From ‘I, Robot’ to YuMi: the inspiring world of robotics
20 June 2018
Isaac Asimov science fiction writer with fictional robot (Credit: Shutterstock)
Ever read Isaac Asimov’s book ‘I, Robot’? Mike Wilson, Business Development Manager, UK & IE, ABB Robotics has! Understand how it was this dystopian novel that first sparked Mike’s interest in robotics and how, since then, he’s gone on to pioneer the use of robots in a number of applications and still continues to strongly argue that they won’t destroy the world of work but play a positive role in the future of humankind.
How did you first become interested in robots?
My first encounter with robots came when I read Isaac Asimov’s book, ‘I, Robot’, a collection of short stories which focused around the interaction between humans and robots and the ideas around morality and robotics. I liked the optimism in Asimov’s work and the idea that robots could play a positive part in the future of humankind, with humans and robots able to benefit from one another – this is something I still believe in and use as an argument against those who see robotic automation as a threat.
How did you take your first steps into the world of robotic automation?
As my father was an engineer, engineering was something I grew up with and was interested in. As a result, I ended up studying for a B.Sc in Engineering Science at the University of Warwick and was thrilled to then be able to follow up on my interest in robots by becoming part of the first year’s intake of Cranfield University’s M.Sc in Industrial Robotics, which was the first course of its kind to be run in the UK.
The course gave me a good grounding in the principles of applying robotic technology to industrial processes. It gave me the opportunity to program some of the earliest industrial robots, including an early Asea IRB60 robot (a forerunner of today’s ABB’s robots); a Little Giant from a UK robot company called Hall Automation; and a Unimation Puma robot.
After that, it was a case of trying to then find opportunities to put what I’d learned into practice, which wasn’t easy back in 1982 as robotics was still very much in its infancy. Fortunately, British Leyland, which subsequently became the Rover Group, was starting to look for ways to introduce robots into its manufacturing processes and I was taken on as an engineer as part of a dedicated group focused on advanced automation projects.
In the six years I was there, I helped to pioneer the use of robots in a growing range of applications, including adhesives, welding and painting. Compared to what’s possible now, a lot of how we programmed and configured our robots was very belt and braces, with lots of trial and error and with programming and set up often taking several weeks, depending on the application.
A good example of this is a project I was involved in which entailed setting up and refining a Unimation Puma robot to apply adhesive onto the bonnet of a Triumph Acclaim. The robot needed to be mounted upside down over the cell and was suspended from the factory roof.
Once the robot had been installed and put into operation, we started to receive complaints from the end of the production line by end of first day that the adhesive was being applied in the wrong place, causing the bonnet to rattle. Naturally, as the obvious variable in the process, the robot got blamed.
On investigation, it transpired that the actual cause of the problem was not the robot, but the expansion of the factory roof as it heated up during the day, with the problem becoming most pronounced at around midday. We installed a gantry, located on the floor, and mounted the robot on this instead of the roof, which immediately solved the problem.
It’s important to remember that while nowadays you can set up and test a robot using offline software, back then everything still had to be manually measured, tested and refined, with the resulting knowledge then being used to create the robot program.
A good example was when we had to set up a hydraulic painting robot on British Leyland’s Freight Rover line to paint the inside of their vans. The job involved two people, myself as the programmer and a colleague who specialised in painting and would be able to advice on the best way to set up the robot.
Together, our task was to teach the robot the path using a lead teach technique, with the data being relayed to a first-generation robot controller. The robot itself was mounted on the end of a boom. The process of painting a van involved reversing it over the robot, which then applied the paint, before then moving it back out again on a conveyor, with the robot continuing to apply paint.
The process of teaching the robot to do this task, which lasted a total of 12 minutes, took almost a year to perfect. There were times when it was incredibly frustrating. As well as having to keep swapping around the robot and its teaching arm, the limited memory of the robot controller meant any mistakes could not be edited. Instead, if a mistake was made, the whole process had to be repeated until you got it right. Worse still, the different variants in the Freight Rover line meant that separate programs had to be developed and refined, requiring many hours of crawling around inside vans making painstaking adjustments to the robot’s positions to make sure we got the painting process just right!
