with humans to help them access information or complete tasks. This is usually accomplished with speech recognition technology; think Apple’s Siri, Microsoft’s Cortana or Amazon’s Alexa. Most of the intelligent assistants that we are familiar with today are consumer facing and are somewhat general in the tasks they can complete. However, these applications are now making their way into more advanced customer service settings.

While there is certainly a space for these automated assistants in the enterprise realm, there is a debate as to whether this technology could fully replace a contact centre agent.

Automation is widely recognised as a valuable tool for organisations to route the customer to the correct agent. However, completely handing over the reins of customer management to a machine could to be a step too far for most businesses. Even the most advanced AI platforms only hold an IQ score equivalent to that of a four-year-old, and naturally, businesses are unlikely to entrust their customer service offering to a child.

The human touch
Automated processes are invaluable for speeding up laborious processes and completing monotonous customer service tasks. But as any customer service expert will tell you, the human touch is what elevates good service to an excellent experience for the customer. Simple tasks will no doubt be increasingly managed and completed using automation and AI-enabled agent support systems, whereas complex issues will still require the careful intervention of a human agent.

During a TED Talk on artificial intelligence, philosopher and technologist Nick Bostrom claimed that “machine intelligence is the last invention that humanity will ever need to make.” However, contact centre agents needn’t hang up their headsets just yet.  Artificial intelligence won’t be replacing the call centre agent any time soon. The only guarantee is that the role of a call centre agent will continue to evolve after all, the industrial revolution was only two seconds ago.

@ParkerSoftware #PAuto

It IS rocket science!

13/03/2017

Graham Mackrell, managing director Harmonic Drive, explains why its strain wave gears have been the top choice in space for over forty years.

Anything that goes into space is seen as the pinnacle of human creation. Astronauts are highly trained and are at the peak of physical fitness, space shuttles are crafted by large teams of expert engineers and all the technology used is so high-tech it’s as if it belongs to science fiction.

Driving on Mars!

Many decades ago, the first Harmonic Drive gears were sent into space during the Apollo 15 mission. Even from the beginnings of the space race, the expectations for the technology used were high. The equipment used in space had to be reliable, compact and lightweight and given the increasing demands on equipment in today’s space missions, it must also now be highly accurate with zero backlash and have high torque capacity.

When aerospace engineers were recently designing a new space rover, they looked to Harmonic Drive gears for reliability. Due to the obvious difficulties of performing repairs in space, a high mean time between equipment failures is a high priority. Harmonic Drive products achieve this by prioritising quality throughout the entire design and manufacturing process.

It is vital that aerospace gears are thoroughly tested before they are sent to customers, ensuring that they always receive a quality product. At Harmonic Drive, we test products using finite element method (FEM) testing. This process simulates real world physics to ensure that the product is capable of surviving in space. For example, structural testing is carried out to ensure the product is robust and the space rover travelling over rough terrain will not damage the actuators used in the wheels. Thermodynamic properties are also important as aerospace gears are often exposed to both extremes of the temperature range, which are tested in the initial design process.

Also considered in the design process is the part count of the aerospace gears. Harmonic uses a low part count which means that they are maintenance free. In addition, there is a lower chance of components failing giving the gears a high Mean Time Between Failure (MTBF). This also contributes to the compactness and light weight of the gears, a feature essential in space.

Another key feature for aerospace gears is high torque capacity and zero backlash. This is essential for systems which communicate the location of the rover to the control room. If traditional, high backlash gears were to be used, the system would misreport the rover’s location. This would cause problems when the rover is used to survey uncharted areas of planets and could lead to inaccurate mapping. Due to the emphasis on high precision with Harmonic Drive gears, this problem can be avoided.

The numerous quality processes that Harmonic Drive undertakes have led to recognition from a number of accrediting bodies. Harmonic Drive products are AS9100 certified, a specific aerospace standard for the design, manufacture and sale of precision gear reducers, servo-actuators and electro-mechanical positioning systems.

To be the pinnacle of global technology, there are no shortcuts. Components used in aerospace technology must be subject to vigorous testing in order to be reliable, safe and have a long product life.

• The MARS adventure: The NASA site.
@HarmonicDriveUK #PAuto #Robotics @StoneJunctionPR

Preparing pharmaceutical and medical technology for the future.

