WEBVTT - TechStuff Sends Exoskeletons XOs 0:00:04.400 --> 0:00:07.800 Welcome to text Stuff, a production from my Heart Radio. 0:00:12.080 --> 0:00:14.720 Hey there, and welcome to tech Stuff. I'm your host, 0:00:14.880 --> 0:00:18.360 Jonathan Strickland. I'm an executive producer with I Heart Radio, 0:00:18.400 --> 0:00:21.880 and I love all things tech, and you guys probably 0:00:21.920 --> 0:00:26.079 know that beyond tech, I also love science fiction. I 0:00:26.120 --> 0:00:29.360 grew up reading books and watching movies and TV shows. 0:00:29.400 --> 0:00:33.480 They're all in the science fiction genre, and there are 0:00:33.520 --> 0:00:37.680 a lot of futuristic technologies in those works that I 0:00:37.760 --> 0:00:41.920 find really compelling, from faster than light space travel, which 0:00:42.240 --> 0:00:45.800 I suspect will never be a reality, to flying cars, 0:00:46.320 --> 0:00:49.519 to a replicator that can make lasagna whenever I want it. 0:00:50.280 --> 0:00:53.920 Somebody invent that. But today I wanted to talk about 0:00:53.960 --> 0:00:56.760 one of those technologies that we're actually seeing evolve in 0:00:56.760 --> 0:01:01.800 the real world, and that would be exoskeletons. Now, because 0:01:01.840 --> 0:01:05.240 of my age, when I think exo skeletons, I think 0:01:05.240 --> 0:01:09.560 of Ripley in the film Aliens, the second of the movies. 0:01:09.880 --> 0:01:13.880 She's in that giant cargo loader exo skeleton going toe 0:01:13.959 --> 0:01:17.560 to toe with the alien queen. But there are tons 0:01:17.800 --> 0:01:21.480 of other examples in science fiction, ranging from the industrial 0:01:21.800 --> 0:01:26.440 to more streamlined versions like Iron Man's suit. And we 0:01:26.520 --> 0:01:29.160 are not anywhere close to being able to build a 0:01:29.200 --> 0:01:31.959 suit like Tony Starks, but we do have some pretty 0:01:31.959 --> 0:01:35.600 cool robotic exo skeletons out there. Some are intended to 0:01:35.600 --> 0:01:38.520 help people move heavy objects, kind of like the cargo 0:01:38.600 --> 0:01:41.920 loader from Aliens. Some are intended to provide support to 0:01:41.959 --> 0:01:45.479 give people with mobility issues more independence, Some are meant 0:01:45.520 --> 0:01:48.120 for people in the military to help them walk for 0:01:48.160 --> 0:01:51.440 further distances and carry heavier loads, and some are meant 0:01:51.480 --> 0:01:55.320 to encourage empathy and understanding in others. So today we're 0:01:55.320 --> 0:01:58.560 gonna look at the development of exoskeletons and their uses. 0:01:59.280 --> 0:02:01.600 One thing I want to talk about right at the 0:02:01.640 --> 0:02:07.160 start is how developing exoskeletons is truly a multidisciplinary pursuit, 0:02:07.760 --> 0:02:09.840 And what I mean by that is that it requires 0:02:09.880 --> 0:02:13.440 the expertise of people in very different fields of study. 0:02:14.120 --> 0:02:18.160 One obvious one is robotics. Exoskeletons share a lot of 0:02:18.200 --> 0:02:22.440 common traits with robots, and robotic elements frequently are part 0:02:22.639 --> 0:02:26.520 of powered exoskeletons. But here's the thing about robots. Their 0:02:26.560 --> 0:02:31.079 form doesn't have the limitation of the human form. Robots 0:02:31.080 --> 0:02:34.480 come in all shapes and sizes, Some aren't mobile at 0:02:34.480 --> 0:02:37.359 all the ones that do move may move in very 0:02:37.400 --> 0:02:41.200 different ways from the way we move. There are swimming robots, 0:02:41.240 --> 0:02:46.480 climbing robots, four legged robots, wheeled robots, robots with treads. 0:02:46.680 --> 0:02:48.600 An engineer who wants to build a robot has a 0:02:48.600 --> 0:02:51.600 lot of options and will likely choose whichever one's best 0:02:51.600 --> 0:02:55.040 suit the purpose of that robot. But we humans have 0:02:55.160 --> 0:02:59.480 way more limitations. Our limbs only work certain ways, and 0:02:59.600 --> 0:03:02.320 if you try to make them work outside of those 0:03:02.360 --> 0:03:07.320 certain ways, well you might get some unpleasant crunching and breaking. 