ISO10218 Update Roberta Nelson Shea full interview

At the International Robot Safety Conference, Roberta Nelson Shea, Global Technical Compliance Officer for Universal Robots, shares insights on the latest revisions to ISO 10218-1 and 10218-2. She explains why new, explicit requirements—such as mechanical strength and external axis safety functions—are essential for clarity and compliance. Roberta also discusses the challenge of language in global standards, the importance of application layout in risk reduction, and why evolving use cases drive continual updates. Get a preview of what to expect at Automate 2025 in Detroit and learn why safety technology advances are more evolutionary than revolutionary.

Video Transcription

Interviewer: Roberta Nelson Shea, thank you so much for talking to me again today. It’s great to see you back here at the International Robot Safety Conference. We just listened to a panel where you were talking about some important new safety updates. Can you tell us a little bit about what we can expect in terms of regulatory changes coming for robotic safety?

Roberta Nelson Shea: We have a new revision to our international standard, ISO 10218-1 and ISO 10218-2. These are international standards, which is great because we have broad global acceptance. In the U.S., they are adopted nationally under the cover of RIA R15.06, which has built a strong reputation over the years. There are a lot of changes, but not necessarily a lot of huge changes. That may sound contradictory, but much of it has to do with making requirements more explicitly stated, rather than leaving them to inference. With explicit statements, the standards have more pages, more content.

For example, there’s now an explicit requirement for mechanical strength. I don’t know of any robot manufacturer that would have ever built a robot without sufficient mechanical strength—it’s always been met in practice because otherwise the robot wouldn’t work. But now, the requirement is clearly stated, not just assumed. When you take a robot and integrate it into an application with other equipment, all of the machinery needs sufficient mechanical strength for its intended function. That wasn’t stated before, even though it was typically done. Now, it’s written down.

When I compared page counts to see what really changed, each edition had about 20–30 more pages, but the overall size of the documents increased much more. That’s due to additional figures, informative annexes (which aren’t requirements but are helpful), and more detailed guidance. The new documents are more explicit and should provide better clarity—but, as with anything new, it means learning what’s changed and determining what you need to update.

Interviewer: That really speaks to why standards aim to remove ambiguity. One thing you mentioned in your talk was the word perspicuous. I love that because it shows the attention to detail in word choice.

Roberta Nelson Shea: That word actually came from our Japanese colleagues. It’s not just about concise English—it’s also about how terms translate. In American English, we often use multiple words or sentences that mean the same thing. When you translate them, the meaning can get lost.

Originally, the intended word was “transparent,” as in “clear and easy to understand,” but in translation, that became “see-through,” which was not the meaning. Perspicuous means “clearly understood,” and it captures the intended meaning in one word. None of us had really used the word before, but now we have.

We try to be this thoughtful with wording whenever someone raises a concern. But with so many pages, reading fatigue is real. When you’re reviewing clause after clause, you can start to skim. I’m sure we’ll find things to improve shortly after publishing, but for now, the revisions are complete—until the next update.

Interviewer: Are there any specific new requirements you’d highlight?

Roberta Nelson Shea: Yes. One example is about external axes in applications. If an external axis is controlled by the robot, many end users assume that setting a speed or axis limit in the robot also applies to that external device. It doesn’t—unless it’s specifically implemented.

We’ve added requirements for safety functions with these external axes when they’re under robot control. It’s only a few lines in the standard, but it’s a big deal. Robot manufacturers, integrators, and external axis providers will all need to account for this. It’s the right thing to do because many assumed it was already covered—but it wasn’t.

Interviewer: That really underscores why updates are necessary—new use cases keep emerging.

Roberta Nelson Shea: Exactly. In Part 2 of the standard, which covers integration and applications, we can’t be overly specific because the use cases are so diverse—welding, inspection, pick-and-place, assembly, measurement, and more. Each has different risks.

One point I encourage people to think about from the start is the layout of the application. Don’t just strive for symmetry—placing the robot in the center and surrounding it with equipment. Instead, group things for the shortest distances, best throughput, and easier programming.

If operator intervention is expected, design so that people can work at a safer distance, reducing risk.

Interviewer: One thing I love about this conference is how engaged the audience is. I’ve seen people interrupt presentations with great questions, and others in the audience, including Engelberger Award winners, add to the discussion.

Roberta Nelson Shea: That’s one of the reasons IRSC is unique. I don’t know of any other conference solely dedicated to automation safety. Other safety conferences often have a sales focus—people presenting their products. Here, it’s about sharing knowledge, giving case studies, and explaining what the standards actually say versus what people may have heard.

Everyone here is genuinely interested. In robotics, there’s always public concern—people imagining “robots gone wild.” I’ve joked in the past that we’re not going to see robots running off to Daytona Beach for spring break. What we are focused on is using robots effectively while keeping people safe.

The attendee mix is also unique. Yes, there are EHS professionals, but there are also many controls engineers—people designing circuits, working in CAD, and figuring out how everything integrates. There’s nothing else like it.

Interviewer: Before we wrap up, let’s talk about Automate 2025 in Detroit. What can attendees expect in terms of safety technology and compliance practices?

Roberta Nelson Shea: I don’t expect to see anything absolutely earth-shattering in safety products this year. Safety technology evolves—it’s usually evolutionary, not revolutionary.

There are newer things like radio-based safety sensors and more advanced 3D sensors, but these have already been discussed in recent years. At Automate, you’ll see more examples of these technologies being used in different ways and integrated into booths.

Some exhibitors may discuss the new standards, though adoption will be gradual. Compliance often becomes urgent only when a deadline is real, and right now, companies are still in the early stages of adjusting.

Interviewer: Roberta, thank you again for sharing your insights on safety.

Roberta Nelson Shea: Thank you—it’s been very nice talking with you.


June 23-24, 2026 | McCormick Place

The 2026 Humanoid Robot Forum is happening at Automate in Chicago!

Join industry leaders exploring the technologies, safety, and real-world potential shaping humanoid robotics.

MEET THE AUTHOR

Association for Advancing Automation

Discover how Association for Advancing Automation can support your automation journey with their complete range of solutions and expertise.

Visit Company Website