27/09/2016

Production environment requirements in the pharmaceutical and medical technology sectors are very high and producers need to keep abreast of current industry trends. Such trends include; process optimisation for the purpose of increasing overall equipment effectiveness (OEE), effective asset life cycle management and predictive maintenance using enterprise mobility and intelligent solutions (smart apps). Increasing networking along with the use of automation technology in accordance with Industry 4.0 have paved the way for these developments.

Industry 4.0
The Internet of Things has found its way into production in the form of Industry 4.0: increasingly networked systems with more communication-capable components have meant an ever-increasing volume of data. Thanks to “big data”, production is being made more and more intelligent, with the pinnacle of achievement being the smart factory. The key to success lies in determining the right information from the mass of data available, analysing the data and drawing findings from them. The aim of such Smart Data Management is to optimise the plant in question and prepare it for the future in terms of site operational excellence.

Robots are increasingly taking on handling tasks to reduce human effort in the medical/pharma sectors, for example by supplying filled syringes to the end-of-line packaging station.

Robots are increasingly taking on handling tasks to reduce human effort in the medical/pharma sectors, for example by supplying filled syringes to the end-of-line packaging station.

In addition, overall equipment effectiveness (OEE) has developed into an obvious trend. The main factors that have contributed towards this have been optimised plant utilisation and productivity for which manufacturing execution systems (MES) and enterprise resource planning (ERP) are of vital importance. Interfaces such as the MES interface from Mitsubishi Electric enable data to be collected quickly and easily at plant level and transferred to higher-level MES or ERP systems at the control level for further analysis. OEE can be specifically optimised based on the results without the need for a gateway PC to transfer the data. Based on the System Q PLC from Mitsubishi Electric the MES interface can be configured by a plant engineer in just around 15 minutes.

Cost control strategies look for a more compact design, shorter production cycles and substantially minimised waste. Automation technologies strongly support these approaches with robot technology in particular being used to achieve these aims.

Collaboration between Robots and Humans
Today’s pharmaceutical and medical technology production environments see robots and human operators increasingly working side by side. Mitsubishi Electric’s integrated Robot Safety Solution helps manufacturers to boost productivity and lift human-machine collaboration by allowing the robot to continue operation, within tight constraints, while operators access its work cell. Safety sensing technology detects movements in two predefined zones within the operating environment of the manufacturing cell and transmits the information to the SafePlus safety system. A reduced operating speed or a movement stop is then assigned to the robot in real time, thus enabling operators to work in close proximity to the moving robot without a safety cage. As a result, humans and robots are able to work within an environment where the risk of danger is eliminated.

Robot-assisted handling solutions: compact, flexible and quick
Space is an expensive commodity, especially in cleanrooms. The manufacture and maintenance of these plants where an extremely high level of hygiene is required are extremely costly. Compact components are all the more important as ultimately, the machine needs to be space-saving. Mitsubishi Electric components such as SCARA and articulated arm robots, controllers and servo drives are characterised by their particularly space-saving design and are suitable for flexible applications even when space is restricted. Easy handling enables fast integration, commissioning and adjustment.

One example of a highly compact handling solution came from Robotronic AG. Their required a secondary packaging solution for supplying and packing filled vials of various sizes. The solution needed to be integrated in an existing system with limited available space. The modular design principle of the modular robot technology (MRT) produced by Robotronic provides excellent design flexibility. As a result, the basic module for the MRT cell has a footprint of just 1.0 x 1.30 metres and is approximately 2.20 metres high, so it also meets the minimal space requirements. The solution for the cleanroom class in accordance with GMP standard level D consists of two MRT cells, each with a compact robot from Mitsubishi Electric and a conveyor line with eight positioning screws, driven by Mitsubishi Electric servo motors. The robots place the vials in the blister packs at a processing speed of 300 units per minute.

Hygiene in cleanroom systems
The increasing use of automation technologies, especially robots, has led to an increase in the demand for systems which meet high cleanroom requirements. It is also just as important to be able to clean a plant before a production changeover without any major costs being incurred. That means that it must be possible to clean the components in place (i.e. be CIP-compatible) using aggressive chemicals like H2O2. For that reason, Mitsubishi Electric also offers its customers multi-resistant versions of its new generation of MELFA robots which have been approved for regular CIP cleaning using H2O2. MELFA robots can even meet cleanroom class requirements; ISO 3, are dust proof, and have IP67 environmental protection.

@MitsubishiFAEU #PAuto


Future factory – a moderator’s impression!