0:03:07.639 --> 0:03:10.520 And I have to admit, one of the nightmarish visions 0:03:10.560 --> 0:03:13.120 I have of an exoskeleton is one that begins to 0:03:13.240 --> 0:03:17.040 bend the opposite way of a human joint and just 0:03:17.520 --> 0:03:21.480 you know, keeps going, which is yikes. So my point 0:03:21.520 --> 0:03:24.440 is that when designing exoskeletons, it is important to have 0:03:24.480 --> 0:03:27.680 people with expertise on the physiology of human beings as 0:03:27.680 --> 0:03:29.720 part of the team to make sure that the exoskeleton 0:03:29.760 --> 0:03:33.800 works with, not against, the person wearing it. Add to 0:03:33.840 --> 0:03:35.760 that the fact that we human beings are you know, 0:03:36.480 --> 0:03:39.880 a little squishy, you know what. Admittedly, ever since I 0:03:39.880 --> 0:03:43.040 stopped going to the gem, I've become way more squishy 0:03:43.120 --> 0:03:47.040 and mechanical systems typically are very much not squishy, and 0:03:47.080 --> 0:03:50.880 sometimes pairing those things together results in bad stuff like 0:03:50.960 --> 0:03:54.240 people getting hurt, and has given rise to a discipline 0:03:54.280 --> 0:03:58.560 sometimes called soft robotics, or systems that can interoperate with 0:03:58.760 --> 0:04:03.320 softer stuff like humans. So the development of an exoskeleton 0:04:03.840 --> 0:04:07.760 is an arduous process, and unless some of the people 0:04:07.880 --> 0:04:11.000 actually developing the technology are the ones who will ultimately 0:04:11.120 --> 0:04:13.960 use it, you also have the added challenge of making 0:04:14.000 --> 0:04:17.320 sure that the thing you're creating is actually useful for 0:04:17.480 --> 0:04:22.400 the intended audience or the intended consumer. And I'm sure 0:04:22.440 --> 0:04:24.520 I'm not the only person who, after I got my 0:04:24.560 --> 0:04:28.520 hands on a product, found myself frustrated because it wasn't 0:04:28.520 --> 0:04:31.520 working the way I thought it would work. It worked 0:04:31.560 --> 0:04:34.239 the way the designers thought it should, but they didn't 0:04:34.279 --> 0:04:37.599 necessarily take into account how the end user, who was 0:04:37.760 --> 0:04:41.320 not involved in the development or production of this technology, 0:04:41.560 --> 0:04:44.159 was going to try and actually use this thing. Look, 0:04:44.200 --> 0:04:47.640 I'm saying, sometimes I get stuff and it's not dummy proof, 0:04:47.839 --> 0:04:50.120 so I find myself wondering how the heck to get 0:04:50.120 --> 0:04:53.320 it to work. And yeah, I'm being a bit glib, 0:04:53.360 --> 0:04:56.640 but it really can be an issue. Designers can sometimes 0:04:56.720 --> 0:04:59.920 make something that doesn't really work for anyone because the 0:05:00.000 --> 0:05:04.440 designers got sidetracked solving one or more technical problems and 0:05:04.480 --> 0:05:07.800 didn't create something that is actually helpful. Now, that last 0:05:07.839 --> 0:05:12.520 point applies to any product really, not just exoskeletons. However, 0:05:12.880 --> 0:05:16.160 with exoskeletons you can see how it is particularly important. 0:05:16.560 --> 0:05:21.479 Designing and building even a prototype is incredibly labor intensive 0:05:21.640 --> 0:05:25.760 and thus expensive. It requires a lot of time and research, 0:05:25.839 --> 0:05:28.640 and that is an investment, so you kind of want 0:05:28.680 --> 0:05:31.480 to make sure the end result is something worthwhile, even 0:05:31.480 --> 0:05:36.080 if it's not going to become a commercial product in 0:05:36.120 --> 0:05:38.920 the future. You want to make sure that what you're 0:05:38.920 --> 0:05:43.400 doing actually matters. And you can also think of exoskeletons 0:05:43.400 --> 0:05:47.720 as falling into one of three broad categories. There's actually 0:05:47.720 --> 0:05:50.040 a couple different ways we can classify exo skeletons, but 0:05:50.800 --> 0:05:53.479 one way is to look at what they are intended 0:05:53.560 --> 0:05:57.279 to do. Are they upper extremity exo skeletons meaning stuff 0:05:57.360 --> 0:06:01.680 that helps with moving heavy weights using your arms and 0:06:01.880 --> 0:06:06.080 your your torso. Are they lower extremity exo skeletons something 0:06:06.160 --> 0:06:08.760 that helps, you know, carry weight with the legs or 0:06:08.800 --> 0:06:12.599 the hips, or