1 00:00:04,160 --> 00:00:07,480 Speaker 1: Get in text with technology with tech Stuff from staff 2 00:00:07,520 --> 00:00:13,800 Speaker 1: works dot com. Hey there, and welcome to tech Stuff. 3 00:00:13,840 --> 00:00:16,840 Speaker 1: I am your host, Jonathan Strickland. I'm an executive producer 4 00:00:16,880 --> 00:00:19,079 Speaker 1: here at how Stuff Works, and I love all things tech, 5 00:00:19,600 --> 00:00:22,760 Speaker 1: and today I thought I would look into a pretty 6 00:00:22,840 --> 00:00:27,120 Speaker 1: cool company, one that's been around longer than I realized, 7 00:00:27,640 --> 00:00:29,880 Speaker 1: and one that has created a lot of interesting stuff 8 00:00:29,920 --> 00:00:32,639 Speaker 1: and yet nothing that you can actually buy. So when 9 00:00:32,640 --> 00:00:35,080 Speaker 1: am I talking about? Well, a few years ago, there 10 00:00:35,080 --> 00:00:38,080 Speaker 1: were some videos that came out of a company called 11 00:00:38,120 --> 00:00:42,720 Speaker 1: Boston Dynamics that went absolutely viral, and those videos showed 12 00:00:42,880 --> 00:00:47,280 Speaker 1: four legged robots traversing various types of terrain, sometimes while 13 00:00:47,360 --> 00:00:52,120 Speaker 1: enduring some rather abusive behavior courtesy of some humans, and 14 00:00:52,400 --> 00:00:55,120 Speaker 1: that kind of served as an introduction for a lot 15 00:00:55,160 --> 00:00:57,880 Speaker 1: of people to this company, Boston Dynamics. But as it 16 00:00:57,920 --> 00:01:01,440 Speaker 1: turns out, it had been around a lot longer than 17 00:01:01,600 --> 00:01:04,479 Speaker 1: you might have anticipated based on those videos. In fact, 18 00:01:04,520 --> 00:01:08,480 Speaker 1: it turned twenty five years old in seventeen. So what 19 00:01:08,760 --> 00:01:12,959 Speaker 1: is the story behind the company and its robots. Well, 20 00:01:13,000 --> 00:01:17,080 Speaker 1: the founder of Boston Dynamics is Mark rayber. Raber was 21 00:01:17,120 --> 00:01:20,679 Speaker 1: born in nineteen forty nine and he was a studious Fellow. 22 00:01:21,160 --> 00:01:25,720 Speaker 1: He attended Northeastern University from nineteen sixty eight to nineteen 23 00:01:25,840 --> 00:01:29,120 Speaker 1: seventy three and earned a degree in electrical engineering. He 24 00:01:29,200 --> 00:01:32,479 Speaker 1: then went on to pursue his PhD at the Massachusetts 25 00:01:32,520 --> 00:01:36,959 Speaker 1: Institute of Technology better known as m i T, and 26 00:01:37,040 --> 00:01:40,240 Speaker 1: upon graduating, Dr Rayber got a job at the NASA 27 00:01:40,319 --> 00:01:43,880 Speaker 1: Jet Propulsion Laboratory as a member of the technacal staff. 28 00:01:43,959 --> 00:01:47,600 Speaker 1: So you could say that he was a rocket scientist, 29 00:01:47,880 --> 00:01:51,480 Speaker 1: although really he worked with rocket scientists. His main focus 30 00:01:51,520 --> 00:01:53,919 Speaker 1: was obviously on robotics, and he worked there for about 31 00:01:53,920 --> 00:01:58,200 Speaker 1: three years, leaving in nineteen eighty. His next gig was 32 00:01:58,680 --> 00:02:01,480 Speaker 1: a pretty just went as well. He was an associate 33 00:02:01,480 --> 00:02:05,040 Speaker 1: professor at Carnegie Melon University's Computer science department and the 34 00:02:05,160 --> 00:02:08,920 Speaker 1: Robotics Institute. Now, if you've been listening to tech stuff recently, 35 00:02:09,000 --> 00:02:11,840 Speaker 1: you may remember I mentioned this particular group in a 36 00:02:11,880 --> 00:02:17,360 Speaker 1: recent episode about Uber, because Uber rated Carnegie Melon and 37 00:02:17,480 --> 00:02:21,680 Speaker 1: hired about forty of their robotics experts over to help 38 00:02:21,760 --> 00:02:26,920 Speaker 1: them with their efforts in developing autonomous car programs. Of course, 39 00:02:26,960 --> 00:02:30,520 Speaker 1: by that time Dr Raber was long gone from Carnegie Melon. 40 00:02:30,600 --> 00:02:33,440 Speaker 1: But I just I mentioned it because it's another link 41 00:02:33,520 --> 00:02:37,320 Speaker 1: to a recent episode. In nine six, Dr rayber took 42 00:02:37,360 --> 00:02:41,120 Speaker 1: a position of Professor of Electrical Engineering and Computer Science 43 00:02:41,160 --> 00:02:43,360 Speaker 1: over at m I T, and there he founded the 44 00:02:43,480 --> 00:02:46,639 Speaker 1: m I T Leg Lab, which sounds a bit weird, 45 00:02:47,480 --> 00:02:50,160 Speaker 1: but stick with me. So this was a lab specifically 46 00:02:50,200 --> 00:02:55,280 Speaker 1: dedicated to researching and designing legged robots, not leg lamps, 47 00:02:55,320 --> 00:02:58,560 Speaker 1: as I was first led to believe. It was a 48 00:02:58,680 --> 00:03:02,920 Speaker 1: very difficult realization to say that it wasn't a major reward. 49 00:03:03,000 --> 00:03:05,280 Speaker 1: But it turns out building robots that use legs for 50 00:03:05,360 --> 00:03:10,679 Speaker 1: locomotion it's really really hard. Now why is that? Why 51 00:03:10,720 --> 00:03:13,280 Speaker 1: is it so difficult to build a robot that uses 52 00:03:13,400 --> 00:03:17,680 Speaker 1: legs to get around? Well, first, let's consider the wheel. Now, 53 00:03:18,360 --> 00:03:22,760 Speaker 1: wheels aren't really found in organisms. There are some microscopic 54 00:03:22,840 --> 00:03:26,680 Speaker 1: critters that have components that are vaguely wheel and axle like, 55 00:03:27,480 --> 00:03:32,240 Speaker 1: but they're not really wheels the way are various inventions 56 00:03:32,280 --> 00:03:35,640 Speaker 1: have wheels. A wheel as a limb is not likely 57 00:03:36,120 --> 00:03:39,480 Speaker 1: to ever evolve in nature for several reasons. For one, 58 00:03:39,960 --> 00:03:43,040 Speaker 1: any mutation that would set an organism on the path 59 00:03:43,760 --> 00:03:47,840 Speaker 1: figuratively speaking to developing wheels would likely at first be 60 00:03:47,920 --> 00:03:51,520 Speaker 1: a negative rather than a positive mutation. So in other words, 61 00:03:51,720 --> 00:03:54,680 Speaker 1: an animal with some sort of pro to wheel limb, 62 00:03:54,880 --> 00:03:57,560 Speaker 1: something that's maybe like part of a wheel but not 63 00:03:57,640 --> 00:04:01,240 Speaker 1: a complete wheel, probably wouldn't have very many advantages and 64 00:04:01,280 --> 00:04:06,240 Speaker 1: perhaps have distinct disadvantages with regards to survival, and thus 65 00:04:06,360 --> 00:04:09,200 Speaker 1: it would be unlikely that such a mutation would be 66 00:04:09,200 --> 00:04:13,760 Speaker 1: passed down to future generations to evolve. So that's one thing, 67 00:04:13,840 --> 00:04:17,560 Speaker 1: is that the pathway for evolution is complicated, and if 68 00:04:17,760 --> 00:04:21,279 Speaker 1: you have a mutation that is more likely to get 69 00:04:21,320 --> 00:04:24,960 Speaker 1: you eaten by predators, probably not going to get passed 70 00:04:24,960 --> 00:04:27,840 Speaker 1: down to a lot of offspring. For another reason, the 71 00:04:27,960 --> 00:04:31,200 Speaker 1: rotation of a wheel, if not due to gravity, has 72 00:04:31,240 --> 00:04:34,680 Speaker 1: to come from some source of torque, that rotational power. 73 00:04:34,760 --> 00:04:37,120 Speaker 1: You have to be able to provide power to a 74 00:04:37,120 --> 00:04:41,000 Speaker 1: wheel to make it turn. So locomotion for multicellular organisms 75 00:04:41,040 --> 00:04:44,520 Speaker 1: typically comes from muscles, right, we have our muscles that 76 00:04:44,560 --> 00:04:48,200 Speaker 1: allow us to move our limbs. But there's not really 77 00:04:48,240 --> 00:04:51,479 Speaker 1: a way you could affix muscles to a wheel and 78 00:04:51,520 --> 00:04:55,160 Speaker 1: allow it to turn freely indefinitely. You know, just think 79 00:04:55,200 --> 00:04:57,760 Speaker 1: about how your wrist turns. Like even if you had 80 00:04:58,360 --> 00:05:01,960 Speaker 1: the ability to turn your wrist completely around with no 81 00:05:02,080 --> 00:05:05,000 Speaker 1: pain or anything like that, you couldn't do it indefinitely. 82 00:05:05,040 --> 00:05:08,240 Speaker 1: Your muscles would continuously twist until eventually they would have 83 00:05:08,279 --> 00:05:10,960 Speaker 1: to untwist like a rubber band. You couldn't just keep ongoing, 84 00:05:11,080 --> 00:05:15,680 Speaker 1: so that's another problem. Wheels are almost exclusively in the 85 00:05:15,720 --> 00:05:21,000 Speaker 1: domain of manufactured objects. Now that being said, they're darn't efficient. 