Our recent robotics trends webinar included three great speakers. Here’s the presentation from Mark Bunger. He is a Vice President at Lux Research where he and his teams work with client innovation strategies across industries and scientific and technology domain. He joined Lux in 2005 and has launched and led many of the company’s intelligence services in consulting areas, ranging from bio based fuels and chemicals to big data and agriculture and healthcare. His current focus is on the intersection of information and matter in biogenetic, industrial, and environmental market.
To watch the webinar on demand, simply visit: http://www.designworldonline.com/innovative-trends-in-robotics/
The topic I wanted to talk about is Soft Robotics and if you haven’t heard about Soft Robotics, you can think of it as a new approach that fits in with a lot of other segments and branches of overall robotics taxonomy where you can get, for example, industrial robots.
Typically, like a robotic arm or one that has wheels and is driving around, swarm robots, bio-mimetic robots that mimic humans and things like that.
Soft Robots is a category really that talks more about the materials that are used to make the robots, so really, any of these approaches, you can look at a robot that typically would be made of metal or hard plastics and steel. People are looking at using soft polymers and textiles and other materials to create new kinds of robots that can work in new ways.
One of the aphorisms about the future and innovation is something that Alan Kay, a researcher at Xerox PARC said and, “The best way to predict the future is to invent it.” When you have a new idea like this or you’ve heard about some things might be happening in the future, one of the things that we do at my company or with research is we bring together those adjacent communities that we think are starting to converge, to create a new discipline or a new technology, in this case, software products.
Robotics obviously requires a lot of robotics experts but then there are also materials so these materials again might be new types of polymers or soft, flexible polymers that have specific, say stiffness, strength, even transparency. Weight, obviously a consideration or textiles like nylon and so things that we wouldn’t normally think of as good materials for making robots out of.
Then, of course, there are applications and users so a lot of times when you see a robot in an industrial setting, either it’s behind a cage or the human is behind a cage because the basically the robot doesn’t have a nice guidance. They’re showing it can kill you, it can knock your head off, basically. It’s a dangerous place to be. In these areas where the objects to be picked up or the humans present in the environment could potentially be at risk by a conventional hard robot, we’re looking at areas like agriculture, for example, or human care, where a softer approach might work.
To pull on a Pop Culture reference, if anybody saw the movie Big Hero Six, there was a big white puffy caretaker robot, that is precisely what soft robot is. It’s a robot, that again uses air pressure or fluid pressure and soft materials to do its job.
In the beginning, just basically had everybody talk to each other, explain their disciplines to one another so that if you’re a material scientist, you probably don’t know much about robotics. If you’re in robotics, you might not know much about, say agricultural needs for how to draw, for example, in the video where you saw pick off leaves from plants that might be diseased or dry.
These three communities talked to each other. They did presentations and collaborated on different types of, for example, prioritizing things that would be tasks to tackle together or even made prototypes. You can see some of the just paper and craft based quick prototypes that these people came together and put together in the space of a couple of hours. Which, again, going back to the point of this workshop and of this space in general, we’ve got three different communities where everybody knows their colleagues within their own community.
The robotic people know the robotic people. The materials people know the materials people, but they don’t generally know each other there.
If you’re interested in what you’re seeing here, there is a … on the black plate on the table, one of the teams had somebody from NASA on this and they basically had the idea that you could use wax melting and then solidifying to create locomotion in a tripod robot. Over to the left, you see somebody who’s wearing a soft prosthetic, a knee basically, a potentially better way to integrate partial robot prosthetics with their human wearers.
The partners, like I said, really brought a lot to the table. One of them was SRI and SRI, if you haven’t heard of SRI, like I said they mention … As I mentioned, they invented SIRI, SIRI and SRI, that’s not a coincidence to the name there. They invented the computer mouse but they’ve also done a lot of work in robotics and some of that’s been for DARPA and for government programs. A lot of those have military goals, obviously, but they do try and then spin out what they can into commercial applications.
Some of the companies they’ve spun out are Intuitive Surgical, which is a surgical robotic company, artificial muscle, which is kind of an actuator, but the one that I think is most interesting in this context is Grabit and so and you tell from the logo what this company does. It basically makes a type of grabber that is soft and works for things like textile manufacturing, for example. Apparel manufacturing, getting robots to interact with floppy pieces of cloth can be really challenging. They’ve found a way to tackle that problem.
Specifically what they do is they use a type of gripper called … They talk about a phenomenon, rather, called Electroadhesion and they have grippers which are basically like I think of old 35 millimeter film or something like that. It’s a flexible plastic film that doesn’t have any rigidity or structure in and of itself but when you apply a current to it, it basically creates electrostatic adhesion to, with the thing that it’s … and forms a very tight bond with that.
