Cyborg or avatar: will you wear a robot, or operate it?

robots-cyborg-sarcos-work

Some robots won’t replace us. Some will augment us: make us faster, stronger, bigger, more capable. Sarcos Robotics has built both robots that you can wear and robots you can operate, and just recently unveiled the Guardian XT to complement the Guardian XO.

One you wear, and it helps you lift 200 pounds with minimal effort. The other you teleoperate in VR while wearing motion-capturing clothing. Weld, cut, lift, bolt … you can do it all in dangerous situations from the safety of the ground, or your home.

(Subscribe to my YouTube channel so you’ll get notified when I go live with future guests, or see the videos later.)

In this episode of TechFirst with John Koetsier, we chat with Ben Wolff from Sarcos about the Guardians, how they work, and what they mean about the future of work. Check out the story on Forbes here, or keep scrolling for the full transcript …


Support TechFirst: Become a $SMRT stakeholder


What we talk about:

  • 0:00 Fleet of thousands
  • 1:06 Robot avatars
  • 2:11 Wearable robots
  • 4:43 Merging human and machine
  • 5:08 Where these robots work
  • 8:41 Fine motor control
  • 10:03 Power supply
  • 10:58 Range of motion
  • 13:30 Robots & work in 5 years

Subscribe to TechFirst: robots that cyborg us and robots we ride as avatars

 

Full transcript: cyborg or avatar, or how we’ll use robots in the workforce

(This transcript has been lightly edited for length and clarity.)

John Koetsier: We worry a lot about how robots are going to replace us and take our jobs. We don’t think as much about how robots will help us, and do dangerous jobs for us, help us avoid getting electrocuted, crushed, or maybe falling from heights.

That’s what the Guardian XT is. It’s made by a company called Sarcos, based in Salt Lake City. Guardian XT can weld, it can cut, grind, lift up to 200 pounds. It’s controlled remotely by an operator wearing a motion capturing garment and controller — kind of looks like it belongs on a Tesla manufacturing line — but it’s mobile via a lift or other vehicles, and it just got a big 5G integration upgrade.

To learn more, we’re chatting with Ben Wolff, founder and CEO of Sarcos Robotics. Welcome, Ben!

Ben Wolff: Thank you, John.

John Koetsier: Super happy to have you here. Tell us about Guardian XT, what is it? 

Ben Wolff: Guardian XT is an industrial robot that has the ability to move the way humans move with their upper body. So when we talk about dexterity, we talk about being able to do all the kinds of tasks that humans do with their arms and their wrists and their hands, and to do it in a way that keeps humans out of harm’s way.

So think of it as a robotic avatar, if you will. We all kind of know what avatars are in the digital world. Think about this being in the real world, where, in this case, the robot goes and does the work for you in the dangerous or difficult environment, at height or in a confined space, and the robot follows you and performs the motion and movement that you ask it to directly and immediately.

John Koetsier: That sounds so science fiction — I’m wearing something and I move my arm and the robot arm moves and everything — but it’s also so natural, right? I do exactly what I want it to do. I think we see some applications in medical robots for surgery that have similar things like that.

You’re basing this off an existing robot that you have which is the Guardian XO, an exoskeleton. Can you talk about that a little bit?

Ben Wolff: Yeah, you bet.

What we realized as we were developing over the last 20 years the Guardian XO, which is a wearable full body robot — so it’s everything you can imagine in a full body full-sized human scale robot except you can wear it; you get inside of it and it allows you to lift up to 200 pounds with no stress or strain on your body at all — and what we realized as we were bringing that product to market is that there are some applications where our customers were wanting to do things like put our exoskeleton in a bucket truck, so that heavy and dexterous work could be done at height.

And we started thinking to ourselves, well, why do we bother with the legs on that robot? You don’t need the legs, in fact, you don’t even need the human to be in the bucket anymore to be able to lift 200 pounds. Let’s take the torso and the upper body and the arms of the exoskeleton, put it into the bucket truck and then keep the human down on the ground to be able to manipulate and direct what the robot does up at height. And that spawned a whole wide variety of interest from a lot of different industries.