maybe helps correct some sort of physiological issue, 0:06:13.480 --> 0:06:17.320 or are they full body exoskeletons, you know, essentially shoulder 0:06:17.400 --> 0:06:20.360 to toe typically is what we're talking about with these things, 0:06:20.839 --> 0:06:23.600 and it's hard to pin down the earliest appearance of 0:06:23.720 --> 0:06:28.159 the idea for a powered exoskeleton, but one early example 0:06:28.200 --> 0:06:32.160 in fiction comes from Robert Heinlein's novel Starship Troopers, in 0:06:32.240 --> 0:06:35.440 which human soldiers of the future are wearing power suits 0:06:35.480 --> 0:06:39.320 that enhance the soldiers strength and speed and other capabilities. 0:06:39.880 --> 0:06:42.839 Some might argue this is more an empowered armor category 0:06:43.040 --> 0:06:46.040 and not exo skeleton, but I maintain those distinctions are 0:06:46.160 --> 0:06:49.560 kind of meaningless in the real world, or at least 0:06:49.920 --> 0:06:52.680 outside the world of fiction. The earliest example I could 0:06:52.720 --> 0:06:55.520 find dates from a nineteen sixty one article in the 0:06:55.640 --> 0:06:59.080 journal Armor, published by the United States Army. It's a 0:06:59.160 --> 0:07:01.680 write up in an news section, so it's sort of 0:07:01.720 --> 0:07:05.520 a series of headlines and not a full like feature 0:07:05.880 --> 0:07:09.040 in the magazine itself, and this particular little story has 0:07:09.080 --> 0:07:14.040 the headline future soldier human tank charming right. The opening 0:07:14.120 --> 0:07:18.400 sentence reads quote the soldier of the distant future will 0:07:18.440 --> 0:07:22.240 be a human tank equipped with power steering and power brakes. 0:07:22.360 --> 0:07:27.640 The Pentagon said recently end quote golly. The project was 0:07:27.760 --> 0:07:31.920 also called Servo Soldier according to this news piece, and 0:07:32.200 --> 0:07:35.080 it wasn't announcing a new invention. It was rather talking 0:07:35.120 --> 0:07:38.280 about how the Pentagon and the Department of Defense were 0:07:38.280 --> 0:07:42.239 soliciting inventors for ideas that could contribute to this goal 0:07:42.480 --> 0:07:45.760 of creating a servo soldier. The suit was going to 0:07:45.880 --> 0:07:51.040 have a pretty amazing wish list of features, so enhanced 0:07:51.080 --> 0:07:57.880 strength check, enhanced speed check, enhanced resistance to stuff like heat, cold, 0:07:58.360 --> 0:08:04.160 toxic gas, and than radiation check. The military experts acknowledge 0:08:04.200 --> 0:08:08.440 that there were one or two minor obstacles to overcome, 0:08:08.800 --> 0:08:11.040 like creating a way for the suit to interpret the 0:08:11.120 --> 0:08:14.880 moves of the soldier as commands, thus generating an output 0:08:15.000 --> 0:08:18.760 several times the strength of the original motion, like if 0:08:18.800 --> 0:08:20.760 you were to throw a punch, the punch would be 0:08:20.880 --> 0:08:24.440 way more powerful, for example. And finding a quiet, portable 0:08:24.480 --> 0:08:27.560 power source capable of generating at least four horsepower was 0:08:27.600 --> 0:08:31.360 another challenge that they admitted to. And the name gives 0:08:31.400 --> 0:08:34.319 me a chance to explain just what a servo is. 0:08:34.800 --> 0:08:37.600 It's not just the surname of a plucky robot who 0:08:37.679 --> 0:08:41.640 riffs on bad movies Shout Out to Tom's servo. Servo 0:08:42.160 --> 0:08:45.960 is short for servo mechanism. The servo mechanism for an 0:08:46.000 --> 0:08:48.960 exo skeleton like the one the army was proposing, would 0:08:49.000 --> 0:08:52.160 be sort of like our muscles. They would connect linked 0:08:52.280 --> 0:08:56.120 parts of the exoskeleton and provide the motive power so 0:08:56.280 --> 0:08:59.439 that they could move or really to augment the movements 0:08:59.520 --> 0:09:03.959 of who whoever is wearing the exoskeleton. So servos often 0:09:04.360 --> 0:09:08.880 find their way into what is called closed loop applications. 