Mobile-Menu

Robotics "Humanoid Robotics Must Be Approached from a Task-Oriented Perspective."

Source: Press Release | Translated by AI 4 min Reading Time

Related Vendors

Fraunhofer IEM
FAULHABER Drive Systems
Taiwan Machine Tool & Accessory Builders' Association (TMBA)
SIMTOS (Seoul International Machine Tool Show)

For Sebastian Prinz, head of the Robotics and Automation department at Fraunhofer IEM, training and data currently pose the greatest challenges for the industrial deployment of humanoid robots. He knows which applications are promising and why companies should start exploring this technology today.

“Find a job that the humanoid can actually do, and understand your processes”—that’s the advice robotics expert Sebastian Prinz gives companies so they can benefit from the technology in the long term.(Source: Fraunhofer IEM / Janosch Gruschczyk)
“Find a job that the humanoid can actually do, and understand your processes”—that’s the advice robotics expert Sebastian Prinz gives companies so they can benefit from the technology in the long term.
(Source: Fraunhofer IEM / Janosch Gruschczyk)

Why is everyone suddenly talking about humanoid robots?

Sebastian Prinz: Humanoid robots produce spectacular images, even if, in my opinion, a little help was involved. However, they illustrate the vision behind the technology very well. 

Sebastian Prinz is the head of the Robotics and Automation Department at the Fraunhofer Institute for Design Engineering and Mechatronics IEM.(Source: Fraunhofer IEM)
Sebastian Prinz is the head of the Robotics and Automation Department at the Fraunhofer Institute for Design Engineering and Mechatronics IEM.
(Source: Fraunhofer IEM)

Many company representatives come to us with images of dancing and gymnastics-performing humanoids in their minds. A machine with these human-like abilities offers—when applied to industrial tasks—an enormous value proposition. We then say, “Yes, but it really depends on the task at hand.”

Let’s be honest: Who isn’t fascinated by robots that can run faster than Usain Bolt? That grabs people’s attention and helps manufacturers, investors, and buyers get excited. For industrial applications, however, other capabilities—such as precision, payload capacity, reliability, robustness, and safety—are usually the deciding factors.

Behind the manufacturers’ marketing, however, lie truly enormous technological advances. In recent years, there has been massive investment in AI, robotics, and hardware. This is giving rise to capabilities that were barely conceivable just a few years ago—and that will deliver very tangible added value in the future.

How far has humanoid robotics actually come in industry today?

Although the technology is developing rapidly, its industrial application is still in its infancy. According to Bitkom, 97 percent of the industrial companies surveyed can envision humanoid robots being used in production in the long term. In fact, 6 percent are currently using such systems, primarily in pilot projects and test runs.

This is also confirmed by the activities of large companies. For example, Siemens and BMW are already testing humanoid robots in production environments. We are still a long way from widespread productive use.

Why aren't humanoid robots already everywhere in factories?

Many discussions about the limitations of humanoid robots revolve around battery life, the lifespan of the drive system, or fine motor skills and precision. However, technologies in these areas are advancing rapidly. In practice, one issue stands out above all others: the training effort required. A person often needs only a brief explanation. If a red part belongs in the left box and a green one in the right, they understand the task immediately. A humanoid robot, on the other hand, must learn every single step: recognizing the part, grasping it, placing it down, and deciding between left and right.

This training requires large amounts of data. That’s why manufacturers around the world are currently collecting data from everyday and work situations. Many impressive demonstrations are the result of an enormous amount of training. You usually can’t tell from the videos just how many repetitions, adjustments, or training runs went into what appears to be a simple task. Furthermore, the real world rarely behaves as orderly as a training scenario. Even small changes in the environment, materials, or procedures can necessitate adjustments.

For which applications are humanoid robots worthwhile?

The strength of humanoid robots lies in their ability to work in environments that were originally designed for humans. This means that machines, tools, and workstations do not need to be completely redesigned. Potential applications include, for example, material handling, machine operation, joining processes, assembly tasks, and inspections. Tasks that have so far been difficult or uneconomical to automate using traditional automation methods are of particular interest.

At the same time, the future will likely be more diverse. After all, humanoid robotics is not an end in itself. It depends on the task at hand. Sometimes a humanoid is the right choice; other times, a cobot or a traditional industrial robot is better. This decision should always be based on the specific application. Systems with mobile platforms instead of legs are already being developed today. That’s why, in the future, there will be versions of humanoids optimized for different use cases—whether for high payloads, fast movements, high precision, or limited installation spaces. Depending on the application, different humanoid characteristics may be of interest—such as hands with fine motor skills, versatile arms, or even all-terrain chassis.

Subscribe to the newsletter now

Don't Miss out on Our Best Content

By clicking on „Subscribe to Newsletter“ I agree to the processing and use of my data according to the consent form (please expand for details) and accept the Terms of Use. For more information, please see our Privacy Policy. The consent declaration relates, among other things, to the sending of editorial newsletters by email and to data matching for marketing purposes with selected advertising partners (e.g., LinkedIn, Google, Meta).

Unfold for details of your consent

Why is it worthwhile for companies to start exploring humanoid robotics today? 

I am convinced that, within three to five years, humanoid systems will be part of everyday production at some companies. However, those who wait until then to address this issue will first have to deal with integration, infrastructure, acceptance, security, data protection, and so on. It makes sense to tackle all these tasks today.

The first step here is not to rush into purchasing a humanoid robot. It is far more important to focus on the planned task and lay the groundwork—which initially has little to do with hardware. This includes process knowledge, data, and an understanding of your own use cases. This foundation isn’t built overnight, but preparations can begin today.

My advice to companies: First, understand your processes. Find a job that a humanoid robot can do really well. Collect training data. Only then should you buy a robot.

Related content