86 00:05:21,279 --> 00:05:25,039 Speaker 1: They're the easiest way to create locomotion for vehicles. It 87 00:05:25,120 --> 00:05:27,960 Speaker 1: is also the most energy efficient way to travel relative 88 00:05:28,040 --> 00:05:32,120 Speaker 1: to speed. So it requires a relatively simple implementation to 89 00:05:32,200 --> 00:05:34,839 Speaker 1: create a wheeled robot. You don't have to worry so 90 00:05:34,920 --> 00:05:37,039 Speaker 1: much about degrees of freedom. You don't have to worry 91 00:05:37,040 --> 00:05:41,040 Speaker 1: about pivot points, or vector controllers or balance. I mean, 92 00:05:41,080 --> 00:05:43,080 Speaker 1: as long as you've designed the robot in such a 93 00:05:43,120 --> 00:05:46,359 Speaker 1: way that's relatively stable on its wheels, balance is not 94 00:05:46,440 --> 00:05:48,080 Speaker 1: that big of an issue. So if you're talking about 95 00:05:48,120 --> 00:05:51,720 Speaker 1: the three wheeled or four wheeled robot, balance is pretty easy. 96 00:05:51,760 --> 00:05:53,880 Speaker 1: Once you get down to two wheels, then things get 97 00:05:53,880 --> 00:05:55,920 Speaker 1: a little more tricky. You have to build in other 98 00:05:56,000 --> 00:05:58,279 Speaker 1: systems in order for it to maintain balance, but in 99 00:05:58,320 --> 00:06:02,279 Speaker 1: general it's not that hard to do. They also typically 100 00:06:02,279 --> 00:06:05,919 Speaker 1: cost less to develop and less to build than a 101 00:06:06,000 --> 00:06:09,839 Speaker 1: legged robot. A simple three wheel robot is stable on 102 00:06:09,839 --> 00:06:12,560 Speaker 1: its own. If you've got that basic sort of triangle shape, 103 00:06:12,760 --> 00:06:14,960 Speaker 1: it does not require a lot of high end technology 104 00:06:15,000 --> 00:06:17,960 Speaker 1: to work, just some motors and then whatever sensors and 105 00:06:18,080 --> 00:06:21,920 Speaker 1: processors you need to actually control the robot. Uh. Four 106 00:06:21,920 --> 00:06:25,240 Speaker 1: wheeled robots are are even more stable than three wheeled robots, 107 00:06:25,279 --> 00:06:27,680 Speaker 1: and they can excel in applications that a three wheeled 108 00:06:27,720 --> 00:06:30,719 Speaker 1: robot might find difficult. For example, if you plan for 109 00:06:30,760 --> 00:06:33,200 Speaker 1: your robot to lift things, you could have a three 110 00:06:33,200 --> 00:06:36,839 Speaker 1: wheeled robot that would become unbalanced depending upon where you 111 00:06:36,920 --> 00:06:39,560 Speaker 1: put the load on top of the robot right If 112 00:06:39,600 --> 00:06:42,080 Speaker 1: it's too far off to one side of its center 113 00:06:42,120 --> 00:06:44,360 Speaker 1: of gravity, it can make the robots sort of tip over. 114 00:06:45,040 --> 00:06:47,320 Speaker 1: Our four wheeled robot tends to be more stable in 115 00:06:47,320 --> 00:06:50,520 Speaker 1: those cases. You can also design a wheeled robot to 116 00:06:50,520 --> 00:06:54,120 Speaker 1: operate even if it's been flipped upside down. So if 117 00:06:54,160 --> 00:06:56,800 Speaker 1: the robot is such that the wheels can make contact, 118 00:06:56,880 --> 00:06:59,760 Speaker 1: whether it's right side up or upside down, you could 119 00:07:00,000 --> 00:07:03,360 Speaker 1: of the option of continuously operating that robot even if 120 00:07:03,360 --> 00:07:05,320 Speaker 1: it were to flip over. In fact, I remember some 121 00:07:05,480 --> 00:07:07,800 Speaker 1: remote controlled cars that were designed this way. They were 122 00:07:07,839 --> 00:07:11,320 Speaker 1: marketed as that you would purposefully drive your car and 123 00:07:11,360 --> 00:07:13,320 Speaker 1: to say a wall, make it flip over, and then 124 00:07:13,320 --> 00:07:16,000 Speaker 1: you could keep on driving the car just kind of fun. 125 00:07:16,520 --> 00:07:19,400 Speaker 1: That's the same basic idea, but wheels are not ideal 126 00:07:19,440 --> 00:07:22,880 Speaker 1: across all different types of terrain. They work best on 127 00:07:23,000 --> 00:07:26,239 Speaker 1: tracks if you have them, but if you don't have tracks, 128 00:07:26,280 --> 00:07:30,480 Speaker 1: they work best on smooth even ground. Depending upon the 129 00:07:30,520 --> 00:07:33,400 Speaker 1: wheel design and the torque available, they may be able 130 00:07:33,480 --> 00:07:37,280 Speaker 1: to travel over semi rough terrain too fairly rough terrain, 131 00:07:37,360 --> 00:07:39,640 Speaker 1: but they aren't great for everything, and if the ground 132 00:07:39,680 --> 00:07:42,720 Speaker 1: is really challenging, they can prove to be ineffective. So 133 00:07:42,760 --> 00:07:44,760 Speaker 1: that challenge could be that the ground is just too 134 00:07:44,840 --> 00:07:48,200 Speaker 1: rocky or covered in debris, or it's composed of loose 135 00:07:48,240 --> 00:07:51,640 Speaker 1: elements like sand, and in those situations, legs might be 136 00:07:51,640 --> 00:07:56,400 Speaker 1: more effective. Your wheels might otherwise spin needlessly, just completely 137 00:07:56,720 --> 00:08:00,000 Speaker 1: without motion. You're just turning around and you're digging yourself 138 00:07:59,840 --> 00:08:03,440 Speaker 1: into a hole. Or they could end up catching on 139 00:08:03,560 --> 00:08:06,320 Speaker 1: debris and pulling them into the works of the robot, 140 00:08:06,520 --> 00:08:10,480 Speaker 1: gumming everything up. A legged robot with the right control 141 00:08:10,520 --> 00:08:14,520 Speaker 1: system or programming can step over obstacles and move smoothly 142 00:08:14,600 --> 00:08:17,400 Speaker 1: over different types of terrain, robots can use legs to 143 00:08:17,440 --> 00:08:19,960 Speaker 1: select a precise spot on the ground to place its 144 00:08:20,000 --> 00:08:23,600 Speaker 1: weight and then shift to move. A wheel has little 145 00:08:23,680 --> 00:08:26,480 Speaker 1: choice but to just roll ahead, but a legged robot 146 00:08:26,520 --> 00:08:28,720 Speaker 1: can place its legs and then shift its weight in 147 00:08:28,760 --> 00:08:31,800 Speaker 1: a very specific way, which is helpful on uneven ground 148 00:08:31,920 --> 00:08:33,920 Speaker 1: or going up and down the stairs. And depending upon 149 00:08:33,920 --> 00:08:36,319 Speaker 1: the design of the robot, you can make it where 150 00:08:36,880 --> 00:08:39,440 Speaker 1: the way it places its leg down, the way it 151 00:08:39,480 --> 00:08:42,559 Speaker 1: shifts its weight, it can maintain its relative position over 152 00:08:42,600 --> 00:08:44,320 Speaker 1: the ground. So it looks like if you if you 153 00:08:44,360 --> 00:08:47,640 Speaker 1: were to remove the legs magically from the image, like 154 00:08:47,679 --> 00:08:51,600 Speaker 1: it's just floating over the landscape. But it's all because 155 00:08:51,679 --> 00:08:53,679 Speaker 1: the way it's able to place its legs and move 156 00:08:53,720 --> 00:08:57,360 Speaker 1: its weight around. But legs, as it turns out, are 157 00:08:57,400 --> 00:09:01,559 Speaker 1: super hard to design. They require joints and points of articulation, 158 00:09:01,920 --> 00:09:04,000 Speaker 1: and you have to have power to move everything and 159 00:09:04,040 --> 00:09:07,760 Speaker 1: figure out how you're going to create the force necessary 160 00:09:07,800 --> 00:09:10,840 Speaker 1: to open and close those various joints, and you have 161 00:09:10,880 --> 00:09:13,680 Speaker 1: to have lots of precise directions to work effectively. So 162 00:09:13,720 --> 00:09:16,840 Speaker 1: you need a very efficient computer and you need really 163 00:09:16,840 --> 00:09:20,240 Speaker 1: good programming so that it can actually maneuver over these 164 00:09:20,280 --> 00:09:23,679 Speaker 1: different terrain. A leg needs to have more moving parts 165 00:09:23,720 --> 00:09:26,040 Speaker 1: than a wheel, with at least three actuators to make 166 00:09:26,080 --> 00:09:28,440 Speaker 1: the leg more useful than a wheel. You can have 167 00:09:28,520 --> 00:09:30,959 Speaker 1: fewer than three actuators and have a leg, but it's 168 00:09:30,960 --> 00:09:32,679 Speaker 1: not going to be more useful than a wheel unless 169 00:09:32,679 --> 00:09:36,000 Speaker 1: you have at least three actuators. So a lot of 170 00:09:36,000 --> 00:09:39,880 Speaker 1: the work in legged robots concentrates on how to move 171 00:09:40,000 --> 00:09:44,040 Speaker 1: those legs. So, for example, you might use artificial muscles 172 00:09:44,080 --> 00:09:48,280 Speaker 1: with stuff like electroactive polymers. Now, those are materials made 173 00:09:48,280 --> 00:09:51,040 Speaker 1: out of long chain molecules that change their shape when 174 00:09:51,080 --> 00:09:54,240 Speaker 1: they encounter an electric field. So here's an example, and 175 00:09:54,240 --> 00:09:56,000 Speaker 1: this is just one I pulled off the top of 176 00:09:56,040 --> 00:09:58,600 Speaker 1: my head, and it's not really indicative of how they 177 00:09:58,600 --> 00:10:01,280 Speaker 1: all work, but it's way they at work. Let's say 178 00:10:01,280 --> 00:10:03,760 Speaker 1: that you build a structure and it's made off of 179 00:10:03,760 --> 00:10:08,000 Speaker 1: this material, and that structure is elongated, it's extended, but 180 00:10:08,040 --> 00:10:11,400 Speaker 1: if it encounters an electric field, it contracts, it pulls together. 181 00:10:11,880 --> 00:10:14,040 Speaker 1: This would make it work in a way similar to 182 00:10:14,080 --> 00:10:18,480 Speaker 1: the way our muscles work. Now, there's not just that 183 00:10:18,600 --> 00:10:21,160 Speaker 1: kind of method to power legs. In fact, that's a 184 00:10:21,160 --> 00:10:25,280 Speaker 1: pretty rare one. You're more likely to find electric actuators 185 00:10:25,360 --> 00:10:29,319 Speaker 1: or hydraulic actuators. But that's another challenge is how do 186 00:10:29,360 --> 00:10:32,160 Speaker 1: you actually get the legs to move. Then there's a 187 00:10:32,200 --> 00:10:34,680 Speaker 1: ton of work on that processing side to calculate things 188 00:10:34,679 --> 00:10:38,959 Speaker 1: like balanced momentum, weight distribution, path finding, all of these 189 00:10:39,000 --> 00:10:42,640 Speaker 1: different things that we take for granted we learn it intuitively, 190 00:10:43,120 --> 00:10:47,080 Speaker 1: but that doesn't work with robots necessarily. And there's a 191 00:10:47,120 --> 00:10:50,199 Speaker 1: long history of studies that happened before robotics were even 192 00:10:50,240 --> 00:10:54,360 Speaker 1: a thing that would eventually inform the design of legged robots, 193 00:10:54,360 --> 00:10:57,600 Speaker 1: and some of those studies which went on to help 194 00:10:57,720 --> 00:11:00,960 Speaker 1: future engineers when they were building out these robot legs 195 00:11:01,000 --> 00:11:04,360 Speaker 1: for the first time. We're pretty grim, as it turns out. 196 00:11:04,760 --> 00:11:09,439 Speaker 1: So get ready for a gross story that I'm about 197 00:11:09,440 --> 00:11:13,320 Speaker 1: to tell you. One study that the Leg Lab cites 198 00:11:13,400 --> 00:11:16,720 Speaker 1: as being an important milestone in the development of legged 199 00:11:16,840 --> 00:11:21,960 Speaker 1: robots comes from an eighteen thirty six journal and it 200 00:11:22,000 --> 00:11:27,040 