As you can see from this example, the box is a very square structure but the grippers conform to its various angles and you can see where a lot of the lack of precision that some environments are constrained by aren’t a problem for this approach. Basically, the grippers find their way around the corners and the edges and find a surface that they can latch onto.
This company, like I said, was spun out of SRI. They’ve recently raised $6 million, Series A, and they’ve gotten strategic investments from ABB which obviously makes robots but also Nike, which again is looking at the … so where able to manufacture jackets and pants and things like that more effectively.
That’s one example of a software product application. SoftRobotics is actually the company name in this case, as well as being the category that it’s in. What they basically do is they make squishy grippers. If you don’t know the shape of, say, a tomato or an egg or something like that that you want to try and pick up, basically this was a technology developed at the Whitesides Research Group at Harvard that then spun out and they have about five employees today.
Basically, what they do is make, as you can see, this blue soft grippers that allow a lot of applications again where the object to be picked up is very soft or is a bit unusual or unpredictable shape. Again, we talked earlier about using vision as a control mechanism, feedback and control mechanism when the environment is unstructured.
You can think in Soft Robotics as an environment where the robot is unstructured as well. It’s less precisely controlled and geometry is less precise than a classic hard robot with separate motors and real closed feedback loops built into it. A lot of Soft Robotics we’ve included use vision. In other words they see when they’re grasping something and they use that as the deducting controller.
Again, here’s another company that we’re watching that’s in this space and then a third one that I think is really interesting and really different from the other two. It is doing grippers but it’s actually doing the entire arm and using textiles for all of that and in addition using textiles, you can see. It uses that air pressure for its own structure but it also uses it for some movement.
You can see on the elbows and a little bit on the flow of the gripper itself by designing the geometry of the textile in such a way that increasing or decreasing the pressure in these tubes at the joints, basically is able to then in fairly precise ways. Precision that it can’t get from a priori, it uses vision to maybe again to be able to hone in on its target, know when it’s got a hold of something and move on.
As you can see from the short video of the engineer here who is wrestling with the robot to the ground and beating it up. If there ever is a robot takeover of humanity, hopefully, it’ll be these guys because we can at least fight back. You can also, obviously … If you have this feel, like sacred to be working around robotics, so that’s new Pneubotics.
A sister company to Pneubotics is both working on soft exoskeletons. If you basically take that same ideas that I just showed you where you’ve got joints that are actuated by air pressure or fluid pressure and put a person inside that. I was actually just at SRI this week and saw some of their advances in this area but. The advantages in this approach are not only the inherent safety that I … As I said earlier, but it’s considerably lighter than even rigid, metal or plastic materials.
One of the issues with exoskeletons is attachment to the body, so if you have a really strong actuator but it’s not actually, say, moving your leg, it’s moving your clothes, it’s not going to basically give you the effect that you want. The softer exoskeletons to tend to conform to your body more so power gets transferred from the machine to the motion of the human.
This is a way, like I said, it’s pretty nascent and hopefully is a lot of interest to you guys. I’ve showed you some examples that are commercial and nearly commercial but there’s a lot of science going on as well. Every so often, there’ll be a special edition of, say, a journal like Advanced Functional Materials or something like that. We’ll basically go in and focus on more some of the new advancements that are happening.
A lot of what is making this go so quickly, advancing quickly, that is is because we’re, as I was mentioning earlier, using the geometry of the textile or say the geometry of the microrobotics inside a polymer part, or those things basically are … Give you a lot of degrees of freedom from the engineering perspective to do things that’ll be difficult to accomplish with the normal motor driven or robotic set up. The way things twist, the way they actuate, you get just a whole new set of possibilities by being able to move things that way.
One final example, I wanted to show you an example of the really out there thinking around Soft Robotics … There’s a researcher who’s name is Vytas SunSpiral working at NASA Ames. He’s been working on Tensegrity robots and those are basically robots that have rigid parts but they’re held together by flexible cables and other types of materials. They work like these children’s toys, if you’ve seen them, these sticks they fit together with the rubbery strings.
By varying length and the tension on those cables, you can make something like this boxy thing you see in the picture, actually move. They can take a fall very nicely and so if you were designing a robot to explore Mars or someplace like that, you can see that this would be an example. There are actually a few science fiction books and movies that have started to take this idea and at least put it into a fictional play. Neil Stevenson’s book, Seveneves, it came out about a year ago, uses one of these tumbler robots basically.