01/02/2016

Read-out was asked to moderate the automation stream at the National Manufacturing & Supplies conference held last week outside Dublin. (26th January 2016). In their wisdom the organisers selected “Future Factory!” as a title for this half day seminar and there were 11 speakers organised to speak on their particular subjects for about 15 minutes each. This was replicated in the the over a dozen different seminars held on this one day.

q#MSC16

Long queues lasted well into the morning to enter the event!

We were a little sceptical that this would work but with the help of the organisers and the discipline of the speakers the time targets were achieved. Another target achieved was the number of attendees at the event as well as those who attended this particular seminar.
In all between exhibitors, speakers and visitors well over 3000 packed the venue. Probably far more than the organisers had anticipated and hopefully a potent sign that the economy is again on the upturn. Indeed it was so successful that it was trending (#MSC16) on twitter for most of the day.

Seminar
But back to our seminar. If you google the term Future Factory you get back 207million links, yet it is difficult to find a simple definition as to what it means. The term automation similarly is a very difficult term to define though the term in Irish “uathoibriú” perhaps is a bit clearer literally meaning “self-working.”

uturefactory.jpg

Good attendance at the Seminar

Background
The world of automation has changed to an extrordinary degree and yet in other ways it remains the same. The areas where it has experienced least change is in the areas of sensing – a thermometer is a thermometer – and final control – a valve is a valve. Where it has changed almost to the point of unrecognisability is in that bit in the middle, what one does with the signal from the sensor to activate the final control element.

From single parameter dedicated Indicator/Controller/Recorders in the sixties which transmitted either pnuematically (3-15psi) or electrically (4-20mA). Gradually (relatively speaking) most instruments became electronic, smaller in size and multifunctional. The means of communication changed too and fieldbus communication became more common to intercact with computors which themselves were developing at breaknech speed. Then transmission via wireless became more common and finally the internet and the ability to control a process from the computer that we call the intelligent phone. There are problems with these latter, internet/cellphone, of course. One is that the reach of the internet is focussed at present on areas of high population. Another is the danger of infiltration of systems by hostile or mischivous strangers. The importance of security protocols is one that has only recently been apparent to Automation professionals.

• Many of the presentations are available on-line here. The password is manufac2016

The Presentations
Maria Archer of Ericsson spoke on the enabling and facilitating IoT in the manufacturing industry. Diving straight into topic she drew on her experience of big data, e-commerce, media, cyber security, IOT and connected devices.

The second speaker was Cormac Garvey of Hal Software who addressed Supply Chain prototyping. The Supply Chain ecosystem is incredibly complex, usually requiring significant integration of each suppliers’ standards and processes to the manufacturer’s. Cormac will introduce the concept of supply chain prototyping, where easy-to-use, standards-based technology is used to wireframe out the entire supply chain ecosystem prior to integration, thus significantly reducing cost, time and risk on the project. This wireframe can then be used as a model for future integration projects.

Two speakers from the Tralee Institute of Technology, Dr. Pat Doody and Dr. Daniel Riordan spoke on RFID, IoT, Sensor & Process Automation for Industry 4.0. They explained how IMaR’s (Intelligent Mechatronics and RFID) expertise is delivering for their industrial partners and is available to those aiming to become a part of Industry 4.0.

Smart Manufacturing – the power of actionable data was the topic addressed by Mark Higgins of Fast Technology. He shared his understanding of the acute issues companies face on their journey to Business Excellence and how leveraging IT solutions can elevate the business to a new point on that journey.

Assistant Professor (Mechanical & Manuf. Eng) at TCD, Dr Garret O’Donnell,   explained how one of the most significant initiatives in the last 2 years has been the concept of the 4th industrial revolution promoted by the National Academy for Science and Engineering in Germany- ACATECH, known as Industrie 4.0. (Industrie 4.0 was first used as a term in Germany in 2011).

Another speaker from Fast Technologies, Joe Gallaher, addressed the area of Robotics and how Collaborative Robots are the “Game Changer” in the modern manufacturing facility.

Dr. Hassan Kaghazchi of the University of Limerick and Profibus spoke on PROFINET and Industrie 4.0. Industrial communications systems play a major role in today’s manufacturing systems. The ability to provide connectivity, handle large amount of data, uptime, open standards, safety, and security are the major deciding factors. This presentation shows how PROFINET fits into Industrial Internet of Things (Industrie 4.0).