Speaker 1: involves dead people. The title of this study was Mechanic 201 00:11:27,360 --> 00:11:33,079 Speaker 1: derminsch lichen I and a Thomish physiologic unter schung fondap 202 00:11:33,120 --> 00:11:38,040 Speaker 1: rudin Wilhelm Weber w Edward web Webber. All right, so 203 00:11:38,120 --> 00:11:40,280 Speaker 1: translated to English because that was a terrible German, I know, 204 00:11:40,360 --> 00:11:44,840 Speaker 1: my German's awful kind of Deutsches. It turns out it 205 00:11:44,920 --> 00:11:49,040 Speaker 1: actually means the mechanics of the human and anatomical physiological 206 00:11:49,080 --> 00:11:53,120 Speaker 1: examination of the brothers Wilhelm and Edward Webber. Now their 207 00:11:53,120 --> 00:11:58,040 Speaker 1: work is available to read if you read German. But 208 00:11:58,200 --> 00:11:59,760 Speaker 1: they found that if you took the leg of a 209 00:12:00,000 --> 00:12:04,360 Speaker 1: corps and you swung the leg of the corpse, so 210 00:12:04,440 --> 00:12:08,920 Speaker 1: it acted like a compound pendulum. The swinging motion was 211 00:12:09,040 --> 00:12:10,959 Speaker 1: very similar to the cadence of a leg with a 212 00:12:10,960 --> 00:12:16,880 Speaker 1: live person is walking around. That's cheerful. But as it 213 00:12:16,920 --> 00:12:19,640 Speaker 1: turns out, that would actually be one of those fundamental 214 00:12:19,720 --> 00:12:22,800 Speaker 1: studies that would go on to inform people who were 215 00:12:22,840 --> 00:12:27,120 Speaker 1: trying to build artificial legs and mechanical legs. Most of 216 00:12:27,120 --> 00:12:30,560 Speaker 1: the studies the lab sites aren't nearly so grim and 217 00:12:30,640 --> 00:12:35,040 Speaker 1: don't involve nearly as many corpses. So for example, another 218 00:12:35,080 --> 00:12:38,800 Speaker 1: important contribution came from a guy named Edward Moybridge. Now 219 00:12:38,840 --> 00:12:42,400 Speaker 1: this was an English photographer who developed stop motion photography. 220 00:12:42,720 --> 00:12:46,199 Speaker 1: Moybridge was hired by a fat cat named Leland Stanford. 221 00:12:46,200 --> 00:12:49,600 Speaker 1: He was a former governor and a railroad tycoon and 222 00:12:49,640 --> 00:12:53,040 Speaker 1: Stanford had this idea, and he needed evidence to support 223 00:12:53,080 --> 00:12:56,760 Speaker 1: his idea. His idea was that when a horse is 224 00:12:56,800 --> 00:13:00,800 Speaker 1: in full gallop, at some point during its gait, it 225 00:13:00,840 --> 00:13:05,120 Speaker 1: will have all four hoofs off the ground simultaneously, and 226 00:13:05,240 --> 00:13:10,480 Speaker 1: moy Bridge was essentially contracted to prove that Stanford had 227 00:13:10,520 --> 00:13:13,320 Speaker 1: a leg to stand on as it were. So Moybridge 228 00:13:13,360 --> 00:13:16,320 Speaker 1: invented away to use a series of twelve cameras to 229 00:13:16,400 --> 00:13:19,200 Speaker 1: take photos with an exposure lasting only a fraction of 230 00:13:19,200 --> 00:13:21,800 Speaker 1: a second, which at the time it was pretty remarkable. 231 00:13:21,880 --> 00:13:26,280 Speaker 1: Most photo photography. Early photography had very long exposure time 232 00:13:26,400 --> 00:13:29,320 Speaker 1: in order to get enough light to actually create the effect, 233 00:13:29,640 --> 00:13:32,280 Speaker 1: which is why you have lots of portraits of very 234 00:13:32,400 --> 00:13:35,720 Speaker 1: dour looking people in early photography, because it meant they 235 00:13:35,760 --> 00:13:39,319 Speaker 1: had to sit still for several minutes at a time, 236 00:13:39,440 --> 00:13:41,440 Speaker 1: and you know, if you're trying to hold a smile, 237 00:13:42,360 --> 00:13:44,320 Speaker 1: pretty soon you hate the world and everything in it, 238 00:13:44,400 --> 00:13:47,200 Speaker 1: so you typically have these very kind of neutral or 239 00:13:47,240 --> 00:13:51,120 Speaker 1: dour expressions in those early photographs. Well, Moybridge wanted to 240 00:13:51,120 --> 00:13:53,000 Speaker 1: find a way of doing this much faster, so he 241 00:13:53,080 --> 00:13:55,839 Speaker 1: created this camera system where in a fraction of a 242 00:13:55,880 --> 00:13:58,400 Speaker 1: second it could expose film to enough light to create 243 00:13:58,400 --> 00:14:01,319 Speaker 1: an image. He said, twelve of these cameras in a row, 244 00:14:01,880 --> 00:14:04,680 Speaker 1: and then he used little trip wires not to trip 245 00:14:04,760 --> 00:14:07,000 Speaker 1: up the horses, but rather the wires themselves would trip 246 00:14:07,040 --> 00:14:09,880 Speaker 1: when a horse ran through them, and it would cause 247 00:14:09,920 --> 00:14:13,520 Speaker 1: the cameras to take these photos in sequence, and then 248 00:14:13,559 --> 00:14:16,480 Speaker 1: you could look at them and see a horse's gallop 249 00:14:16,760 --> 00:14:20,480 Speaker 1: in progress. The photos showed that Stanford was in fact correct. 250 00:14:20,600 --> 00:14:22,680 Speaker 1: There was a brief moment in the horse's stride in 251 00:14:22,720 --> 00:14:25,760 Speaker 1: which all four legs were off the ground, and Moybridge's 252 00:14:25,800 --> 00:14:29,520 Speaker 1: innovation became valuable for scientists who wanted to study animal movement, 253 00:14:29,960 --> 00:14:33,240 Speaker 1: and later for engineers and roboticists who wished to mimic 254 00:14:33,480 --> 00:14:38,040 Speaker 1: those movements when they were developing robotic legs. Uh, Moybridge 255 00:14:38,080 --> 00:14:41,160 Speaker 1: and stop motion animation and stop motion photography, I should say, 256 00:14:41,200 --> 00:14:44,000 Speaker 1: are really fascinating subjects, and I'm sure I'll cover them 257 00:14:44,000 --> 00:14:47,160 Speaker 1: in another episode of tech Stuff at some point. Back 258 00:14:47,200 --> 00:14:50,240 Speaker 1: to milestones, Well, there are a lot of others that 259 00:14:50,280 --> 00:14:52,760 Speaker 1: the lab acknowledgeist as being really important moments in the 260 00:14:52,760 --> 00:14:55,880 Speaker 1: development of robotic legs, and they even include a few 261 00:14:55,920 --> 00:14:59,280 Speaker 1: of Dr Rabur's contributions to the fields, such as three. 262 00:14:59,280 --> 00:15:03,400 Speaker 1: When Rayber build a one legged hopping machine, it could 263 00:15:03,440 --> 00:15:06,000 Speaker 1: keep its own balance and travel at a specified rate. 264 00:15:06,040 --> 00:15:08,920 Speaker 1: It kind of looks like a UFO on a pogo stick. 265 00:15:09,000 --> 00:15:13,520 Speaker 1: It's constantly bouncing around. The main trick here was just 266 00:15:13,600 --> 00:15:17,160 Speaker 1: getting that balance right. So Rayber was able to create 267 00:15:17,240 --> 00:15:20,840 Speaker 1: this robot that uses a computer that calculates where that 268 00:15:21,040 --> 00:15:23,960 Speaker 1: one foot needs to come down in order to keep 269 00:15:24,000 --> 00:15:27,680 Speaker 1: the overall robot balanced while still traveling in the desired direction. 270 00:15:28,560 --> 00:15:31,120 Speaker 1: Those calculations had to be done fast enough so that 271 00:15:31,200 --> 00:15:33,760 Speaker 1: they were complete before the robot would come crashing down 272 00:15:33,840 --> 00:15:36,400 Speaker 1: after its last hops. So as it hops up in 273 00:15:36,400 --> 00:15:38,720 Speaker 1: the air, the computer has to figure out the trajectory, 274 00:15:38,840 --> 00:15:42,320 Speaker 1: the momentum, all of these different elements that the robot 275 00:15:42,400 --> 00:15:46,680 Speaker 1: is experiencing in order to compensate with the leg so 276 00:15:46,760 --> 00:15:48,600 Speaker 1: that it can hit the ground at just the right 277 00:15:48,640 --> 00:15:50,720 Speaker 1: spot with just the right amount of force to propel 278 00:15:50,760 --> 00:15:53,720 Speaker 1: the robot up again and continue moving it in whatever 279 00:15:53,760 --> 00:15:57,080 Speaker 1: direction you wanted to go in. So it's actually pretty complicated, 280 00:15:57,120 --> 00:15:59,720 Speaker 1: It took a lot of work and it's pretty fascinating 281 00:15:59,720 --> 00:16:03,560 Speaker 1: stuff off. That basic principle would then later be adapted 282 00:16:03,560 --> 00:16:07,160 Speaker 1: and applied into multi legged robots with a computer determining 283 00:16:07,200 --> 00:16:09,720 Speaker 1: where each leg needs to go, how it needs to 284 00:16:09,760 --> 00:16:13,120 Speaker 1: make contact with the ground, how much force, all just 285 00:16:13,160 --> 00:16:16,920 Speaker 1: so that the robot doesn't fall over. Rayber would end 286 00:16:16,960 --> 00:16:20,760 Speaker 1: up writing a book about his research. In the book 287 00:16:20,800 --> 00:16:25,520 Speaker 1: has the title legged robots that Balance. Pretty much sums 288 00:16:25,600 --> 00:16:28,240 Speaker 1: up the whole thing, doesn't it. It sounds like it's 289 00:16:28,240 --> 00:16:30,960 Speaker 1: a trivial thing. But again, it was really tricky to 290 00:16:31,040 --> 00:16:35,200 Speaker 1: do because while we're growing up, most of us, you know, 291 00:16:35,240 --> 00:16:38,680 Speaker 1: as kids, we learned to walk, and we kind of 292 00:16:39,000 --> 00:16:42,040 Speaker 1: have an kind of intuitive grasp of what we need 293 00:16:42,080 --> 00:16:45,240 Speaker 1: to do in order to keep our balance barring any 294 00:16:45,920 --> 00:16:50,400 Speaker 1: unforeseen circumstances like stepping on something that is on uh 295 00:16:50,840 --> 00:16:54,040 Speaker 1: it's like uncertain ground without even realizing it, or getting 296 00:16:54,080 --> 00:16:57,240 Speaker 1: shoved by someone. We're pretty good at keeping our balance 297 00:16:57,360 --> 00:17:01,600 Speaker 1: generally speaking, assuming all things are you know, normal, But 298 00:17:01,640 --> 00:17:06,360 Speaker 1: a robot has to think air quotes, think about how 299 00:17:06,400 --> 00:17:09,080 Speaker 1: to do this. Intuition isn't an option, at least not 300 00:17:09,160 --> 00:17:13,080 Speaker 1: without some really advanced machine learning algorithms, which honestly, we're 301 00:17:13,119 --> 00:17:14,960 Speaker 1: not that far along. In the nineteen eighties, this was 302 00:17:15,000 --> 00:17:17,879 Speaker 1: all stuff that had to be worked out by human engineers. 