The robots are coming
Industrial robots can be enormously effective for automating high-value, repetitive, and often unsafe processes, such as handling heavy castings or welding body seams on vehicle production lines. They may also offer advantages to small-and medium-sized companies that need flexible help with a changing roster of process-oriented tasks. The only issue with industrial robots is… using them.
Despite the challenges of doing so, to which we will return shortly, the uptake of robots has accelerated over the past couple of years, especially in the US, where demand for robots is growing as companies invest in automation.
Total US orders for robots were 20% greater in the first quarter of 2021 than in the same quarter in 2020, according to America’s Association for Advancing Automation (A3). Companies working in metals processing spent 86% more on robots, those in life sciences and pharmaceuticals 72% more, those in food and consumer goods 32% more, and the rest of the non-automotive industries spent an aggregate 12% more. Figures from A3 also show that 2020 was the first year in which companies outside the automotive sector spent more on robotics than the car makers.
Speaking on publication of the results in May 2021, Jeff Burnstein, President of A3, said:
“While advances in robot technology, ease of use, and new applications remain key drivers in robot adoption, worker shortages in manufacturing, warehousing, and other industries are a significant factor in the current expansion of robot use. COVID didn’t create the move toward automation, but certainly has accelerated trends that already were underway.”
The industry 4.0 vision
One reason that there is a gap between our view of the value of industrial robots and their actual utility may spring from the Industry 4.0 strategy introduced by the German government in 2011. Industry 4.0 was presented to improve manufacturing by building bridges between the physical world of production lines and the digital world of work scheduling, equipment monitoring, statistical quality control, and predictive maintenance.
Under an Industry 4.0 strategy, factories would evolve inexorably towards ‘smart factories’, in which every physical action on the production lines was matched with real-time data capture, advanced analytics, and action requests handled by cloud-based computers reached over robust communications networks.
This is still a vision of the future for many. A research brief on the status and future of industrial robotics, published by the Interactive Robotics Group at MIT in November 2020, laid out some reasons why this is so.
Industrial robots on a production line
The integration challenge
One of the biggest challenges in using industrial robots is integrating them into production lines. Industrial robots are often large, fast, powerful machines that must be fenced off to protect factory workers, distorting existing workflows.
Industrial robots can be difficult to program: the industry lacks a common language with which to program the movement of robots, each manufacturer tends to have its own user interface, and even the manual controllers differ. This makes programming robots such a specialist skill that it is often outsourced to third-party integrators, whose work can cost more than the robot itself. If the programmed function needs to be changed, for example because a part has changed, companies may need to get the integrators back in to make the adjustment.
Some in the industry have even floated the idea of offering ‘robots as a service’, as a way for companies to outsource the pain of making robots work well for them.
The smart factory challenge
Even companies that embrace industrial robots may be sceptical about other parts of a ‘smart factory’ strategy, such as the Industrial Internet of Things (IIoT), because they're concerned about privacy, security, and keeping control of their data. Factories that do take up the IIoT to monitor robotic production lines need to implement an extremely robust communications network to reach all the distributed sensors and actuators.
For example, when grocery-delivery and fulfilment-centre automation company Ocado wanted to provide communications for its very large robotic warehouses, it installed a private 4G network.
There are also concerns with using cloud computing to gather, store, clean and analyse data for a smart factory. The promise of cloud computing is that it provides processing power on tap, like a utility. The reality is that cloud computing systems can have issues that become mission critical when they are used to manage a production line.
The real elephant in the room, though, is that gathering a lot of data is not the same as creating actionable insights. Senior management may look at the cost of implementing the infrastructure that turns a standalone robot into part of a smart factory and ask, “Where’s the return on my investment?”
The cobot alternative
One alternative way forward is the collaborative robot, or cobot, which enables people and robots to work more closely without endangering worker safety.
Cobots tend to be smaller, lighter robot arms, designed to handle less mass, and move more slowly than standalone robots. Many of these arms include sensing facilities so that they stop moving if they encounter an obstruction – such as a human limb. These can include passive contact sensors, light or laser curtains, proximity sensors within the working area, or even capacitive ‘skins’.
A cobot working with an operator on an assembly line to boost productivity.
This approach, used by Bosch Rexroth in its APAS Production Assistant cobots, can sense people within its working area without making contact, and slow or stop the cobot’s operation. Unsurprisingly, there is an ISO standard, ISO/TS 15066, which defines safety requirements for collaborative robots.
Cobots face similar programming issues as larger robots, but their size and accessibility may make it easier to gain enough familiarity with them to adapt them to new tasks quickly.
Automated guided vehicles
Automated Guided Vehicles (AGVs) are an important subspecies of robot that can navigate factory floors to move material under their own control, using sensors to find their way and avoid obstacles. The good news is that work on developing self-driving cars is advancing the state-of-the-art in key concepts such as simultaneous location and mapping, as well as encouraging the development of more sophisticated sensing technologies such as time-of-flight sensors and LiDAR.
The slightly less good news is that a lack of standards is making it more difficult to manage a fleet of AGVs from multiple manufacturers as a group.
Enabling technologies
As the AGV example illustrates, one of the key enabling technologies for future robotics will be standards that bring a common approach to key development issues such as programming, communications, and co-working.
There are already efforts to achieve this – the open-source Robot Operating System (ROS) is a collection of tools, libraries, and conventions, popular with individuals and academics, to create complex and robust robot behaviours on a wide variety of robotic platforms.
The ROS justifies the project like this: “Why? Because creating truly robust, general-purpose robot software is hard. From the robot's perspective, problems that seem trivial to humans often vary wildly between instances of tasks and environments. Dealing with these variations is so hard that no single individual, laboratory, or institution can hope to do it on their own. As a result, ROS was built from the ground up to encourage collaborative robotics software development.”
The other major category of enabling technology is sensors, which can do everything from monitoring an AGV’s battery status to helping microposition a robot’s manipulator so that it can pick up a component. Such sensors need to be robust, highly accurate, have long operating lifetimes, and produce consistent results in rapidly changing environmental conditions.
For example, the authors of the MIT report found that some sophisticated robot vision systems worked well in lab conditions but failed in the varying lighting of a real production line.
One response to this could be to use infrared time-of-flight sensors, such as OMRON EMC’s B5L sensor (left), to measure distances to objects to create 3D models of their position in space. The sensor is designed to reject the effects of varying ambient light and is protected against mutual interference so that up to 17 of the units can share a working environment.
As previously discussed, though, better sensors are only valuable if they are backed up by a communications, computing and decision-making infrastructure that can absorb, analyse, and act fast enough to deliver some practical advantage in the way that the robot operates.
Conclusion
Industrial robotics has a long history of successful application, especially in the car industry. But it still faces some of the same issues it has always had, such as understanding the position of objects in space, and programming them to manipulate those objects at a distance. Although there are pathways and strategies for wider application of robotics, turning them into a reality will involve sorting out a lot of details on multiple fronts.
If you want to find out more about solutions for industrial robots and vehicles from our leading suppliers, visit our non-automotive transportation page. Alternatively, if you have a question or you would like to discuss your design in more detail, you can reach out to our team of field applications engineers in your local language.
WANT MORE LIKE THIS? SUBSCRIBE TO OUR NEWSLETTERS
About Author
Hagen Goetze
Hagen Goetze, Senior Director Marketing at Avnet Abacus, has leadership of Supplier and Product Mana...
abacus content library/articles/the-robots-are-coming
The Robots Are Coming | Avnet Abacus