0:09:09.320 --> 0:09:11.719 A closed loop system is one in which there is 0:09:11.760 --> 0:09:14.880 a process that has an outcome and the state of 0:09:15.000 --> 0:09:19.319 that outcome then comes back to affect the process. So 0:09:19.480 --> 0:09:21.959 one example of this is a clothes dryer with a 0:09:22.080 --> 0:09:25.079 dryness sensor. So you throw some wet clothes into the 0:09:25.160 --> 0:09:28.079 dryer and you set the dryer to to dry. The 0:09:28.200 --> 0:09:31.400 dryer tosses your unmentionables around while blowing hot air to 0:09:31.520 --> 0:09:34.520 dry them out, and a sensor stays on the lookout 0:09:34.600 --> 0:09:37.079 for signs of moisture. And so as the cycle is 0:09:37.160 --> 0:09:40.080 coming towards an end, the sensor is detecting if there's 0:09:40.080 --> 0:09:42.760 any moisture present, and if it does pick up that 0:09:42.880 --> 0:09:46.400 there's moisture, there the clothes are not really dry. It 0:09:46.480 --> 0:09:49.560 then feeds back into the system, and thus the dryer 0:09:49.720 --> 0:09:52.760 just keeps going for a while longer. If the sensor 0:09:52.880 --> 0:09:56.040 does not detect moisture, then it says, okay, it's good 0:09:56.080 --> 0:09:58.720 for us to stop now, and the whole process stops. 0:09:58.840 --> 0:10:01.240 This is a closed loop because it relies on that 0:10:01.400 --> 0:10:06.559 feedback signal. So a servo motor, at its heart is 0:10:06.600 --> 0:10:10.599 a device that rotates parts of a machine precisely and efficiently, 0:10:10.960 --> 0:10:15.439 so that rotation might be translated into a different type 0:10:15.520 --> 0:10:19.480 of motion, usually through gears of some sort, so that 0:10:19.640 --> 0:10:22.120 you have an element inside the servo that's rotating, but 0:10:22.400 --> 0:10:25.599 the part that you might see might be uh a 0:10:25.880 --> 0:10:29.599 platform that raises and lowers. But it can do so 0:10:29.960 --> 0:10:33.640 at a very precise distance and a precise speed, and 0:10:33.760 --> 0:10:35.959 that's it, which is pretty simple, right. They're able to 0:10:36.000 --> 0:10:39.520 move at varying velocities and two different positions. But that 0:10:39.720 --> 0:10:44.280 differentiates servos from other types of motors. Servo motors have 0:10:44.640 --> 0:10:47.280 some form of sensor that measures the position of the 0:10:47.360 --> 0:10:51.640 motors moving components, and the controller that dictates how and 0:10:51.800 --> 0:10:55.160 when and how far and how fast the servocean move. 0:10:55.559 --> 0:10:59.400 So imagine a robot arm with a joint like an elbow, 0:11:00.040 --> 0:11:02.360 and imagine it has the same range of motion as 0:11:02.440 --> 0:11:06.079 a typical human elbow. The servo motor would provide the 0:11:06.120 --> 0:11:09.040 force to extend the arm or to bend it back, 0:11:09.760 --> 0:11:13.040 and the sensor would detect precisely where in the range 0:11:13.080 --> 0:11:17.040 of motion the arm was at any given time, and 0:11:17.120 --> 0:11:19.719 the controller would be the element providing the instructions on 0:11:19.840 --> 0:11:23.320 what should be happening at any given moment. Now, there's 0:11:23.320 --> 0:11:26.079 a lot more to be said about servo motors, but 0:11:26.200 --> 0:11:28.920 I'll have to dedicate a full episode to explain them 0:11:28.960 --> 0:11:31.959 in greater detail. For our purposes, it's good to know 0:11:32.240 --> 0:11:35.360 that it's an electro mechanical element that acts like a muscle, 0:11:35.880 --> 0:11:39.160 largely in robotics. Uh not, that's not the only application 0:11:39.200 --> 0:11:41.599 we find servo motors in all sorts of stuff, but 0:11:41.920 --> 0:11:45.480 in robots in particular, it is an incredibly important component. 0:11:46.280 --> 0:11:48.880 Uh It is not the only mechanism in robotics that 0:11:48.960 --> 0:11:50.599 does this, and I'm sure I'm going to touch on 0:11:50.880 --> 0:11:54.839 several others in this episode. Now, I do not know 0:11:55.480 --> 0:11:59.679 how long the servo Soldier project lasted, or whether it 0:12:00.000 --> 0:12:04.199 ever got beyond the initial stage of just soliciting design submissions, 0:12:04.640 --> 0:12:06.959 but I do know the piece and armor came out 0:12:07.040 --> 0:12:09.679 a full two years before the first appearance of the 0:12:09.760 --> 0:12:13.720 comic book character Iron Man, and there were other attempts 0:12:13.880 --> 0:12:18.040 at creating exoskeleton technology as far back as the nineteen sixties. 