303 00:17:18,359 --> 00:17:21,880 Speaker 1: These days if you created a really advanced machine learning algorithm, 304 00:17:22,000 --> 00:17:25,080 Speaker 1: I suppose it won't be long before we start seeing 305 00:17:25,160 --> 00:17:27,720 Speaker 1: robots that teach themselves how they need to walk. And 306 00:17:27,800 --> 00:17:31,000 Speaker 1: we've seen some examples of that, and they look really weird, 307 00:17:32,119 --> 00:17:35,159 Speaker 1: but these are mostly simulations, like we've seen computer simulations 308 00:17:35,160 --> 00:17:37,760 Speaker 1: of what it would look like, where there's a virtual 309 00:17:37,880 --> 00:17:43,199 Speaker 1: robot that has virtual statistics including its virtual weight and 310 00:17:43,359 --> 00:17:45,359 Speaker 1: height and all these sort of things that have real 311 00:17:45,440 --> 00:17:49,200 Speaker 1: world impact if you were to create an actual robot, 312 00:17:49,560 --> 00:17:51,960 Speaker 1: And then there are simulations that show how a computer 313 00:17:52,040 --> 00:17:54,320 Speaker 1: works out how this robot needs to move in order 314 00:17:54,359 --> 00:17:57,640 Speaker 1: to actually walk without falling over, and in some cases 315 00:17:57,960 --> 00:18:02,280 Speaker 1: the animations are hysterical and also nightmare inducing. But as 316 00:18:02,320 --> 00:18:04,680 Speaker 1: far as I know, we don't have any actual physical 317 00:18:04,800 --> 00:18:07,359 Speaker 1: robots that follow that model. I could be wrong about that. 318 00:18:07,400 --> 00:18:10,160 Speaker 1: By the way, there's advances in robotics all the time, 319 00:18:10,280 --> 00:18:13,040 Speaker 1: and they're very well. Maybe a project out there that's 320 00:18:13,119 --> 00:18:15,840 Speaker 1: much further along than what I'm suggesting, but the ones 321 00:18:16,000 --> 00:18:20,159 Speaker 1: I have seen of all been virtual representations rabs. Robots. 322 00:18:20,160 --> 00:18:24,440 Speaker 1: By the way, we're mostly dynamic robots, meaning they were 323 00:18:24,520 --> 00:18:27,120 Speaker 1: dynamic in the sense they needed to move in order 324 00:18:27,200 --> 00:18:32,760 Speaker 1: to remain stable. You could have static stable robots, but 325 00:18:32,800 --> 00:18:34,960 Speaker 1: these were ones that needed to move in order to 326 00:18:35,200 --> 00:18:39,240 Speaker 1: maintain that upright position and not just fall over. And 327 00:18:39,280 --> 00:18:42,680 Speaker 1: even some of those early robots, which often were tethered 328 00:18:42,920 --> 00:18:47,159 Speaker 1: to control and power systems, could do really amazing things. 329 00:18:47,280 --> 00:18:51,280 Speaker 1: So most of the early robots, if you see the 330 00:18:51,320 --> 00:18:54,480 Speaker 1: footage from Boston Dynamics, they have all these different cables 331 00:18:54,480 --> 00:18:57,360 Speaker 1: and stuff coming off of them. Those cables are frequently 332 00:18:57,400 --> 00:19:00,880 Speaker 1: going to power systems and to controls the dumbs. Because 333 00:19:00,920 --> 00:19:04,840 Speaker 1: the robots were meant to demonstrate specific types of locomotion, 334 00:19:04,880 --> 00:19:07,520 Speaker 1: they weren't going to be used as a finished product. 335 00:19:08,040 --> 00:19:10,679 Speaker 1: It was more about studying the way to build the 336 00:19:10,760 --> 00:19:14,639 Speaker 1: best systems, and so you didn't have to worry about 337 00:19:14,680 --> 00:19:16,920 Speaker 1: making something that was all self contained. You were studying 338 00:19:16,920 --> 00:19:19,199 Speaker 1: all of this in the lab anyway. So a lot 339 00:19:19,240 --> 00:19:21,480 Speaker 1: of those videos you'll see just have the this like 340 00:19:21,680 --> 00:19:26,040 Speaker 1: nest of chords that are grouped up over the robot 341 00:19:26,119 --> 00:19:29,040 Speaker 1: and go into some or you know, system or other 342 00:19:29,240 --> 00:19:34,000 Speaker 1: multiple systems in some cases. But even in those early robots, 343 00:19:34,040 --> 00:19:38,440 Speaker 1: there's some video clips of phenomenal robotic gymnastics, Like there's 344 00:19:38,440 --> 00:19:40,879 Speaker 1: somewhere engineers are controlling the robots and making them do 345 00:19:40,960 --> 00:19:44,160 Speaker 1: things like forward flips while traveling a path. I'm talking 346 00:19:44,160 --> 00:19:46,639 Speaker 1: about like hopping robots. They'll be hopping in a circle 347 00:19:47,119 --> 00:19:50,520 Speaker 1: and then they'll do an extra tall hop and do 348 00:19:50,600 --> 00:19:53,800 Speaker 1: a full forward flip and land on their foot and 349 00:19:53,920 --> 00:19:57,120 Speaker 1: catch their balance and continue going in the circle. This 350 00:19:57,160 --> 00:19:59,920 Speaker 1: is decades before we saw the viral clips of youtu 351 00:20:00,160 --> 00:20:03,879 Speaker 1: both Big Dog and the other famous robots from Boston Dynamics. 352 00:20:03,880 --> 00:20:07,239 Speaker 1: So if you get a chance, do some digging on 353 00:20:07,280 --> 00:20:09,840 Speaker 1: YouTube for Boston Dynamics and take a look at some 354 00:20:09,880 --> 00:20:13,440 Speaker 1: of those early videos, because they're really really impressive and 355 00:20:13,560 --> 00:20:18,439 Speaker 1: in some cases extremely creepy. Another Raybors contributions on the 356 00:20:18,480 --> 00:20:22,480 Speaker 1: list dates from when he and his team developed quadruped 357 00:20:22,600 --> 00:20:26,480 Speaker 1: robots that can move at different speeds like trotting, pacing, 358 00:20:26,560 --> 00:20:29,639 Speaker 1: and bounding, and they could switch between those different modes 359 00:20:29,720 --> 00:20:31,240 Speaker 1: so you didn't just have to turn it on and 360 00:20:31,240 --> 00:20:33,640 Speaker 1: then put it down on the ground. You could actually 361 00:20:33,800 --> 00:20:36,879 Speaker 1: go from one speed to the other and you know, 362 00:20:36,960 --> 00:20:39,520 Speaker 1: knock up a notch or down a notch. If you 363 00:20:39,560 --> 00:20:42,479 Speaker 1: watch slow motion footage of animals moving at different speeds, 364 00:20:42,480 --> 00:20:45,720 Speaker 1: you'll see that their legs move differently at each rate 365 00:20:45,720 --> 00:20:48,440 Speaker 1: of travel. So like a horse at a trot and 366 00:20:48,440 --> 00:20:50,679 Speaker 1: a horse at a gallop, their legs are behaving in 367 00:20:50,720 --> 00:20:53,200 Speaker 1: different ways. Well, Raybor was working very hard to create 368 00:20:53,280 --> 00:20:56,120 Speaker 1: the same sort of thing in his robots. So Rayber 369 00:20:56,200 --> 00:20:58,800 Speaker 1: and his team worked to push the envelope in the 370 00:20:58,800 --> 00:21:02,440 Speaker 1: field of robotic low emotion. They experimented with different designs 371 00:21:02,520 --> 00:21:05,440 Speaker 1: create more stable, capable robots, and many of those early 372 00:21:05,480 --> 00:21:09,040 Speaker 1: designs used humans to guide the path of the robots, 373 00:21:09,480 --> 00:21:13,440 Speaker 1: so it was kind of a partly human controlled, partly 374 00:21:13,560 --> 00:21:19,040 Speaker 1: robotic controlled system. Frequently, the robots themselves had computers, either 375 00:21:19,160 --> 00:21:21,600 Speaker 1: on board or just off board that did all the 376 00:21:21,640 --> 00:21:24,800 Speaker 1: actual work to determine the particulars like where does the 377 00:21:24,880 --> 00:21:28,080 Speaker 1: leg go and how would it need to shift its 378 00:21:28,119 --> 00:21:30,560 Speaker 1: weight in order to obey the commands. So the human 379 00:21:30,640 --> 00:21:32,960 Speaker 1: is essentially giving commands like I want you to move 380 00:21:33,040 --> 00:21:36,960 Speaker 1: over forward five feet, but it was the robots computer 381 00:21:37,080 --> 00:21:38,960 Speaker 1: that was actually doing the work of how does it 382 00:21:39,119 --> 00:21:43,399 Speaker 1: achieve that goal. In n Rayber founded a company to 383 00:21:43,440 --> 00:21:47,119 Speaker 1: continue this work and find commercial applications for the technologies 384 00:21:47,119 --> 00:21:51,639 Speaker 1: he was developing, and that company was Boston Dynamics. So 385 00:21:51,680 --> 00:21:54,800 Speaker 1: we're finally at the point where Boston Dynamics is its 386 00:21:54,840 --> 00:21:58,200 Speaker 1: own company, which means it's time to take a quick 387 00:21:58,200 --> 00:22:07,600 Speaker 1: break and thank our sponsor. All right, So it's the 388 00:22:07,600 --> 00:22:10,679 Speaker 1: early nine nineties and bust and Dynamics as a new company. 389 00:22:10,760 --> 00:22:14,760 Speaker 1: So do they start turning out commercial robots right away? Nope. 390 00:22:14,960 --> 00:22:16,600 Speaker 1: As a matter of fact, as of the recording of 391 00:22:16,640 --> 00:22:19,400 Speaker 1: this podcast, none of the company's work has ever been 392 00:22:19,400 --> 00:22:23,280 Speaker 1: turned into a commercial product. Directly, you cannot go out 393 00:22:23,320 --> 00:22:26,399 Speaker 1: and buy a Boston Dynamics robot for yourself. But the 394 00:22:26,400 --> 00:22:29,159 Speaker 1: company's work has been extremely important to push forward the 395 00:22:29,200 --> 00:22:32,560 Speaker 1: evolution of robotics, and there are some really big organizations 396 00:22:32,560 --> 00:22:35,760 Speaker 1: that have been extremely interested in that, such as the 397 00:22:35,800 --> 00:22:39,720 Speaker 1: Department of Defense. Now, much of Buston Dynamics funding has 398 00:22:39,760 --> 00:22:43,000 Speaker 1: come from really big contracts awarded to the company by 399 00:22:43,040 --> 00:22:47,760 Speaker 1: the Defense Advanced Research Projects Agency, also known as DARPA. 