Once we’d got them running properly though, the robots ran for around 10 years. Interestingly, having moved on from British Leyland to another company, I was also involved in replacing them with electrically driven robots. In a massive contrast to the time the original robots had taken to install and set up, the new robots were installed, programmed and operational within just three weeks, which shows how much the technology had moved on in the space of just a decade.
What do you think have been the biggest developments in robotics since you started?
Since I first started working with robots in the early 1980s, robotic technology has experienced massive leaps forward that have totally transformed how and where they can be used. One of the biggest gamechangers has been the switch from hydraulics to electric drives, which has enabled robots to be used for an expanded range of tasks requiring precise and accurate movement such as welding and finishing. Although initially limited in terms of their payload, electrically-driven industrial robots have gone on to dominate the market, thanks to developments in motor technology that have substantially increased their load capacity.
Coupled with this have been massive improvements in robot control technology. ABB’s current generation of robot controllers, for example, can do things that I could only have dreamed of back when I first started in robotics. Examples include the use of technology that can accurately control the precision of the robot regardless of its operating speed. The controllers can also control up to 36 axes of movement, enabling users to coordinate their robot with external equipment such as work-piece positioners and track-motion devices or even run multiple robots.
It’s also now much easier to program and use a robot, with even the least experienced user able to be trained in basic robot operation in a matter of hours.
Today’s robots can also be used in ways that were not possible back in the early 1980s. Thanks to developments in technology, the barriers between robots and humans have quite literally been removed to enable unprecedented levels of collaboration. ABB’s YuMi robot is a great example of this. Featuring extensive safety measures, YuMi can work safely alongside human workers, helping them to improve their productivity by handling difficult or monotonous tasks.
In your experience, what are the biggest misconceptions about robots?
One of the biggest misconceptions about robots is the belief they are new technology, even though the world’s first ever production robot – the Unimate – started its first day at work on a General Motors production line in 1961. If that robot was a human, it would be retired by now!
In the 57 years since that time, the capabilities of robots have increased exponentially, as has the number of robots in active service. Today there is something like 2.6M robots in use worldwide, with plenty of evidence to show the benefits they can bring.
Despite this, the rate of adoption of robots in the UK continues to trail behind that of countries where companies have taken the plunge and found ways to use robots in their production processes.
A lot of the reason for this is that people believe they lack the experience and training needed to make a success out of using robots. As a result, they prefer to stick with what they know rather than taking a perceived risk and investing in new technologies that could help them achieve more.
The danger is that the longer this situation persists, the more we get left behind. You only have to look at the staggering 87,000 robots installed in China last year to see how quickly the world is moving towards manufacturing automation.
Introducing robots to the factory floor doesn’t have to be difficult. In my experience, a lot of companies tend to be over-ambitious in their expectations, not necessarily of automation, but of their own ability to handle the change needed to bring it about.
Instead, it’s often better to start small and work on the things that are easy to automate first. Once you’ve got that experience under your belt, it’s much easier to move on to the bigger things.
Another misconception is that putting robots into factories somehow heralds the end of human employment. This is more of a sensationalist headline created by the media than a reflection of reality.
There are plenty of examples to suggest that the best outcomes for companies happen where robots are used to enhance the productive capabilities of human workers. This may either be directly, in terms of the workers using the robots to help them do their jobs better; or indirectly, where improved product throughput and/or shorter delivery times lead to expansion and there’s a need to take on more staff to handle areas such as logistics and distribution or to process orders.
While it is true that robots are now handling many of the jobs that were performed by humans, a lot of the tasks they are used for are those that are unpleasant, arduous or repetitive. The question really is why use people like robots when you can use robots to do the things that people shouldn’t be doing. It makes no sense for an advanced economy to be using people to shift boxes around or spend all day, every day, doing the same thing.
With the average hourly wage for factory workers in the UK starting at around £8.01 (source: Indeed.co.uk), it makes more sense instead for companies to use staff to handle higher level tasks that make use of their skills and attributes. In my own experience of visiting factories, improvements almost always stem from people using their initiative and finding new ways to do things better or more quickly, which is something robots cannot do.