0:12:18.480 --> 0:12:21.160 A project involving the U. S. Army, the U. S. Navy, 0:12:21.280 --> 0:12:26.040 and General Electric or GE produced a prototype exoskeleton dubbed 0:12:26.240 --> 0:12:30.280 Hardyman h A R D I M A N. The 0:12:30.360 --> 0:12:33.600 goal for this project was to design a wearable exoskeleton 0:12:33.720 --> 0:12:36.760 that would give the operator a lifting amplification to a 0:12:36.880 --> 0:12:40.400 factor of twenty five. Now, that would mean you could 0:12:40.440 --> 0:12:44.080 lift a payload of one thousand, two hundred fifty pounds 0:12:44.240 --> 0:12:45.800 and it would feel like you were picking up a 0:12:45.920 --> 0:12:49.480 fifty pound weight. In kilograms, that would be a mass 0:12:49.559 --> 0:12:52.400 of around five hundred sixty seven kilograms, but it would 0:12:52.440 --> 0:12:55.760 feel like a mass of twenty two points seven kilograms. 0:12:56.240 --> 0:12:58.920 That is amazing, and it goes right in line with 0:12:59.000 --> 0:13:02.520 that cargo load I mentioned from the film Aliens. In fact, 0:13:02.559 --> 0:13:05.360 if you look at images of this thing, it kind 0:13:05.440 --> 0:13:08.720 of resembles that fictional piece of hardware. So I imagined 0:13:08.800 --> 0:13:12.959 that when James Cameron was thinking about the film that 0:13:13.400 --> 0:13:16.480 the hardy Man design was one of the influences for 0:13:17.080 --> 0:13:20.000 the cargo loader in that film. The hearty Man was 0:13:20.080 --> 0:13:23.400 technically two systems that were connected together. First, you had 0:13:23.440 --> 0:13:26.120 the pieces that connected to the operator. This is an 0:13:26.280 --> 0:13:31.640 internal exoskeleton, internal with perspective to the device, not to 0:13:31.800 --> 0:13:34.559 the operator. No one needed to have surgery performed in 0:13:34.640 --> 0:13:37.280 order to have this put on them, thank goodness. No, 0:13:37.480 --> 0:13:40.959 this was a skeletal framework that had more than twenty 0:13:41.080 --> 0:13:44.360 five different joints. In fact, I found two separate web pages, 0:13:44.720 --> 0:13:49.160 both on g s own website, that gave the numbers 0:13:49.360 --> 0:13:52.199 of twenty eight joints and thirty joints, so I'd say 0:13:52.240 --> 0:13:55.200 somewhere in that range as a safe bet. Then there 0:13:55.400 --> 0:14:00.600 was a very large mechanical system that connected to this framework. 0:14:01.000 --> 0:14:04.880 So the mechanical system did not connect directly to the operator. 0:14:05.000 --> 0:14:08.920 You had this separate skeletal system that connected to the 0:14:09.000 --> 0:14:12.920 mechanical system, and then the operator connected to the skeletal system. 0:14:13.320 --> 0:14:16.400 But the mechanical system was outfitted with hydraulic and electronic 0:14:16.480 --> 0:14:20.600 systems to provide the force needed to amplify the wearer's strength. 0:14:21.080 --> 0:14:25.200 According to some fairly sketchy records. The design team was 0:14:25.240 --> 0:14:28.120 able to create some form of feedback system, which is 0:14:28.160 --> 0:14:30.920 a really important part of this equation and one that 0:14:31.040 --> 0:14:33.000 really needs to extend all the way back to the 0:14:33.120 --> 0:14:36.040 human operator. So in other words, imagine that you're wearing 0:14:36.080 --> 0:14:39.760 a suit, but there's no feedback, Like you can't tell 0:14:40.320 --> 0:14:44.400 the difference between pushing lightly and pushing really hard, So 0:14:44.480 --> 0:14:46.520 you wouldn't know if you were using enough pressure to 0:14:46.600 --> 0:14:49.360 pick up a heavy load, or maybe you might use 0:14:49.480 --> 0:14:51.680 too much then you damage whatever it is you're trying 0:14:51.720 --> 0:14:55.560 to pick up. So in one case, you might try 0:14:55.680 --> 0:14:58.400 to lift something but because of the pressure you're using 0:14:58.480 --> 0:15:01.880 with your hands, the grip isn't strong enough and that's 0:15:01.920 --> 0:15:05.680 something slips from your grasp, which could be potentially very dangerous, 0:15:06.360 --> 0:15:10.280 or you're using way too much pressure and you actually 0:15:10.360 --> 0:15:14.200 cause damage to the thing you're trying to move. Exoskeletons 0:15:14.240 --> 0:15:16.680 need a way to indicate to the operator what the 0:15:16.760 --> 0:15:19.280 interactions with the environment are actually like, there needs to 0:15:19.320 --> 0:15:23.840 be some form of feedback, whether it's audible, visual, or tactile. 