Because, number one, so many people get injured doing work at height. It’s more expensive, it’s more time consuming, there’s injuries involved. And the same thing goes for doing work in, for example, nuclear power plants — might not be at height, but you’re talking about situations where humans shouldn’t be exposed to those kinds of environments. So what if we can take a robot that can follow our most direct direction and do those tasks in those kinds of environments, keeping humans out of harm’s way. And that’s what our whole focus was.

So the exoskeleton is about having a human in the machine, in the workplace. And the Guardian XT is about those environments where you don’t want to have the human inside of it and instead you’ve got your robotic avatar. 

John Koetsier: You have gone full-on Avatar, hey? I mean, you’ve got the exoskeleton, that’s in the movie. You’ve got [laughing] the remote teleoperation, that’s in the movie, kind of. Very interesting.

I hadn’t seen on your site anything about the nuclear reactor application, but that is really interesting as well because you have remote operation. Because we’ve seen robots used in high-radiation scenarios before. We’ve seen that for instance in Japan, where they have that ongoing issue with that one reactor that got devastated by a tsunami. But they have an issue, because they have sensitive electronics on board and that gets fried out. It might be a little bit better to have teleoperation because you don’t need as many sophisticated electronics — the brain doesn’t have to be on board. Is that correct? 

Ben Wolff: Exactly. Our whole concept here is that we’re leveraging the best of what humans can offer — human intelligence, wisdom, judgment — with the capabilities of a machine. And so it’s kind of this combination of man and machine to get the best result possible. And so we’re relying very much on human intelligence.

Nothing about what we do today is about artificial intelligence. It’s all about human intelligence and finding this great synergy between what the best of humans can offer and the best of machines.

John Koetsier: Where do you see XT being most useful? What industries are you looking at? What kinds of jobs will it do? 

Ben Wolff: I think that first and foremost, as we launch this product at the end of next year, we’re going to see real applicability in the construction industry … doing things at height, whether it’s welding and sanding and grinding and cutting and all of those kinds of things that you see construction workers doing either in bucket trucks or on scissor lifts. Anything that’s an elevated platform, we can now do that with keeping humans safely down on the ground.

I also think we’ll see it in the power industry. So doing things like limbing trees over power lines, maybe even getting to the point that we’re managing some of the fine motor skill work that needs to be done on the top of the power pole. When you talk about doing things like transformer installation, removal, replacement, insulators, those kinds of things — those are all things that are highly dangerous jobs.

And I’ll tell you one of the things that many of these industries that we’re working with struggle with is it’s hard to attract young people today to go into these physically demanding jobs that are dangerous. I mean, all of these industries are having labor challenges. So if now we can attract young people to say, ‘You know what? You can still do this really cool work, really meaningful work, but instead of having to be at height in a bucket truck, now you get to stay on the ground where it’s safe; you can stay dry, and you can manipulate the robot because it’s the one doing the hard work up in the air.’

So, the power industry, construction industry, even the infrastructure inspection industry — doing things like inspecting elevated pipelines in chemical plants. You know, oftentimes you have to put scaffolding up and get humans up 30, 40 feet in the air with sensitive devices to be able to determine whether the infrastructure is solid, whether it’s performing as expected, whether there’s defects or deficiencies. Now we can keep the human on the ground and have that work being done by the robot at height. 

John Koetsier: I found it so interesting you were talking about construction and robots, because that’s been an industry really resistant to the introduction of robots, not because they don’t want it necessarily, but it’s so hard. It’s such a dynamic environment, having a robot learn where to go, what to do in this constantly changing world of construction — there’s new materials here, that building wasn’t there yesterday, right? You know, that’s a really challenging space to work in. And if you can help with that, I think that’s really, really interesting.

What’s kind of the status here … when is this shipping? 

