Adaptive Systems: Is the corona crisis taking robotics to a new level?

Already before corona, but at the latest now: A robotics expert explains why adaptive robots play an important role in times of crisis and even after them.

Could German industry have been better prepared, or even should it have? “The coronavirus and the Covid-19 crisis are affecting exactly those places where digital, highly automated production is operating,” Ronnie Vuine stated, CEO of Micropsi Industries, a Berlin-based robotics start-up. As a result, it is in the very best company – as Klaus Helmrich, Member of the Board of Management of Siemens, wrote in a LinkedIn article: “The benefits of digitization and automation for the manufacturing industry are particularly clear in the face of the pandemic, because automated and digital solutions enable technological innovations that can be employed in a targeted manner to counter the effects of the virus.” Therefore, corona is also a wake-up call in Mr. Vuine’s opinion. “People have been talking about Industry 4.0 and digitization for many years. This was always considered more or less a topic for the future, which you could do if you had time and then use it a bit as an embellishment. Now, however, we are in a situation where all those who neglected this say, ‘Oh no, if I had only’…"

What is the status of automation in Germany's factories?

More highly automated production – “co-worker” robots are especially well suited for this. They are not contagious, because “if production is to be maintained during a pandemic, certain distances between workers must be maintained,” Patrick Schwarzkopf stated, Managing Director of the VDMA Robotics + Automation Association, explaining the advantages of the technology in current times. Admittedly, no one could have expected a health crisis of this magnitude in Germany before the COVID‑19 pandemic. Small or medium-sized companies in particular are not developing under the same economic conditions as a key player from the automotive industry.

In many cases, the necessary business cases simply did not exist either, Mr. Vuine argued. “Not everyone has half a million on hand and the luxury of predictability beyond twelve or eighteen months,” among other things for the integration of an automation solution, as the entrepreneur is well aware of. Consequently, the costs had to be reduced first: for the components and also for engineering. In addition, many companies simply lacked – or still lack – the know-how for meaningful automation of production. Mr. Vuine knows: “For a long time, automation was considered an expensive expert topic that was outsourced to system integrators – and rightly so. After all, the consequences of incorrect system integration can be severe and can even threaten the existence of a company.”

Adaptive robots instead of specialized machines

But there are also solutions for smaller companies – without the need for machinery specifically tailored to their products: “You need flexible automation that can quickly adapt to new situations,” Mr. Vuine explained. "And this can only be done if one accepts that the use of automation technology means the use of software to a large extent – because it can be changed quickly. Instead of using specialized machines, it is better to work with robots that can be reprogrammed and that can ideally learn something new quickly.” With adaptive systems, even small and medium-sized companies can benefit from automation – even after corona.

For this purpose, Micropsi Industries is developing real-time, sensor-driven robot controllers, which should make robots capable of learning and adaptive. The Mirai robot controller, which can be trained, differs from conventional robot systems in one important point: The robot is no longer programmed for its abilities, but is taught by humans. Mr. Vuine explained: “You do this for two or three afternoons; you train the robot on hand-eye-coordinated skills, similar to sports or playing an instrument. This works by watching the robot through a camera that is installed on your wrist or on a mounting table, then looking at the new motion data in the cloud, and afterward the robot can do it.”

Mr. Vuine explained the typical application scenario of his AI-driven robot automation: "We are not replacing conventional robot programming at all – whether it is writing codes or showing poses, as is already done with cobots.” In this area, there are already sufficiently simplified programming interfaces from KUKA, ABB, Fanuc or also with cobots from Universal Robots, as Mr. Vuine knows. He explained in more detail: “Large movements through open space are relatively easy for robots to make. It always gets exciting when they have to make contact with the world – for example, with a human being or a work piece – and do not move through open spaces. The conventional task: A workpiece has to be picked up that is not always in the same place. This part, where the robot has to react when something unexpected happens, is taken over by adaptive robot systems, so to speak, we are then controlling the robot.”

How robots learn via a pen

The Dresden robotics start-up Wandelbots has developed a new solution for teaching robots. Instead of complicated programming of the systems, laymen should now be able to assign tasks to the robot with a wireless pen, called “TracePen”, easily and without any programming knowledge – remotely controlled wherever the user is. Especially in corona times, “lights-out production” enables critical production processes to run completely without human interaction. The user simply performs the work of deburring, inspecting or gluing with the pen in his hand by way of example. The intelligent software then converts the movements into automation scripts for the robot. According to the company, this not only solves the problem of lack of expertise, but also reduces the costs for companies, since software alone currently accounts for 70 percent of the life cycle costs of a robot. The first delivery of the solution began in August 2020.

For production managers, this means: In the future, it will be a question of a new variance in automation and of movements that have so far been reserved for humans. “Put a piece of plastic on here, pull out a cable there, everything that is done during assembly work”—and where people previously had to measure in time-consuming work to know exactly where something was before a person could program a robot so that it could precisely view the measurement and only then react to it. “Robots have not yet been able to do this,” Mr. Vuine explained. Acceptance by the employees is not a challenge for him, but instead the abilities of robotics to map the handling of such variances in sufficient quality and process reliability.

Robotics in times of crisis: Learning from corona for exceptional situations in the future

Mr. Vuine described what this means for the current crisis and also for future exceptional situations using the example of the protective mask bottleneck. “How could it be that an industrial nation like Germany cannot just start up production for paper masks for a short time? The reason is automation! Paper requires special machines, and it takes time to set them up.” To emerge strengthened from what is probably the biggest crisis since the Second World War, Mr. Vuine has a tip: try it out. “It already works a lot more often that you simply buy a robot and look for a 20-year-old employee who wants to try it out. Then you let him experiment with it for a few weeks, and an automation solution is found. Such did not exist in the old world; it's completely new.”

Therefore, Micropsi Industries lends systems to companies in the current situation, for example, in the metal, plastic or electronics industry. "We believe that now is the time when people have some time to think about new things, because they cannot produce in part because they do not receive material, and it is the time when people can think about the next cycle of their manufacturing and how they can save money in the future and make their production more robust for what is still to come. Once the awareness of a life with change and the associated demands for flexibility in production has been created, supply chains will be more secure again for a long time.