0:15:23.880 --> 0:15:27.160 If it's a haptic feedback system, maybe a rumble system, 0:15:27.240 --> 0:15:30.360 so that way you can tell how much pressure you're 0:15:30.480 --> 0:15:35.240 using and the rumble increases in amplitude as you start 0:15:35.280 --> 0:15:37.960 to add more pressure. Something needs to be there in 0:15:38.120 --> 0:15:41.080 order for the operator to have a sense of how 0:15:41.200 --> 0:15:44.480 much force they are using. And while the scant records 0:15:44.520 --> 0:15:47.000 on the Hearty Man suggested that the team found some 0:15:47.400 --> 0:15:51.520 solutions to this problem, they also don't go into any detail, 0:15:51.960 --> 0:15:55.120 so if they did find solutions to it, I don't 0:15:55.160 --> 0:15:58.640 know what those were. The Hearty Man never progressed beyond 0:15:58.840 --> 0:16:02.840 the prototype stage and reportedly it never had anyone inside 0:16:02.920 --> 0:16:05.400 it while it was actually turned on, not not the 0:16:05.520 --> 0:16:08.600 full suit. Maybe parts of it, but not the full thing. 0:16:09.000 --> 0:16:12.160 There are photos of people in the suit, presumably the 0:16:12.240 --> 0:16:14.640 suit was off at the time those photos were taken. 0:16:15.240 --> 0:16:17.080 But the question is what went wrong? Why didn't this 0:16:17.200 --> 0:16:20.560 thing get further refined and go into production. Well, for 0:16:20.680 --> 0:16:23.880 one thing, the suit was just impractical. The suit itself 0:16:24.120 --> 0:16:27.960 weighed in at a hefty one thousand, five hundred pounds 0:16:28.040 --> 0:16:32.840 that's about six ms. Providing power to it was a challenge. 0:16:33.120 --> 0:16:36.120 I'm guessing it required a tethered power line at the time, 0:16:36.760 --> 0:16:40.400 and it wasn't exactly the most stable of suits. According 0:16:40.440 --> 0:16:43.920 to reports, they had this tiny issue of quote violent 0:16:44.160 --> 0:16:48.640 and uncontrollable motion in the quote, which, as you can imagine, 0:16:48.920 --> 0:16:51.920 is not the best feature if you're putting someone in 0:16:52.040 --> 0:16:54.880 the darn thing. And so it never went further than 0:16:54.920 --> 0:16:58.040 the prototype stage. When we come back, I'll give a 0:16:58.120 --> 0:17:01.560 word or two about hydraulics systems, and then we'll continue 0:17:01.720 --> 0:17:13.720 looking at exoskeleton projects. But first let's take a quick break. Okay, 0:17:13.760 --> 0:17:16.360 before the break I mentioned, I would talk about hydraulic 0:17:16.440 --> 0:17:20.280 systems and why they are used in heavy duty, industrial 0:17:20.480 --> 0:17:23.440 strength machinery. And this has a lot to do with physics, 0:17:23.560 --> 0:17:27.800 particularly the physics of specific types of liquids. Now, basically, 0:17:27.920 --> 0:17:31.600 hydraulic systems are using some form of pressurized fluid to 0:17:31.800 --> 0:17:35.520 do work. So imagine you've got a pair of syringes 0:17:36.160 --> 0:17:39.520 and the first syringe is filled with a liquid and 0:17:40.000 --> 0:17:44.040 it's connected by a tube which is also filled with liquid. 0:17:44.119 --> 0:17:47.280 There's no gaps there, and this is connected to an 0:17:47.359 --> 0:17:52.400 empty syringe. But this second syringe has its plunger fully depressed, 0:17:52.840 --> 0:17:55.280 so there's nowhere in the system. Right, You've got a 0:17:55.400 --> 0:17:59.760 fully filled syringe with liquid in it plunger fully back. 0:18:00.240 --> 0:18:03.240 You've got a tube filled with liquid connected to the 0:18:03.520 --> 0:18:06.400 end of a second syringe. With its plunger fully depressed. 0:18:06.920 --> 0:18:10.200 If you push on the full syringe is plunger, then 0:18:10.280 --> 0:18:12.879 the plunger pushes against the liquid, and the liquid in 0:18:12.960 --> 0:18:16.320 turn pushes against the plunger on the empty syringe, and 0:18:16.440 --> 0:18:20.680 it pushes that plunger outward as the first syringe empties. 