400 00:22:47,920 --> 00:22:50,160 Speaker 1: And you might remember DARPA as being the group responsible 401 00:22:50,200 --> 00:22:53,040 Speaker 1: for the Grand Challenge, in which self driving cars raced 402 00:22:53,080 --> 00:22:56,159 Speaker 1: through courses ranging from desert landscapes to a mock up 403 00:22:56,200 --> 00:22:58,400 Speaker 1: of a working city to see who could come out 404 00:22:58,440 --> 00:23:00,760 Speaker 1: on top. Or maybe you remember brit DARPA as the 405 00:23:00,800 --> 00:23:03,760 Speaker 1: agency which at the time was known as ARPA that 406 00:23:03,920 --> 00:23:06,879 Speaker 1: funded the development of technology that became the backbone of 407 00:23:06,920 --> 00:23:09,640 Speaker 1: the Internet. We wouldn't have the Internet as it stands 408 00:23:09,640 --> 00:23:11,760 Speaker 1: today without the R and D arm of the Department 409 00:23:11,760 --> 00:23:15,280 Speaker 1: of Defense. As it turns out, DARPA has been very 410 00:23:15,359 --> 00:23:18,919 Speaker 1: much interested in robotics, and the US Army, Navy, and 411 00:23:19,000 --> 00:23:22,639 Speaker 1: Marine Corps have similarly been interested in potential applications of 412 00:23:22,760 --> 00:23:26,199 Speaker 1: robots that are either inspired by man or beasts. So 413 00:23:26,280 --> 00:23:30,439 Speaker 1: Boston Dynamics has frequently operated as a research facility on 414 00:23:30,560 --> 00:23:35,119 Speaker 1: behalf of these organizations. Now. One of the earliest projects 415 00:23:35,119 --> 00:23:37,399 Speaker 1: that we ever got to hear about from Boston Dynamics 416 00:23:37,520 --> 00:23:42,000 Speaker 1: was Big Dog, dynamically stable four legged robot that debuted 417 00:23:42,040 --> 00:23:45,000 Speaker 1: in two thousand five. It was a predecessor of a 418 00:23:45,080 --> 00:23:49,760 Speaker 1: larger project titled Legged Squad Support System or LS three. 419 00:23:50,520 --> 00:23:54,480 Speaker 1: Boston Dynamics partnered with the Jet Propulsion Laboratory, the Concord 420 00:23:54,840 --> 00:23:58,080 Speaker 1: Field Station at Harvard University, and a robotics company called 421 00:23:58,119 --> 00:24:02,719 Speaker 1: Foster Miller on this The concept behind the design was 422 00:24:02,800 --> 00:24:05,240 Speaker 1: a robot that would be able to carry field equipment 423 00:24:05,280 --> 00:24:08,679 Speaker 1: in support of deployed military squads. So rather than load 424 00:24:08,720 --> 00:24:12,600 Speaker 1: down soldiers with dozens of pounds of gear, the robot 425 00:24:12,640 --> 00:24:14,600 Speaker 1: would be able to carry those loads and free up 426 00:24:14,640 --> 00:24:17,000 Speaker 1: the humans so they didn't have to tire so quickly 427 00:24:17,040 --> 00:24:21,120 Speaker 1: and they can move around with fewer restrictions. Dr Martin Bueller, 428 00:24:21,320 --> 00:24:25,080 Speaker 1: formerly of McGill University, led the Big Dog project. Dr 429 00:24:25,119 --> 00:24:28,040 Speaker 1: Bueller would be the Director of robotics Research at Boston 430 00:24:28,119 --> 00:24:30,960 Speaker 1: Dynamics from two thousand three to two thousand eight, but 431 00:24:31,000 --> 00:24:33,840 Speaker 1: would eventually move on to work for another corporation, one 432 00:24:33,880 --> 00:24:36,560 Speaker 1: called I Robot in two thousand and eight, and then 433 00:24:36,640 --> 00:24:39,520 Speaker 1: later on would move on to other projects. Big Dog 434 00:24:39,680 --> 00:24:44,479 Speaker 1: could carry forty or more than ninety pounds. The robot 435 00:24:44,520 --> 00:24:47,560 Speaker 1: itself weighed in at a hundred nine ms. It was 436 00:24:47,600 --> 00:24:52,040 Speaker 1: a meter tall, and it wasn't some battery operated robot. Instead, 437 00:24:52,119 --> 00:24:56,080 Speaker 1: it actually used a gasoline fueled engine. The joints used 438 00:24:56,160 --> 00:25:00,000 Speaker 1: hydraulic actuators, meaning the robot would use liquids under press 439 00:25:00,040 --> 00:25:03,080 Speaker 1: sure to extend limbs or reduce pressure to allow limbs 440 00:25:03,119 --> 00:25:07,720 Speaker 1: to contract to bend backward. For sensing its environment, Big 441 00:25:07,760 --> 00:25:11,000 Speaker 1: Dog also had stereo cameras so it could perceive depth, 442 00:25:11,400 --> 00:25:13,240 Speaker 1: and a light oar system to get an idea of 443 00:25:13,240 --> 00:25:16,560 Speaker 1: the orientation of various elements in an environment. Now, a 444 00:25:16,640 --> 00:25:19,480 Speaker 1: light OAR stands where light detection and ranging. Here's how 445 00:25:19,480 --> 00:25:22,720 Speaker 1: it works in a nutshell, a light ar detector has 446 00:25:22,800 --> 00:25:25,800 Speaker 1: two main components. First is an emitter or a light 447 00:25:25,800 --> 00:25:28,720 Speaker 1: gun of some sort, which shoots out light, typically in 448 00:25:28,760 --> 00:25:31,520 Speaker 1: the infrared range. The light shoots out from the emitter 449 00:25:31,640 --> 00:25:34,439 Speaker 1: until it hits some sort of object, and then some 450 00:25:34,560 --> 00:25:37,639 Speaker 1: of that light bounces back, and that's when light ars. 451 00:25:37,640 --> 00:25:41,000 Speaker 1: Second big component, a detector picks up the returning signals. 452 00:25:41,840 --> 00:25:44,760 Speaker 1: By doing some quick math, the lightar system can determine 453 00:25:44,800 --> 00:25:47,000 Speaker 1: how far away the object was based on the delay 454 00:25:47,080 --> 00:25:51,000 Speaker 1: between the emitted infrared light and then the detection of 455 00:25:51,000 --> 00:25:53,359 Speaker 1: the infrared light when it comes back. This is the 456 00:25:53,440 --> 00:25:56,800 Speaker 1: system used in modern speed guns. Oh and in a 457 00:25:56,840 --> 00:25:59,040 Speaker 1: speed gun, the emitter typically shoots out a burst of 458 00:25:59,080 --> 00:26:01,359 Speaker 1: these infrared way wives. So when the emitter picks up 459 00:26:01,400 --> 00:26:03,800 Speaker 1: the returning waves, it can look at the difference in 460 00:26:03,880 --> 00:26:07,720 Speaker 1: distances from the emitter and thus and for the speed 461 00:26:07,800 --> 00:26:10,000 Speaker 1: of the object that the gun was pointed at, such 462 00:26:10,000 --> 00:26:12,760 Speaker 1: as a super sweet red sports car, because we all 463 00:26:12,800 --> 00:26:15,959 Speaker 1: know the red ones go faster. This is also the 464 00:26:16,080 --> 00:26:20,199 Speaker 1: robot that could keep kicking and keep on ticking. I 465 00:26:20,240 --> 00:26:24,040 Speaker 1: guess anyway, there are many videos showing people giving big 466 00:26:24,080 --> 00:26:27,520 Speaker 1: Dog a let's call it a healthy shove with a 467 00:26:27,560 --> 00:26:30,760 Speaker 1: foot to throw the robot off balance. You've probably seen 468 00:26:30,800 --> 00:26:33,160 Speaker 1: one of these videos at least where there's this four 469 00:26:33,240 --> 00:26:37,399 Speaker 1: legged robot and some dude comes up, puts his foot 470 00:26:37,560 --> 00:26:39,200 Speaker 1: right up against the robot and gives us a big 471 00:26:39,200 --> 00:26:41,280 Speaker 1: old shove, and the robot just stumbles around for a 472 00:26:41,280 --> 00:26:43,720 Speaker 1: bit before picking itself back up. Now, the purpose of 473 00:26:43,720 --> 00:26:46,600 Speaker 1: those videos isn't a show that robots, hey, we're still 474 00:26:46,640 --> 00:26:49,480 Speaker 1: the boss of you, but rather to demonstrate the ability 475 00:26:49,480 --> 00:26:52,359 Speaker 1: of the robots to catch themselves even after they've been shoved, 476 00:26:52,359 --> 00:26:56,200 Speaker 1: and to regain their balance through careful, though sometimes seemingly 477 00:26:56,280 --> 00:27:00,679 Speaker 1: haphazard placement of their legs. Now doesn't work every time, no, 478 00:27:01,200 --> 00:27:04,560 Speaker 1: because terrain can be really unpredictable. Sometimes the robots best 479 00:27:04,560 --> 00:27:06,960 Speaker 1: efforts just aren't enough to keep it from toppling over. 480 00:27:07,400 --> 00:27:10,040 Speaker 1: But there are plenty of video examples of those robots, 481 00:27:10,320 --> 00:27:14,320 Speaker 1: which look fairly lifelike in their locomotion getting kicked before 482 00:27:14,320 --> 00:27:17,280 Speaker 1: regaining their footing. And it shouldn't come as any surprise 483 00:27:17,320 --> 00:27:20,120 Speaker 1: that the robots movements look lifelike since the engineers were 484 00:27:20,119 --> 00:27:23,919 Speaker 1: frequently relying upon animal movement as a model for robotic movement. 485 00:27:24,800 --> 00:27:27,280 Speaker 1: As for the l S three project, it added some 486 00:27:27,320 --> 00:27:30,600 Speaker 1: more requirements. This is the bigger version of Big Dog. 487 00:27:31,280 --> 00:27:34,200 Speaker 1: Sometimes they would call this Alpha Dog, and it could 488 00:27:34,240 --> 00:27:37,320 Speaker 1: carry up to a hundred two of gear for a 489 00:27:37,359 --> 00:27:40,880 Speaker 1: distance of thirty two kilometers or duration of twenty four hours. 490 00:27:41,359 --> 00:27:44,240 Speaker 1: According to Boston Dynamics, during one control test, it was 491 00:27:44,280 --> 00:27:49,160 Speaker 1: actually able to carry five hundreds of mass it's more 492 00:27:49,160 --> 00:27:53,280 Speaker 1: than a thousand pounds. It also could recognize a designated leader, 493 00:27:53,440 --> 00:27:56,359 Speaker 1: so an actual human being, and then follow behind that 494 00:27:56,480 --> 00:28:00,560 Speaker 1: specific person that leader without needing any other operators. On 495 00:28:00,680 --> 00:28:03,119 Speaker 1: top of the sensors that it had from its a 496 00:28:03,160 --> 00:28:07,199 Speaker 1: smaller cousin Big Dog, it also had GPS, so it 497 00:28:07,280 --> 00:28:11,040 Speaker 1: had some additional sensors inside of it. And ultimately, though 498 00:28:11,520 --> 00:28:14,000 Speaker 1: it wouldn't go very far beyond a few test missions 499 00:28:14,000 --> 00:28:18,679 Speaker 1: with the military. So the advances in robotics locomotion were considerable, 500 00:28:19,320 --> 00:28:21,959 Speaker 1: and there are many videos showing how impressive this robot 501 00:28:22,160 --> 00:28:24,520 Speaker 1: is and how it maneuvers across different types of terrain. 