보스턴 다이내믹스의 새 주인 현대차가 보인 두개의 로봇 VIDEO: Hyundai Motor Group Introduces Two New Robots

Hyundai Motor Group Introduces Two New Robots

The new owner of Boston Dynamics has some interesting ideas about what robots should be doing

By Evan Ackerman

Over the past few weeks, we’ve seen a couple of new robots from Hyundai Motor Group. This is a couple more robots than I think I’ve seen from Hyundai Motor Group, like, ever. We’re particularly interested in them right now mostly because Hyundai Motor Group are the new owners of Boston Dynamics, and so far, these robots represent one of the most explicit indications we’ve got about exactly what Hyundai Motor Group wants their robots to be doing.

Photo: Hyundai

Hyundai Motor Group's "Advanced Humanoid Robot" DAL-e.

   지난 몇 주 동안, 우리는 현대자동차 그룹의 새로운 로봇 두 개를 보았다. 이것은 제가 현대 자동차 그룹에서 본 것 보다 몇 가지 더 많은 로봇이다. 우리는 특히 지금 그들에게 관심이 있다. 왜냐하면 현대 자동차 그룹이 보스턴 다이내믹스의 새로운 소유주이기 때문이다. 그리고 지금까지 이 로봇들은 현대 자동차 그룹이 그들의 로봇들이 정확히 무엇을 하기를 원하는지에 대해 우리가 알고 있는 가장 명백한 증거들 중 하나이기 때문이다.

그럼 현대자동차그룹이 최근 어떤 성과를 내고 있는지 한번 살펴보자. 이 첫 번째 로봇은 DAL-e로 HMG는 "고급 휴머노이드 로봇"이라고 부른다.

현대차에 따르면 DAL-e는 '자동화 고객 서비스의 미래 개척을 위해 설계됐다'며 '얼굴 인식을 위한 최첨단 인공지능 기술은 물론 언어 이해 플랫폼 기반의 자동 통신 시스템'을 탑재했다. 차 전시실에서는 볼 수 있지만, 지금은 서울에서만 볼 수 있다.

DAL-e는 확실히 아무 문제가 없다. 적당히 귀엽고 적당히 기능적인 것 같다. Pepper와 같은 다른 플랫폼도 비슷한 역할을 하는 것을 보았으며, 이러한 더 높은 로봇의 장기적 비용 효율성은 다소 제한적인 경향이 있다는 인상을 받았다. 그리고 우리가 알지 못하는 숨겨진 기능들이 없다면, 이 로봇은 겉으로 드러나지 않는 것처럼 보인다. 

황기철 콘페이퍼 에디터

Ki Chul Hwang Conpaper editor 

---

Related Article

현대차의 걸어다니는 무인 자동차 ‘타이거’ VIDEO: Hyundai Motor Group Unveils TIGER Uncrewed Ultimate Mobility Vehicle Concept

edited by kcontents

We know it would be a mistake to read too much into these new announcements, but we can’t help reading something into them, right? So let’s take a look at what Hyundai Motor Group has been up to recently. This first robot is DAL-e, what HMG is calling an “Advanced Humanoid Robot.”

According to Hyundai, DAL-e is “designed to pioneer the future of automated customer services,” and is equipped with “state-of-the-art artificial intelligence technology for facial recognition as well as an automatic communication system based on a language-comprehension platform.” You’ll find it in car showrooms, but only in Seoul, for now.

We don’t normally write about robots like these because they tend not to represent much that’s especially new or interesting in terms of robotic technology, capabilities, or commercial potential. There’s certainly nothing wrong with DAL-e—it’s moderately cute and appears to be moderately functional. We’ve seen other platforms (like Pepper) take on similar roles, and our impression is that the long-term cost effectiveness of these greeter robots tends to be somewhat limited. And unless there’s some hidden functionality that we’re not aware of, this robot doesn’t really seem to be pushing the envelope, but we’d love to be wrong about that.

The other new robot, announced yesterday, is TIGER (Transforming Intelligent Ground Excursion Robot). It’s a bit more interesting, although you’ll have to skip ahead about 1:30 in the video to get to it.

We’ve talked about how adding wheels can make legged robots faster and more efficient, but I’m honestly not sure that it works all that well going the other way (adding legs to wheeled robots) because rather than adding a little complexity to get a multi-modal system that you can use much of the time, you’re instead adding a lot of complexity to get a multi-modal system that you’re going to use sometimes.

chosun.com

edited by kcontents

You could argue, as perhaps Hyundai would, that the multi-modal system is critical to get TIGER to do what they want it to do, which seems to be primarily remote delivery. They mention operating in urban areas as well, where TIGER could use its legs to climb stairs, but I think it would be beat by more traditional wheeled platforms, or even whegged platforms, that are almost as capable while being much simpler and cheaper. For remote delivery, though, legs might be a necessary feature.

That is, if you assume that using a ground-based system is really the best way to go.

The TIGER concept can be integrated with a drone to transport it from place to place, so why not just use the drone to make the remote delivery instead? I guess maybe if you’re dealing with a thick tree canopy, the drone could drop TIGER off in a clearing and the robot could drive to its destination, but now we’re talking about developing a very complex system for a very specific use case. Even though Hyundai has said that they’re going to attempt to commercialize TIGER over the next five years, I think it’ll be tricky for them to successfully do so.

The best part about these robots from Hyundai is that between the two of them, they suggest that the company is serious about developing commercial robots as well as willing to invest in something that seems a little crazy. And you know who else is both of those things? Boston Dynamics. To be clear, it’s almost certain that both of Hyundai’s robots were developed well before the company was even thinking about acquiring Boston Dynamics, so the real question is: Where do these two companies go from here?

https://spectrum.ieee.org/automaton/robotics/industrial-robots/hyundai-boston-dynamics-new-robots

이제는 로봇이 차를 안내합니다! 현대차 DAL e달이 소개

See what an Ultimate Mobility Vehicle can do | TIGER | Hyundai

kcontents