0:18:21.119 --> 0:18:25.760 So fluids like water don't compress very much, so when 0:18:25.800 --> 0:18:28.399 you push on them, they transfer that force in whatever 0:18:28.480 --> 0:18:32.080 direction happens to be available. In fact, they press outward 0:18:32.119 --> 0:18:36.359 in every direction without diminishing the force. But if the 0:18:36.440 --> 0:18:40.440 fluid hits an unyielding edge, then the pressure will act 0:18:40.560 --> 0:18:44.040 against that space at a right angle. This is called 0:18:44.119 --> 0:18:48.200 Pascal's law after Blaze Pascal. And practically what this means 0:18:48.280 --> 0:18:50.600 is you can apply force to a small amount of 0:18:50.640 --> 0:18:53.760 liquid and get a lot of power as an output 0:18:54.200 --> 0:18:58.119 as you get a proportionally bigger force on a bigger area. 0:18:58.640 --> 0:19:00.480 And in the example I just mentioned, you can easily 0:19:00.520 --> 0:19:04.040 imagine reversing this process simply by picking up syringe number 0:19:04.119 --> 0:19:08.000 two that is now full and pushing down on its plunger, 0:19:08.200 --> 0:19:11.119 forcing the pressurized liquid to push back against the plunger 0:19:11.240 --> 0:19:14.320 of syringe number one. And there are several parts to 0:19:14.440 --> 0:19:18.200 a typical hydraulic system, but that's not super important for 0:19:18.280 --> 0:19:21.320 this episode. I've done a few episodes talking about hydraulic 0:19:21.400 --> 0:19:24.000 systems and greater detail in the past, but the important 0:19:24.000 --> 0:19:26.679 things to remember about hydraulic systems is that they can 0:19:26.760 --> 0:19:31.360 generate a lot of power output, they can do very 0:19:31.520 --> 0:19:37.320 heavy duty work with a relatively small amount of input required, 0:19:37.840 --> 0:19:42.280 and hydraulic machines can lift heavy payloads, so they are 0:19:42.440 --> 0:19:48.720 often part of heavy duty industrial equipment. We're talking about 0:19:48.760 --> 0:19:54.760 stuff like construction machines, you know, cranes and bulldozers and 0:19:55.080 --> 0:19:56.919 back hose and that kind of thing, and they are 0:19:57.040 --> 0:20:00.360 part of heavy duty exoskeletons as a result as well. 0:20:00.880 --> 0:20:03.560 And I should also mention the difference between hydraulic and 0:20:03.680 --> 0:20:07.880 pneumatic systems. Hydraulics rely on pressurized liquids, and again, those 0:20:07.960 --> 0:20:11.919 don't compress. Pneumatic systems are kind of similar, but they 0:20:12.000 --> 0:20:16.560 rely on pressurized gas and gas does compress. So hydraulic 0:20:16.600 --> 0:20:19.879 systems are better for those super heavy duty functions and 0:20:19.960 --> 0:20:23.119 they are incredibly precise, but they are also slower than 0:20:23.200 --> 0:20:27.200 pneumatic systems. Pneumatic systems are better for smaller applications that 0:20:27.320 --> 0:20:30.200 don't require as much force or precision, but they do 0:20:30.400 --> 0:20:35.800 require faster action. So there are two different technologies that 0:20:35.880 --> 0:20:39.640 work on similar principles, but you would want to choose 0:20:39.720 --> 0:20:42.399 one or the other based upon whatever application you had 0:20:42.440 --> 0:20:46.399 in mind. While ge produced a powered exoskeleton that was 0:20:46.480 --> 0:20:49.119 potentially too dangerous to put someone in when it was 0:20:49.200 --> 0:20:52.800 you know, actually on, a team at Cornell led by 0:20:53.040 --> 0:20:55.960 Neil Maisen was working on a different design called the 0:20:56.320 --> 0:21:01.640 man Amplifier, which yikes, anyway, that name seems pretty loaded 0:21:01.680 --> 0:21:04.280 to me. But the concept was similar to gas approach, 0:21:04.320 --> 0:21:06.720 although it was meant to be much smaller and it 0:21:06.760 --> 0:21:10.120 would also rely on hydraulics and servos and it would 0:21:10.119 --> 0:21:14.920 also amplify the operator's strength. The team designed the underlying exoskeleton, 0:21:15.200 --> 0:21:17.320 which was the part that would attach directly to the 0:21:17.440 --> 0:21:22.040 operator's body. This was the unpowered section. They never reached 0:21:22.080 --> 0:21:24.440 a point in which they were able