502 00:28:24,640 --> 00:28:27,480 Speaker 1: But it had a big, big problem, and that was 503 00:28:27,520 --> 00:28:31,439 Speaker 1: it was noisy. The robots gasoline engine created a huge 504 00:28:31,440 --> 00:28:34,119 Speaker 1: amount of noise, which meant that if you were a 505 00:28:34,160 --> 00:28:37,000 Speaker 1: military squad and you had one of these in your group, 506 00:28:38,200 --> 00:28:40,920 Speaker 1: your position would immediately be given away. There was nothing 507 00:28:40,960 --> 00:28:45,040 Speaker 1: stealthy about it. Boston Dynamics would create a smaller, battery 508 00:28:45,080 --> 00:28:49,400 Speaker 1: operated version of that robot called Spot, But spots problem 509 00:28:49,840 --> 00:28:53,600 Speaker 1: was not noise, it was just carrying capacity. It wasn't 510 00:28:53,640 --> 00:28:57,720 Speaker 1: as hefty, It wasn't as strong as Big Dog or 511 00:28:57,720 --> 00:29:00,240 Speaker 1: Alpha Dog. It could only carry forty pounds or about 512 00:29:00,240 --> 00:29:04,440 Speaker 1: eighteen kims, which wasn't really considered to be enough to 513 00:29:04,520 --> 00:29:07,600 Speaker 1: be useful to the military, so it was much more 514 00:29:07,680 --> 00:29:11,040 Speaker 1: quiet but had less utility. Both Big Dog and Spot 515 00:29:11,320 --> 00:29:14,560 Speaker 1: would come to an end. Their projects would end, sometimes 516 00:29:14,560 --> 00:29:17,560 Speaker 1: around two thousand and fifteen, but the progress made during 517 00:29:17,560 --> 00:29:19,720 Speaker 1: the design and build out of the robots would inform 518 00:29:19,800 --> 00:29:22,400 Speaker 1: future efforts at Boston Dynamics, So it wasn't like it 519 00:29:22,440 --> 00:29:26,080 Speaker 1: was a total loss or anything. The value was really 520 00:29:26,240 --> 00:29:30,280 Speaker 1: in the actual design and build out of these robots, 521 00:29:30,640 --> 00:29:33,840 Speaker 1: and that becomes a story, kind of exploratory research, kind 522 00:29:33,840 --> 00:29:38,160 Speaker 1: of story for Boston Dynamics throughout its entire existence. Another 523 00:29:38,280 --> 00:29:40,800 Speaker 1: robot that would come out of this effort would be 524 00:29:40,880 --> 00:29:44,560 Speaker 1: spot Many. This is a much more recent robot. It's 525 00:29:44,560 --> 00:29:48,400 Speaker 1: a smaller variation on that four legged design spot Many 526 00:29:48,440 --> 00:29:52,480 Speaker 1: debuted in two thousand seventeen, and it has a much 527 00:29:52,520 --> 00:29:54,840 Speaker 1: more friendly looking appearance. A lot of the you know, 528 00:29:54,880 --> 00:30:00,120 Speaker 1: Big Dog and Spot were kind of industrial and in 529 00:30:00,560 --> 00:30:02,920 Speaker 1: looks like they weren't they weren't meant to be pretty. 530 00:30:03,120 --> 00:30:05,480 Speaker 1: Spot Many has a little bit of a friendlier look 531 00:30:05,520 --> 00:30:08,640 Speaker 1: with some three D printed parts, and it makes a 532 00:30:08,640 --> 00:30:11,280 Speaker 1: little more sleek and it also has a nice bright 533 00:30:11,360 --> 00:30:15,920 Speaker 1: yellow color to the ones that are on the videos anyway, um, 534 00:30:15,960 --> 00:30:19,400 Speaker 1: and it's also smaller. Buston Dynamics called it their quietest 535 00:30:19,480 --> 00:30:22,000 Speaker 1: robot to date, and of course it operates on battery 536 00:30:22,040 --> 00:30:25,880 Speaker 1: power because hydraulics would be very loud and also unnecessarily 537 00:30:25,920 --> 00:30:28,800 Speaker 1: strong for such a small form factor. The robot moves 538 00:30:28,800 --> 00:30:31,760 Speaker 1: in a very fluid, organic way, and certain motions kind 539 00:30:31,760 --> 00:30:34,280 Speaker 1: of give off the illusion that it is actually alive, 540 00:30:34,720 --> 00:30:37,920 Speaker 1: but we have been assured by Boston Dynamics that as 541 00:30:37,960 --> 00:30:41,360 Speaker 1: not the case, so I'm gonna trust them. The robot 542 00:30:41,400 --> 00:30:45,160 Speaker 1: weighs as the base model, but you can also get 543 00:30:45,240 --> 00:30:48,520 Speaker 1: one while you can't. But the Boston Dynamics also makes 544 00:30:48,560 --> 00:30:53,040 Speaker 1: one that has an articulated arm that is mounted to 545 00:30:53,080 --> 00:30:55,240 Speaker 1: the top of the robot. So think of the robot 546 00:30:55,280 --> 00:30:57,880 Speaker 1: is like a platform. You've got four legs attached to 547 00:30:57,920 --> 00:31:00,800 Speaker 1: it on the top front part of it. You've also 548 00:31:00,800 --> 00:31:04,320 Speaker 1: got an articulated arm that can reach out and grab 549 00:31:04,440 --> 00:31:07,880 Speaker 1: stuff is on a joint, so it's like it's got 550 00:31:07,880 --> 00:31:11,080 Speaker 1: an elbow. And this adds an extra five kilograms to 551 00:31:11,120 --> 00:31:14,560 Speaker 1: the weight of the robot. And if you just take 552 00:31:14,600 --> 00:31:17,920 Speaker 1: a look at the design, the hand and arm makes 553 00:31:17,960 --> 00:31:19,840 Speaker 1: it look like it's kind of a head and neck 554 00:31:20,080 --> 00:31:22,080 Speaker 1: of an animal, so kind of like the spot Mini 555 00:31:22,200 --> 00:31:26,560 Speaker 1: is actually a tiny frisky a pedasaurus or something. According 556 00:31:26,560 --> 00:31:29,000 Speaker 1: to the company's spot Mini can prance around for about 557 00:31:29,120 --> 00:31:31,960 Speaker 1: ninety minutes before it needs a recharge, and it can 558 00:31:32,000 --> 00:31:36,160 Speaker 1: carry about fourteen kilograms of payload and has seventeen joints 559 00:31:36,160 --> 00:31:39,240 Speaker 1: of articulation. If you haven't seen videos of that thing 560 00:31:39,240 --> 00:31:41,040 Speaker 1: in motion, I highly recommend you check it out. They 561 00:31:41,040 --> 00:31:44,920 Speaker 1: are fascinating and again a little unnerving because you know, 562 00:31:45,120 --> 00:31:48,800 Speaker 1: they again are mocking organic creatures. So there's this sort 563 00:31:48,840 --> 00:31:54,200 Speaker 1: of kind of Uncanny Valley thing going on with these 564 00:31:54,320 --> 00:31:57,280 Speaker 1: robots that move like living things, but clearly are not 565 00:31:57,400 --> 00:32:00,560 Speaker 1: living things. There's also Little Dog, which is a much 566 00:32:00,600 --> 00:32:04,320 Speaker 1: smaller robot that Boston Dynamics created for DARPA. DARPA, in turn, 567 00:32:04,400 --> 00:32:07,840 Speaker 1: relies on Little Dog as a development platform for other 568 00:32:07,880 --> 00:32:10,360 Speaker 1: advances in the robotics. So if you've listened to my 569 00:32:10,400 --> 00:32:13,240 Speaker 1: other episodes that involved DARPA, you know that DARPA itself 570 00:32:13,280 --> 00:32:17,480 Speaker 1: does not conduct original research. Rather, it's an agency that 571 00:32:17,520 --> 00:32:20,560 Speaker 1: offers up funding to other companies and organizations to do 572 00:32:20,600 --> 00:32:23,560 Speaker 1: that kind of work, and that's in the interest of 573 00:32:23,560 --> 00:32:26,720 Speaker 1: the Department of Defense. Little Dogs Men is kind of 574 00:32:26,720 --> 00:32:30,320 Speaker 1: a standard platform upon which robotics organizations can develop new 575 00:32:30,360 --> 00:32:34,760 Speaker 1: technology for DARPA projects. So that might involve things like pathfinding, 576 00:32:35,080 --> 00:32:38,760 Speaker 1: object recognition, uh, you know that sort of stuff, things 577 00:32:38,840 --> 00:32:40,960 Speaker 1: that are maybe more on the software side of things 578 00:32:40,960 --> 00:32:44,280 Speaker 1: than the hardware side of things. Other four legged robots 579 00:32:44,280 --> 00:32:47,320 Speaker 1: built by Boston Dynamics include the Cheetah, which could hit 580 00:32:47,360 --> 00:32:49,480 Speaker 1: a top speed of twenty eight miles per hour that's 581 00:32:49,480 --> 00:32:52,600 Speaker 1: about forty five kilometers per hour. The Cheetah was not 582 00:32:52,800 --> 00:32:56,320 Speaker 1: built to run free across the landscape. It was running 583 00:32:56,360 --> 00:33:00,240 Speaker 1: on a treadmill. It actually had tethered power supply. Eye 584 00:33:00,240 --> 00:33:04,960 Speaker 1: tethered controls, so again, this wasn't a a all inclusive robot. 