Ben Wolff: We’ve just done some of our first field deploy alpha trials where we’ve gotten our alpha version of the machine out with customers. We’ve gotten great feedback. We’re taking the feedback that we got from those customers and incorporating all of that into our beta version which is in the design process now.

We expect to have the first beta units completed towards the end of this year. We’ll go out and do the same thing … again, trial, trial, trial with our customers. Make sure we’re hitting the mark with what their needs are, what their use cases are, and then our expectation is to start low rate initial production at the end of next year and getting the first commercial units out into the field then.

It seems like every week that goes by we get more use cases, more customers from different industries saying, ‘Well, if it can really do that, boy, I’ve got three things that it can go do.’ So we love getting the robots out in the field and getting customers to experience it first hand.

John Koetsier: Are you building that with sort of an extensible platform as well? I know that some industrial robots, you can take an arm off and put a different arm on for different attachments. You talked about fine motor control as well, and are you thinking along those lines?

Ben Wolff: We are in a number of respects. We think of it being a modular robot in terms of the bases that it can affix to. So whether it’s a scissor lift, or a boom truck, or a teleoperator, or even a rolling mobile base that doesn’t need to do anything at height, but just like in the nuclear power plant example where it just needs to roll around the floor. So it’s modular in that respect and customizable in that respect.

In addition, the end effectors or the hands as you were just referring to, we’re coming up with a version that is an actuated three-finger robotic hand that can grasp a lot of different kinds of tools. But we also recognize that may not be the choice for all applications. And so we will have a version of this machine that will have the ability to swap out tools on a quick disconnect basis, so that when the three-finger end effector isn’t the right solution, you might want to attach a saw or a welding torch right to the wrist and do the job consistently with that type of approach.

John Koetsier: Amazing. I could see that working. You were talking about limbing trees or something like that — just attach a saw and there you go, right? Very, very useful. What do you need on site to make Guardian work? Like what kind of power supply? How’s that run? 

Ben Wolff: So the robot is entirely powered by lithium ion batteries. So there is nothing required on site. This is such an incredibly extensible platform because you mount it on whatever your mobile platform or base is, and it’s all self-contained from a power perspective.

So we can put the combined base on a truck, roll it out to a job site, and immediately deploy it with no customization or infrastructure required on site.

John Koetsier: Amazing! What kind of a charge does it have and are those quick swap batteries? 

Ben Wolff: The batteries are hot swappable in the field without even powering down the machine. We get about two to three hours of runtime out of the machine today between charges, and we ship every unit with an extra set of batteries. So again, when you think about being able to swap them out in 10 or 15 seconds, you can have effectively unlimited runtime with this machine. 

John Koetsier: Amazing. Amazing. We talked a little bit about the fingers. The full name of the robot is Guardian XT Highly Dextrous Mobile Robot.

How dextrous are we talking about here? What kind of fine motor control are you thinking about building into this thing? 

Ben Wolff: First and foremost, think about how dextrous the arms are. The fact that the arms have the same kind of range of motion as humans have. And in effect, we’ve built the machine to be patterned after the way the human moves, so that it’s what we call kinematically equivalent to the human body.

So what that means is when you’re wearing the control device and you just start moving your arms or your limbs the way you would normally do so in the real world, the robot just automatically follows you. You don’t have to think about managing the robot, and that’s because of the level of dexterity that we have.

Then when you get to the hands or the end effectors, it really does get down to the specific tool or job that it’s being deployed for. And it really just depends on what that tool and use case is, but we expect to be able to do almost all of the same kinds of things that humans can do, but with the safety and precision and strength of this robot.

So, you know, I couldn’t be in a bucket truck lifting 200 pounds with my own arms, but this robot can. 

John Koetsier: I love it. I mean, we’re in a human world, we’re in a human-designed world — the trucks we have, the tools we have are designed for humans — so if you’ve got something that uses tools like a human, fits like a human, has reach and extensibility like a human, it works in that reality. That makes a ton of sense.

I wonder if you could project out for us a few years, maybe three to five years or so, what does this look like? How capable is it? What is it doing? 

