What does the future hold for industrial automation?

UL’s Dr Eoin Hinchy talks about the past, present and future of industrial automation, and how trends such as AI and mechatronics are making their mark.

In order to understand just how far industrial automation has come, Dr Eoin Hinchy says you have to look at how things were made in the past.

Hundreds of years ago, he explains, products were manufactured individually by hand as one-offs. “So while every product might be designed to be identical, the handmade nature resulted in unique differences between products,” he says.

Along came the first industrial revolution, where mechanised production shifted “from cottage industries to loud centralised factories” and power sources such as the steam engine were introduced, leading to higher volume production and improved standardisation across manufacturing.

Each subsequent industrial revolution brought with it some major changes to the manufacturing process. The second (which took place from the late 19th to early 20th century) introduced new organisational principles, explains Hinchy, while the widespread adoption of electricity transformed factories, allowing for cleaner, more flexible layouts and continuous operation.

Emerging during the latter half of the 20th century, the third revolution marked the shift from analogue mechanical and electronic technology to digital technology, leading to the rise of information technology and operational technology.

According to Hinchy, an associate professor in digital manufacturing and control at the University of Limerick (UL), many of today’s manufacturing facilities are still in the third industrial revolution stage. However, in 2011, along came its successor – industry 4.0.

Industry 4.0 combines cyber-physical systems, the internet of things (IoT), cloud computing and artificial intelligence (AI) into the concept of smart factories, where machines, sensors and products “communicate and cooperate with each other in real time”.

“While there have been many research articles published regarding these technologies, their implementation in factories across the world is limited,” adds Hinchy.

And although industry 4.0 has not been fully realised yet, talk of industry 5.0 has been ongoing for the past few years. “The EU introduced the concept of industry 5.0 in 2021, as a vision that complements and extends industry 4.0 rather than replacing it,” explains Hinchy. “Industry 5.0 emphasises a transition towards a more human-centric, sustainable and resilient industry – and thus, this will be the future of industrial automation.”

A combined effort

In this transitionary period between industry 4.0 and 5.0, Hinchy says that a number of modern technologies are shaping the industrial automation industry.

Advanced sensors, IoT, 3D printing, AI, as well as cloud and edge computing, are some of the technologies driving the sector, according to Hinchy. But one concept stands out as a driving force for the industry.

“Consider a traditional robotic arm on a manufacturing line. These machines are incredibly fast, accurate and repeatable – the trouble is that they can’t see what is in front of them,” he says. “This means that if they are doing a pick-and-place operation, for example putting parts from a conveyor into a box, if the part they are picking is not in the correct place, they don’t know.

“Humans, on the other hand, have excellent awareness, so if they are doing the same pick-and-place operation, they can clearly see where the parts to be picked are. However, repetitive pick-and-place operations for humans are laborious, tiring and can lead to ergonomic strain.”

This, according to Hinchy, is where human-robot collaboration systems come in.

“What if we could develop a system which has the repetitive excellence of a robot arm, with the ability to contextualise the real world as well as humans can?” he proposes.

“Advanced sensing, such as high-resolution vision systems, Lidar and 3D time of flight, coupled with AI and edge-based processing, are enabling the accurate detection of humans on the factory floor,” he says. “These systems can then be coupled with collaborative robotics to develop human-robot collaborative manufacturing systems.

“Such systems let robots and humans work safely together, allowing robots to conduct operations like lifting heavy objects during assembly operations, while humans can do complex tasks like lining bolts into holes.”

Mechatronics

One concept that’s gaining traction in industrial automation is mechatronics. As course director for UL’s mechatronics master’s degree – which is accredited by Engineers Ireland – Hinchy is more than familiar with the topic.

Mechatronics, or mechatronic engineering, is a multidisciplinary topic, according to Hinchy. The word ‘mechatronics’ comes from a blend of mechanical and electronic engineering.

“There is much more to mechatronics than just mechanical and electronic engineering, however,” says Hinchy.

Mechatronics incorporates elements of industrial automation, control engineering, computer science, programming and data analytics. More recently, Hinchy says, AI and machine learning have also entered the field.

“Think of it like automation engineering,” he says.

According to Hinchy, the real value of mechatronics in the industrial automation sector is that it provides graduates with the “core capabilities for developing automated systems”.

“While mechanical engineering focuses on the physical structure and motion of machines, and electronic engineering on circuitry and signal processing, mechatronics distinguishes itself by its holistic, integrative approach to systems and control,” he explains.

“Rather than specialising in isolated components, mechatronics engineers view automated systems as complex, interconnected entities, blending mechanical design, electronic hardware, software control and potentially intelligent algorithms to achieve precise functionality.”

What lies ahead?

Pondering the future of industrial automation, a few trending concepts come to mind for Hinchy such as digital twins, which he says can be utilised to instantly identify bottlenecks, inefficiencies and deviations from optimal performance thanks to the tech’s collection of real-time data.

Most of all though, as with many other sectors, Hinchy says that AI and machine learning are the dominant trends that are “creeping into” industrial automation.

“For example, machine learning, a subset of AI, is excellent in detecting patterns in large datasets,” he says. “This is particularly useful for high-volume production, where large volumes of data are generated. Such data can be used for process predictions and optimisation – this is one area of my research.

“Without doubt developments in AI and machine learning are going to put their stamp on the sector. We are seeing this already beginning in some manufacturing facilities and this is likely to grow further as the AI boom continues. This could be in the guise of AI-edge processing or vision systems or system optimisation.”

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.