585 00:33:05,360 --> 00:33:07,880 Speaker 1: The purpose of the Cheetah was for robotics experts to 586 00:33:08,000 --> 00:33:11,720 Speaker 1: experiment with ways to build a fast moving robot. So again, 587 00:33:11,800 --> 00:33:14,080 Speaker 1: this was for them to kind of build out the 588 00:33:14,120 --> 00:33:18,600 Speaker 1: technologies that would then be incorporated into future designs, rather 589 00:33:18,640 --> 00:33:21,320 Speaker 1: than something that was its own self contained robot. So 590 00:33:21,400 --> 00:33:25,120 Speaker 1: again it kind of a development platform. Uh. It was 591 00:33:25,320 --> 00:33:28,080 Speaker 1: really fascinating to to watch though. It would run on 592 00:33:28,120 --> 00:33:31,520 Speaker 1: these treadmills at blistering speeds. You'd see this treadmill start 593 00:33:31,560 --> 00:33:34,320 Speaker 1: to ramp up, and at twenty eight miles per hour, 594 00:33:34,440 --> 00:33:37,959 Speaker 1: it meant that it was moving faster even than Hussain Bolt. However, 595 00:33:38,320 --> 00:33:43,360 Speaker 1: it couldn't balance itself, so unlike Hussain Bolt, you would 596 00:33:43,360 --> 00:33:46,280 Speaker 1: not be able to Like, it couldn't take a turn 597 00:33:46,360 --> 00:33:48,760 Speaker 1: to save its life. So if somehow one of these 598 00:33:48,880 --> 00:33:51,800 Speaker 1: Cheetah robots were to get after you, all you need 599 00:33:51,800 --> 00:33:55,760 Speaker 1: to do is take a left or right you'd be fine. Uh. Also, 600 00:33:55,840 --> 00:33:57,880 Speaker 1: it be tethered to a computer and probably couldn't move 601 00:33:57,960 --> 00:34:02,040 Speaker 1: very far, so you're probably fine either way. But that's 602 00:34:02,080 --> 00:34:05,760 Speaker 1: just one They did, however, use that design to build 603 00:34:05,760 --> 00:34:09,400 Speaker 1: out another four legged fast robot called the Wildcat. This 604 00:34:09,440 --> 00:34:13,759 Speaker 1: one was a self contained robot. Buston Dynamics showed off 605 00:34:13,760 --> 00:34:17,480 Speaker 1: the Wildcat robot in two thousand thirteen with videos of 606 00:34:17,520 --> 00:34:22,160 Speaker 1: this four legged robot bounding and galloping outside. It was 607 00:34:22,520 --> 00:34:24,560 Speaker 1: not tethered so it can run around and had a 608 00:34:24,600 --> 00:34:28,759 Speaker 1: gasoline engine very similar to Big Dog. Pretty loud, but 609 00:34:28,880 --> 00:34:31,200 Speaker 1: also pretty fast. It could get a top speed of 610 00:34:31,239 --> 00:34:34,200 Speaker 1: about sixteen miles per hour, so not quite as fast 611 00:34:34,200 --> 00:34:36,440 Speaker 1: as the Cheetah. If you are Hussain Bolt, you could 612 00:34:36,480 --> 00:34:41,480 Speaker 1: outrun this particular robot. If you are me, you would 613 00:34:41,560 --> 00:34:45,919 Speaker 1: be robot meat. I don't go faster than sixteen miles 614 00:34:45,920 --> 00:34:50,440 Speaker 1: per unless I'm either inside another vehicle or falling off 615 00:34:50,440 --> 00:34:53,879 Speaker 1: a cliff. There is another Cheetah robot by the way, 616 00:34:53,920 --> 00:34:57,800 Speaker 1: if you do searches for Cheetah robot, there's a different 617 00:34:57,800 --> 00:35:00,440 Speaker 1: one that comes out of the m I T Robotics apartment. 618 00:35:01,160 --> 00:35:04,040 Speaker 1: This one's a really cool robot. It's a four legged 619 00:35:04,120 --> 00:35:07,239 Speaker 1: robot that can detect and jump over obstacles. There's some 620 00:35:07,320 --> 00:35:12,120 Speaker 1: great videos of engineers showing it uh seeing an oncoming 621 00:35:12,120 --> 00:35:15,200 Speaker 1: obstacle on a treadmill and leaping over it. Even as 622 00:35:15,200 --> 00:35:17,600 Speaker 1: the obstacles grow taller and taller, it's still able to 623 00:35:17,640 --> 00:35:21,200 Speaker 1: get over them. It's an incredibly impressive robot, but it 624 00:35:21,320 --> 00:35:26,640 Speaker 1: is not a Boston Dynamics robot. Then there's Rise, which 625 00:35:26,680 --> 00:35:29,800 Speaker 1: is a wall climbing robot. This one has six legs 626 00:35:29,880 --> 00:35:32,560 Speaker 1: and it's pretty small. It's about nine point six inches 627 00:35:32,680 --> 00:35:36,640 Speaker 1: or long, has a tail like protrusion that makes it 628 00:35:36,680 --> 00:35:39,919 Speaker 1: look sort of like a robotic horseshoe crab, and each 629 00:35:40,000 --> 00:35:42,279 Speaker 1: leg has two joints, which means it has a total 630 00:35:42,320 --> 00:35:45,880 Speaker 1: of twelve actuated degrees of freedom. The feet have a 631 00:35:45,920 --> 00:35:49,759 Speaker 1: little micro claws on it, so it can grip to surfaces. 632 00:35:49,760 --> 00:35:52,759 Speaker 1: As long as the surface isn't completely smooth, it can 633 00:35:52,800 --> 00:35:55,480 Speaker 1: grip onto it and allow this little robot to climb 634 00:35:55,560 --> 00:35:58,720 Speaker 1: right up a vertical surface. On top of the normal 635 00:35:58,760 --> 00:36:02,120 Speaker 1: sensors that you would expect in any given robot, there 636 00:36:02,160 --> 00:36:04,440 Speaker 1: were some additional ones to ensure that it wouldn't take 637 00:36:04,480 --> 00:36:09,000 Speaker 1: an unplanned tumble, which included foot contact sensors so the 638 00:36:09,080 --> 00:36:11,759 Speaker 1: robot would actually know when its claws had been more 639 00:36:11,800 --> 00:36:15,200 Speaker 1: or less engaged in whatever surface it was climbing. They 640 00:36:15,239 --> 00:36:19,480 Speaker 1: also had legs strain sensors so if any legs were 641 00:36:20,760 --> 00:36:23,799 Speaker 1: having more strain on them than normal, it could either 642 00:36:23,920 --> 00:36:27,160 Speaker 1: redistribute its weight in an attempt to continue its climb, 643 00:36:27,320 --> 00:36:30,800 Speaker 1: or it could stop its climb rather than risking falling 644 00:36:30,840 --> 00:36:33,200 Speaker 1: off of its surface and allow someone to come up 645 00:36:33,200 --> 00:36:38,280 Speaker 1: and retrieve it. According to Boston Dynamics UH, this robot 646 00:36:38,320 --> 00:36:43,120 Speaker 1: could actually climb faster than an average human, and unlike humans, 647 00:36:43,120 --> 00:36:45,400 Speaker 1: it doesn't tire out, although it could run out a 648 00:36:45,400 --> 00:36:48,400 Speaker 1: battery charge, so that's sort of similar. And this was 649 00:36:48,440 --> 00:36:51,920 Speaker 1: not built for a specific purpose. It was more like 650 00:36:52,640 --> 00:36:57,640 Speaker 1: building out capabilities of robots with the ultimate goal of 651 00:36:57,680 --> 00:37:00,239 Speaker 1: creating robots that could be useful for people like first 652 00:37:00,280 --> 00:37:06,760 Speaker 1: responders in various UH emergencies. So there wasn't a specific mission, 653 00:37:07,120 --> 00:37:09,600 Speaker 1: but it was more about let's build out these capabilities 654 00:37:09,640 --> 00:37:12,560 Speaker 1: that could come in handy in future robotic designs. I 655 00:37:12,640 --> 00:37:15,440 Speaker 1: got a lot more to say about Boston Dynamics, but 656 00:37:15,520 --> 00:37:18,040 Speaker 1: before I get into this final segment, let's take another 657 00:37:18,120 --> 00:37:27,640 Speaker 1: quick break to thank our sponsor. Okay, so let's talk 658 00:37:27,680 --> 00:37:30,759 Speaker 1: about a couple more robots before we get into some 659 00:37:30,960 --> 00:37:35,719 Speaker 1: interesting recent stories involving Boston Dynamics. There was also a 660 00:37:35,840 --> 00:37:39,920 Speaker 1: robot called Rex and that's spelled r h e X. 661 00:37:40,000 --> 00:37:42,600 Speaker 1: That's a six legged robot designed to be able to 662 00:37:42,640 --> 00:37:46,680 Speaker 1: tackle particularly tough terrain. It's not a big robot. It 663 00:37:46,960 --> 00:37:50,400 Speaker 1: is about fourteen centimeters tall, so it doesn't exactly tower 664 00:37:50,520 --> 00:37:53,799 Speaker 1: above the landscape, but it's curved legs could propel it 665 00:37:53,840 --> 00:37:56,840 Speaker 1: over lots of different surfaces better than robots much larger 666 00:37:56,880 --> 00:38:00,200 Speaker 1: than itself. It relied on TELA operations, which means you know, 667 00:38:00,280 --> 00:38:03,480 Speaker 1: someone's controlling it remotely using a camera to get a 668 00:38:03,560 --> 00:38:05,720 Speaker 1: robot's eye view of what was in front of the critter. 669 00:38:06,200 --> 00:38:09,279 Speaker 1: The legs rotate kind of like wheels, so they're these 670 00:38:09,280 --> 00:38:12,320 Speaker 1: sort of semi circular legs. They don't make a full circle, 671 00:38:12,640 --> 00:38:15,839 Speaker 1: but they do rotate around and they can go through 672 00:38:15,920 --> 00:38:18,760 Speaker 1: pretty much anything like that shows robots going through mud 673 00:38:18,760 --> 00:38:22,200 Speaker 1: and water, and because it's a fully sealed robot, it 674 00:38:22,239 --> 00:38:24,680 Speaker 1: doesn't you know, it's not in danger of having water, 675 00:38:25,239 --> 00:38:28,480 Speaker 1: give it any short circuits or anything like that, no disassemble. 676 00:38:28,920 --> 00:38:33,760 Speaker 1: It can go over rocks and twigs and even logs 677 00:38:33,800 --> 00:38:36,319 Speaker 1: and stuff. So it's a pretty interesting and robust looking 678 00:38:36,360 --> 00:38:39,360 Speaker 1: little critter. And again it's sort of a proof of 679 00:38:39,440 --> 00:38:42,680 Speaker 1: concept of technologies that could be used in future robots. 680 00:38:42,960 --> 00:38:46,239 Speaker 1: There's also sand Flea, which is a wheeled robot, an 681 00:38:46,320 --> 00:38:51,000 Speaker 1: unusual thing for Boston Dynamic or Boston Dynamics, I should say. 682 00:38:51,480 --> 00:38:53,840 Speaker 1: It can jump actually, so it's a wheeled robot, but 683 00:38:53,840 --> 00:38:56,759 Speaker 1: it can also leap into the air to the tune 684 00:38:56,800 --> 00:39:00,000 Speaker 1: of about thirty feet in the air. It uses gyroscope 685 00:39:00,000 --> 00:39:04,000 Speaker 1: hopes to maintain its orientation during jumps so it doesn't 686 00:39:04,040 --> 00:39:06,839 Speaker 1: just tumble to the ground and you know, crash into 687 00:39:06,880 --> 00:39:08,520 Speaker 1: lots of little bit so it can actually land on 688 00:39:08,520 --> 00:39:11,440 Speaker 1: its wheels properly and then jump again. There's some great 689 00:39:11,560 --> 00:39:14,800 Speaker 1: footage of it jumping from the ground onto the top 690 00:39:14,880 --> 00:39:17,799 Speaker 1: of a building that I was really impressed by, so 691 00:39:17,840 --> 00:39:20,000 Speaker 1: you should check that out to sand Flea is the 692 00:39:20,080 --> 00:39:22,360 Speaker 1: name of that one. Then we get to the humanoid 693 00:39:22,480 --> 00:39:26,040 Speaker 1: robots that Boston Dynamics has built. Now, the ones I've 694 00:39:26,080 --> 00:39:29,239 Speaker 1: mentioned earlier were mostly based off of animal designs, with 695 00:39:29,360 --> 00:39:31,880 Speaker 1: that exception of sand Flea because that's a wheeled robot. 