Ben Wolff: I’ll really focus on kind of that three to five-year period. I think that we’ve got a machine that is so intuitive to use, it becomes a natural part of the workforce. You know, our customers will put these machines on their payroll, just like humans, and it becomes part of a work team.

So we don’t actually sell the machines. We provide them as a service to our customers, just like it is a unit of labor … but it’s next generation labor and it’s a true partner with its human operator. So just like you think about other types of machines that are used in construction industries or the power industry where there’s an operator involved, same thing in our case.

And so what we see three to five years from now is a fleet of tens of thousands of these machines deployed around the world that are partnered with their human work teams, and they’re able to get far more work done and to do it far more safely. 

John Koetsier:  That’s amazing, Ben. There’s a lot of work to be done. We talked before this show about the wildfires that are raging through the Midwest, the Pacific Northwest, other areas. We’ve seen flooding in Turkey and fires in Greece and all over the world. We’ve got a need to drive solar, a need to drive a lot of different alternative forms of energy and a lot of work to be done in the world.

And a lot of it will be dangerous and in remote areas, so the more of this sort of solution that we can get, the better. You’ve added 5G capability and we didn’t actually talk about that yet here. I forgot to ask you … what is the point of the 5G capability and how does that make it more useful, more capable? 

Ben Wolff: So I mentioned before how important it is for this machine to be intuitive to use. You don’t want to have to have the operator think a lot about managing the machine when they’re also doing a dangerous job, right? So they want to be able to focus all of their attention on the job. Well, we accomplished that today by having very low latency in the machine and the communications link between the human operator and the machine.

And we achieved that today with a fiber optic tether because we can send the control at the speed of light, the control signal at the speed of light. And as a result, it feels very intuitive. It feels like a natural extension of the human body because there’s no delay in what the operator does versus what the robot does.

If we tried to deploy a wireless solution over something like 3 or 4G today, there simply isn’t the bandwidth to be able to provide the operator with that same kind of no-latency experience, where it just feels like a natural extension. So we’re excited about 5G. We’re excited about our partnership with T-Mobile because they have a tremendous 5G network with a lot of bandwidth that is really specifically designed to be able to address these kinds of high bandwidth applications that we have. It is a perfect application for 5G.

So now instead of having to have a fiber optic tether between the human operator and the robot, now we could actually have a situation — as long as we’ve got our high bandwidth, low latency, 5G network available — you could theoretically have somebody on the other side of the country managing one of these robots … or a fleet of these robots. And that’s really where it gets exciting and you start seeing some exponential gains in the use cases and applicability, when you can maybe have a handful of robot operators sitting in a control room in one part of the country and they’re managing a fleet of these robots doing dangerous and difficult work in another place. 

John Koetsier: It’s really interesting to hear you say that, because I was just thinking of, you know, sometimes you’re in a Google doc or some form and you’re typing and it’s not showing up at the same speed because [laughing] your internet’s bad or something like that, and how many mistakes do you make on a simple job like typing out a simple sentence because you don’t have that direct immediate feedback … how much worse could that be, you’re working on a power line, you’re working on a dam and you’re going to hit stuff, you’re going to break stuff if you don’t have that immediate feedback.

That makes a ton of sense. Ben, I want to thank you for your time. It’s been super interesting. 

Ben Wolff: Great, thank you very much, John. I appreciate it.

Wow … you read all of that?

Made it all the way down here? Who are you?!? 🙂

The TechFirst with John Koetsier podcast is about tech that is changing the world, including wearable tech, and innovators who are shaping the future. Guests include former Apple CEO John Scully. The head of Facebook gaming. Amazon’s head of robotics. GitHub’s CTO. Twitter’s chief information security officer, and much more. Scientists inventing smart contact lenses. Startup entrepreneurs. Google executives. Former Microsoft CTO Nathan Myhrvold. And much, much more.

Consider supporting TechFirst by becoming a $SMRT stakeholder, and subscribe on your podcast platform of choice:

 


Want weekly updates? Of course you do …