to add powered 0:21:24.560 --> 0:21:29.200 components to it, but designing just the skeletal structure itself 0:21:29.320 --> 0:21:31.520 was a big deal because they had to figure out, 0:21:31.600 --> 0:21:34.719 how can we do something that would support weight, How 0:21:34.800 --> 0:21:37.800 can we do it so that it moves with the operator, 0:21:38.320 --> 0:21:41.720 and that it uh it interferes with the operator's range 0:21:41.760 --> 0:21:45.080 of movement as little as possible, and those are all 0:21:45.720 --> 0:21:50.720 conflicting priorities. There's no easy way to satisfy all of 0:21:50.880 --> 0:21:55.280 those requirements. Over in Europe, at the Mahilo Institute in Serbia, 0:21:55.440 --> 0:21:59.919 Professor Miomir Vukabanovich led efforts to develop a wearable system 0:22:00.320 --> 0:22:03.359 that would replicate the walking gait of a human, with 0:22:03.440 --> 0:22:06.040 the ultimate goal of providing assistance to people who had 0:22:06.119 --> 0:22:09.560 limited or no mobility in their legs. The work in 0:22:09.680 --> 0:22:14.000 that field would also advance our understanding of bipedal robotics 0:22:14.200 --> 0:22:16.920 systems in general, meaning that while the goal was to 0:22:17.000 --> 0:22:20.119 create an exoskeleton to give mobility to people who might 0:22:20.240 --> 0:22:23.919 otherwise not have that independence, it also contributed to our 0:22:24.000 --> 0:22:27.359 progress towards bipedal robots, you know, robots with two legs 0:22:27.760 --> 0:22:31.359 that are capable of maintaining their balance and also moving 0:22:31.480 --> 0:22:36.240 through environments. This is easier said than done. It's actually 0:22:36.440 --> 0:22:41.560 incredibly difficult to achieve. So the reason why this is 0:22:41.600 --> 0:22:43.840 such a hard challenge is that you've got to think 0:22:43.880 --> 0:22:47.399 about what walking actually is. You know, for a lot 0:22:47.440 --> 0:22:50.040 of people, walking is just something we can do without 0:22:50.119 --> 0:22:52.040 really thinking about it. You know, you just you just 0:22:52.160 --> 0:22:55.000 do it. But if you want to replicate walking, you 0:22:55.119 --> 0:22:57.399 have to figure out what walking really is from a 0:22:57.760 --> 0:23:00.240 physics perspective. You've got to figure out stuf off like 0:23:00.680 --> 0:23:05.480 momentum and balance. You have to marry a stable repeated 0:23:05.600 --> 0:23:09.960 phase that is moving the legs in a pattern that 0:23:10.480 --> 0:23:15.040 the works and creates a stable, you know system. But 0:23:15.160 --> 0:23:17.520 you also have to do that within the context of 0:23:17.600 --> 0:23:21.600 an unstable phase. And by that I mean Bipedal walking 0:23:22.200 --> 0:23:25.159 is essentially a series of shifting our balance forward so 0:23:25.320 --> 0:23:30.040 that we're falling. Walking is just falling and catching ourselves 0:23:30.160 --> 0:23:33.280 over and over by extending a foot and then we 0:23:33.400 --> 0:23:35.280 catch ourselves with that foot. We shift our weight and 0:23:35.320 --> 0:23:38.960 our balance so that we continue our fall forward, but 0:23:39.040 --> 0:23:42.320 we catch ourselves with our other foot. Now, this process 0:23:43.040 --> 0:23:45.520 is natural to us. It is easy for us to do, 0:23:45.680 --> 0:23:47.960 and we don't feel like we're falling, right, We're just 0:23:48.320 --> 0:23:51.680 pushing ourselves forward. But if you were to try and 0:23:51.920 --> 0:23:55.080 replicate that in a machine system, you literally are having 0:23:55.160 --> 0:23:58.440 to cause the machine to fall forward and then catch 0:23:58.520 --> 0:24:02.680 itself over it over again, And that requires you to 0:24:03.359 --> 0:24:06.320 figure out how to do things like maintain balance, how 0:24:06.440 --> 0:24:10.280 do you maintain speed, how do you allow for stopping, 0:24:10.560 --> 0:24:13.600 how do you allow for the change of direction. I mean, 0:24:13.680 --> 0:24:17.000 all of the acceleration forces are really tricky. And one 0:24:17.119 --> 0:24:20.320 concept that Vukabanovich put forth was what he called the 0:24:20.520 --> 0:24:24.440 zero moment point, and this describes a condition in which 0:24:24.520 --> 0:24:28.440