Even where robots are used for more skilled roles, this often tends to be because of a shortage in workers to perform those roles. In the UK, for example, estimates point to a serious shortage of skilled welders, with companies desperate to find workers to fulfil their requirements. The danger is that if companies cannot perform welding operations due to a lack of skilled workers, then they will need to either move their operations to somewhere where they can be found or shut down completely. In either scenario, more people are likely to lose their jobs than would do so if robots were introduced to take up the slack.
What do you think Isaac Asimov would make of today’s robots?
In terms of technology, I think he would be excited to see how far things have moved on and about the range of things that today’s robots are able to do. I think he would also be pleased to see how many robots are being used worldwide, both in our factories and, increasingly, in our homes.
However, I think Asimov would also be disappointed by the pessimistic view that a lot of people have about robots. Despite the word ‘robot’ originating from the Czech word for slave, there seems to be a general sense of resignation that robots taking over the world is a done deal and that it’s just a matter of time until human employment either disappears or diminishes to a handful of roles.
In my opinion, this is a bleak view of the future which misses out on the potential positives that could be achieved by using robots to do the jobs that humans either don’t want or shouldn’t be doing.
Mike Wilson, Business Development Manager, UK & IE, ABB Robotics
Do you agree with those who say that the rise of robots will destroy the future of work?
I don’t believe that robots will destroy the future of work, but I do think that it will change it and that this is something we need to be preparing ourselves for.
Part of the problem is that we are living in a rapidly changing present equipping ourselves for an unfamiliar but exciting future, but still using ideals and approaches from the past.
Our expectations of work and our definitions of jobs are based on increasingly outdated assumptions and models that need to be challenged. Organisational hierarchies built on seniority for example, should arguably be providing greater opportunities for younger staff with awareness of how to use and apply technology to make their voices heard.
Sometimes it actually takes an inexperienced eye to point out new areas for innovation that can easily be missed by those more focused on doing things a particular way.
We recently had the heads of a start-up company presenting at one of our Switch to Robots events at the ABB Robotics training centre in Milton Keynes. Both in their early 20s, they have started a company using a combination of technologies, including an ABB robot and a 3D printer to produce bespoke confectionery. When asked why they had decided to use the technology, their response was ‘Why wouldn’t we? This is technology we have been brought up with’. They made a very valid point that highlights the importance that combining innovation and technology can deliver – of all of the major confectionery producers worldwide, not one is doing anything similar.
As for jobs disappearing and the financial implications this would bring, common sense dictates there is little to be gained by having automated factories churning out thousands of products if there is no-one with the means of buying them. In the same way that nature abhors a vacuum, capitalism abhors not having a market. The very nature of capitalism relies on maximising profits. Maximising profits only works if you have customers. Using robots to save money on employing people is therefore a bit like robbing Peter to pay Paul, where any short-term focus on profit will end up causing the collapse of the very markets you need to stay profitable.
If robotic automation and wider developments such as machine learning and Artificial Intelligence (AI) are going to be part of the future, then we need to be learning to prepare for them and to work with them as tools. To this end, everyone from children in classrooms through to workers in factories need to be educated and trained about how to use the technology, in the same way as people in the past were taught how to use physical tools.
The creation of initiatives such as University Technical Colleges is a step in the right direction to creating and preparing the next generation of technology-savvy workers with the skills needed to thrive in the factories and organisations of the future.
In short, I believe that the future is bright both for people and robots, but that more needs to happen if we are to fully maximise the opportunities that technology can bring.
Mike Wilson M.Sc B.Sc
Mike Wilson has over 35 years of experience in the application of automation to manufacturing across a broad range of industry sectors throughout Europe, Asia and America. Previously he has worked for both users and suppliers of automation and also as an independent consultant providing automation expertise to industry and also training via Warwick Manufacturing Group. He is widely recognised as one of the leading authorities on the application of industrial robot systems in the UK.
Mike is Business Development Manager for ABB Robotics leading the drive to increase the use of robots across UK manufacturing. In addition, he is Chairman of the British Automation and Robot Association, Director of the Processing and Packaging Machinery Association and Board member of the Engineering and Machinery Alliance. He has previously been Chairman of the International Federation of Robotics from 2000 to 2003. His book “Implementation of Robot Systems” was published in 2014.
Contact Details and Archive...