White Andreetto

Maurice Buckley CEO NSAI

The CEO of NSAI, the Irish National Standards Authority, Maurice Buckley explained how standards and the National Standards Authority of Ireland can help Irish businesses take advantage of the fourth industrial revolution and become more prepared to reap the rewards digitisation can bring.

The next two speakers stressed the impact of low forecast accuracy on the bottom line and how this coulbe be addressed. Jaap Piersma a consultant with SAS UK & Ireland explained that low forecast accuracies on the business performance is high in industry but with the right tools, the right approach and experienced resources you can achieve very significant result and benefits for your business. Following him Dave Clarke, Chief Data Scientist at Asystec, who mantains the company strategy for big data analytics service development for customers. He showed how are incredible business opportunities possible by harnessing the massive data sets generated in the machine to machine and person to machine hyper connected IoT world.

The final speaker David Goodstein, Connected Living Project Director, GSMA, described new form factor mobile SIMs which are robust, remotely manageable which are an essential enabler for applications and services in the connected world.

All in all a very interesting event and useful to attendees. Papers are being collected and should be available shortly on-line.

It is hoped to do it all again next year on 24th January 2017- #MSC17.

See you there.

@NationalMSC #MSC16 #PAuto #IoT


Man or machine? Is HAL taking over?

04/08/2015
As we enter the golden age of robotics, the fear that robots will take human jobs has slowly spread. Jobs such as assembly, farming and surgery are already being delegated to robots. Here, Darren Halford of European Automation considers if our jobs are really at risk.

epa245This speed of technological change has led Google’s director of engineering, Ray Kurzwell, to estimate that robots will, “reach human levels of intelligence by 2029”.

For many, the idea of artificial intelligence surpassing human intellect is a daunting thought. Whilst certainly not the first example of evil artificial intelligence in pop culture, HAL the homicidal computer from 2001: A Space Odyssey is a prime example of the recurring ‘robot uprising’ theme we see depicted in film and literature.

However, even in the far-flung worlds of science fiction, robots have proven to be predominately helpful – just as they have in manufacturing. From traditional six-axis, SCARA and Cartesian robots to Automated Guided Vehicles (AGVs) robots generally create jobs; increasing the overall number of positions available.

One of the automation trends for 2015, AGVs are mobile robots that navigate independently using magnets, lasers, vision and geoguidence and are used commonly in industrial settings to transport materials and goods in a factory or warehouse. AGVs can increase efficiency, ultimately reducing costs and, because of this, their market is growing at a rapid pace.

Unlike automated guided vehicles, more complex, manufacturing robots are usually confined to operating inside cages known as robotic work cells. These physical barriers protect human workers from potential accidents and the sheer power and speed of malfunctioning robotics. If you’ve ever seen a robot at a trade show try to return to zero without back up in the event of power cut, you will know exactly what I mean.

Despite this restriction, advances in programming mean that some robots can now operate, without enclosures, alongside humans on the factory floor. This integration revolutionises the job roles of both robot and human workers. This increases the productivity of menial tasks and frees human workers to focus on jobs that are more sophisticated.

Although still in its early stages, this man-machine collaboration is a huge step towards humans and robots working harmoniously together.

Robot trends
Cell free robotics was a theme at this year’s Hannover Messe, which also featured ABB’s wonderfully cool YuMi, a two-armed collaborative assembly assistant that can see and feel its way around an application. It has soft, padded arms that allow it to interact safely with its human counterparts.

There is no denying that some very menial labour will be replaced with technology. In fact, Deloitte and the University of Oxford predict that robots could ultimately replace ten million unskilled workers.

However, throughout history, technology has created thousands of new jobs while eliminating old ones. Consider the first half of the twentieth century, where a large percentage of working Londoners were limited to work in manufacturing and heavy industry.

Whilst some might argue that IT and communications led to a decline in heavy industry, others would say it freed workers to ‘break the habit’ and pursue a wide range of vocations outside of the factory. This, in turn, established London as the cosmopolitan metropolis and services hub that it is today.

The first robots might have been installed in factories in the 1960s, but we are only now truly entering the golden age of robotics. It will open doors to new industries and generate new roles requiring creativity, judgment, empathy and a thirst for innovation – human skills which robots can’t yet replicate. So it’s not time to worry about HAL and his compatriots just yet; your job is safe.