696 00:39:32,239 --> 00:39:34,600 Speaker 1: But these are all bipedal machines that I'm going to 697 00:39:34,680 --> 00:39:38,680 Speaker 1: talk about at this point. First, there's the Protection Ensemble 698 00:39:38,880 --> 00:39:42,880 Speaker 1: Test mannequin, also known as pet Man. This is an 699 00:39:42,880 --> 00:39:47,360 Speaker 1: anthropomorphic robot that's pretty basic. Its movements were largely influenced 700 00:39:47,360 --> 00:39:50,360 Speaker 1: by the work done on the Big Dog project. In 701 00:39:50,400 --> 00:39:54,640 Speaker 1: other words, it lumbers around a little bit. It when 702 00:39:54,640 --> 00:39:58,080 Speaker 1: it's not wearing anything, it does not look terribly human. 703 00:39:58,120 --> 00:40:01,400 Speaker 1: I mean, it's basic human shape that's got legs and arms. 704 00:40:02,120 --> 00:40:04,680 Speaker 1: But it does important work for humans, and that is 705 00:40:04,719 --> 00:40:07,400 Speaker 1: to test out chemical protection suits to make sure they 706 00:40:07,400 --> 00:40:10,960 Speaker 1: are of suitable durability for use by real, live human beings. 707 00:40:11,960 --> 00:40:15,359 Speaker 1: So a suit might protect against chemicals when it's nice 708 00:40:15,360 --> 00:40:17,160 Speaker 1: and new and when it's just kind of folded up, 709 00:40:17,760 --> 00:40:20,480 Speaker 1: But on the real world, a suit's gonna be exposed 710 00:40:20,480 --> 00:40:24,280 Speaker 1: to stresses and wear and tear, and so pet Man 711 00:40:24,400 --> 00:40:27,920 Speaker 1: could simulate that by moving around while a suit is 712 00:40:27,960 --> 00:40:30,080 Speaker 1: actually on the device. It would be wearing one of 713 00:40:30,080 --> 00:40:33,440 Speaker 1: these chemical protection suits, and it could test the durability 714 00:40:33,440 --> 00:40:36,479 Speaker 1: of the suit and actually even do this while being 715 00:40:36,520 --> 00:40:41,479 Speaker 1: exposed to different types of stresses like temperature changes or 716 00:40:41,520 --> 00:40:45,320 Speaker 1: even chemicals. You could spray chemicals on this thing, because 717 00:40:45,840 --> 00:40:47,640 Speaker 1: it's a robot, you know, you wouldn't want to do 718 00:40:47,680 --> 00:40:49,560 Speaker 1: that to a real human. You'd put them to way 719 00:40:49,640 --> 00:40:53,720 Speaker 1: too much risk. If in fact, the chemical resistance suit 720 00:40:53,840 --> 00:40:57,000 Speaker 1: didn't live up to its design, but with a robot, 721 00:40:57,040 --> 00:40:59,640 Speaker 1: you don't have to worry about that, So that's exactly 722 00:40:59,760 --> 00:41:02,440 Speaker 1: what they would use it for. It's kind of kind 723 00:41:02,440 --> 00:41:06,120 Speaker 1: of interesting. Um and it's also fun to watch videos 724 00:41:06,160 --> 00:41:08,799 Speaker 1: of it wearing the suits because then it just looks 725 00:41:08,800 --> 00:41:13,160 Speaker 1: like a It kind of looks like a slasher movie monster, 726 00:41:13,320 --> 00:41:18,000 Speaker 1: like Jason or Mike Myers, Mike Wires. That's just this big, 727 00:41:18,120 --> 00:41:23,240 Speaker 1: lumbering hulk of a thing and uh typically camouflage chemical suits. 728 00:41:23,520 --> 00:41:28,400 Speaker 1: It's kind of terrifying. Then there's Atlas, the agile anthropomorphic robot. 729 00:41:28,520 --> 00:41:32,480 Speaker 1: The first generation of Atlas was a pretty monstrous beast itself. 730 00:41:32,520 --> 00:41:36,880 Speaker 1: It measured six ft tall uh three centimeters in other words, 731 00:41:37,040 --> 00:41:40,480 Speaker 1: and waited a hundred fifty ms or three thirty pounds. 732 00:41:41,520 --> 00:41:44,640 Speaker 1: It's sensor's range finders and optical cameras communicated with a 733 00:41:44,640 --> 00:41:48,120 Speaker 1: computer that was separate from the robot itself in the 734 00:41:48,160 --> 00:41:51,960 Speaker 1: original prototype model, so in other words, all the computational 735 00:41:51,960 --> 00:41:56,320 Speaker 1: stuff was offloaded onto a nearby unit. According to Boston Dynamics, 736 00:41:56,680 --> 00:41:59,160 Speaker 1: the robot had a total of twenty eight degrees of 737 00:41:59,239 --> 00:42:02,520 Speaker 1: freedom when you added up all of its various actuated joints. 738 00:42:02,800 --> 00:42:06,160 Speaker 1: The earliest version required a tethered power supply, and then 739 00:42:06,280 --> 00:42:09,000 Speaker 1: later builds would incorporate the power onto the robot itself, 740 00:42:09,440 --> 00:42:11,719 Speaker 1: and its movement was largely based off the results of 741 00:42:11,760 --> 00:42:15,360 Speaker 1: developing pet Man. So while pet Man was kind of 742 00:42:15,360 --> 00:42:18,200 Speaker 1: an evolution of the work that was done on Big Dog, 743 00:42:18,880 --> 00:42:20,799 Speaker 1: Atlas was an evolution of the work that was done 744 00:42:20,800 --> 00:42:23,719 Speaker 1: on pet Man. And Atlas could do some pretty cool stuff. 745 00:42:23,719 --> 00:42:26,800 Speaker 1: It could balance on one leg, which is really remarkable 746 00:42:26,800 --> 00:42:29,960 Speaker 1: when you think about how big this thing was, and 747 00:42:30,000 --> 00:42:34,200 Speaker 1: also it could recover from being shoved uh or sometimes 748 00:42:34,280 --> 00:42:36,160 Speaker 1: it could. Other times it would lose its balance and 749 00:42:36,200 --> 00:42:40,719 Speaker 1: unpredictably fall over, sometimes without even being shoved. First, it 750 00:42:40,760 --> 00:42:43,279 Speaker 1: was that generation of Atlas that would be featured in 751 00:42:43,320 --> 00:42:47,000 Speaker 1: the Team DARPA Robotics Challenge. I did a full episode 752 00:42:47,400 --> 00:42:51,120 Speaker 1: about the Robotics Challenge, but I would give a quick summary. Here. 753 00:42:52,160 --> 00:42:54,359 Speaker 1: You had a bunch of teams that were competing in 754 00:42:54,400 --> 00:42:57,719 Speaker 1: designing a robotics solution to a series of problems that 755 00:42:57,800 --> 00:43:00,440 Speaker 1: simulated the sorts of task a robot might have to 756 00:43:00,480 --> 00:43:03,640 Speaker 1: complete in the wake of a disaster. The inspiration for 757 00:43:03,760 --> 00:43:07,680 Speaker 1: the challenge was a real world disaster The Fukushima nuclear 758 00:43:07,719 --> 00:43:11,400 Speaker 1: reactors when they malfunctioned after its tsunami um It was 759 00:43:11,520 --> 00:43:14,080 Speaker 1: very dangerous to send people into that area, but a 760 00:43:14,200 --> 00:43:17,560 Speaker 1: robot would be a much better choice. So this was 761 00:43:17,640 --> 00:43:20,319 Speaker 1: kind of a challenge to see if any teams could 762 00:43:20,360 --> 00:43:23,680 Speaker 1: design robotics platforms that could respond to the sort of 763 00:43:23,680 --> 00:43:27,080 Speaker 1: things that would be necessary in another disaster of that caliber. 764 00:43:28,040 --> 00:43:30,120 Speaker 1: So the robots would have to operate a vehicle, a 765 00:43:30,200 --> 00:43:32,839 Speaker 1: human powered vehicle, typically not a human power, but human 766 00:43:32,880 --> 00:43:36,040 Speaker 1: operated vehicle, to a specific destination, so in other words, 767 00:43:36,160 --> 00:43:38,880 Speaker 1: like a golf cart. They'd have to drive this themselves, 768 00:43:39,600 --> 00:43:43,000 Speaker 1: get out of that vehicle, open a door, walk through 769 00:43:43,000 --> 00:43:46,360 Speaker 1: a doorway, and then complete several other challenges, including the 770 00:43:46,360 --> 00:43:49,760 Speaker 1: operation of a handheld power tool. At one point, six 771 00:43:49,880 --> 00:43:53,399 Speaker 1: of the competing teams would use the Atlas robot as 772 00:43:53,400 --> 00:43:58,000 Speaker 1: their robotics platform. Other teams would rely on different humanoid robots, 773 00:43:58,040 --> 00:44:01,160 Speaker 1: so they weren't required to use Atlas, but six teams 774 00:44:01,200 --> 00:44:04,000 Speaker 1: did use Atlas. The nature of the challenges meant that 775 00:44:04,040 --> 00:44:06,480 Speaker 1: the robot would have to simulate human activities, and the 776 00:44:06,480 --> 00:44:09,880 Speaker 1: winning team, Team KIST, used a robot called Hubo, so 777 00:44:09,960 --> 00:44:12,560 Speaker 1: it was not an Atlas. There are a lot of 778 00:44:12,640 --> 00:44:15,879 Speaker 1: videos of the competition out there, including some of the 779 00:44:15,960 --> 00:44:20,799 Speaker 1: various Atlas robots. There are also some great videos of 780 00:44:20,880 --> 00:44:24,680 Speaker 1: heartbreaking and occasionally incredibly funny moments of robots falling over 781 00:44:24,719 --> 00:44:27,640 Speaker 1: while trying to do seemingly simple things like walk through 782 00:44:27,680 --> 00:44:31,960 Speaker 1: a doorway. As it turns out, that's a lot trickier 783 00:44:31,960 --> 00:44:35,480 Speaker 1: than it sounds. In two thousand and sixteen, Boston Dynamics 784 00:44:35,520 --> 00:44:38,760 Speaker 1: unveiled a second generation of Atlas robots that looked slightly 785 00:44:38,840 --> 00:44:42,600 Speaker 1: less industrial and had more advanced sensors for navigation and 786 00:44:42,680 --> 00:44:47,160 Speaker 1: orientation purposes, as well as more advanced maneuverability technology. It 787 00:44:47,239 --> 00:44:50,640 Speaker 1: could walk on surfaces like snow, and even planned paths 788 00:44:50,640 --> 00:44:54,239 Speaker 1: in areas where other objects are currently in motion. This 789 00:44:54,280 --> 00:44:57,360 Speaker 1: one was a seventy five tall, so a little shorter. 790 00:44:57,600 --> 00:45:01,359 Speaker 1: That's about five ft nine inches tall. It two ms 791 00:45:01,480 --> 00:45:03,880 Speaker 1: or around a d eighty pounds, a little you know, 792 00:45:04,840 --> 00:45:09,000 Speaker 1: significantly lighter than three pounds. Boston Dynamics has credited three 793 00:45:09,080 --> 00:45:12,360 Speaker 1: D printing for the drastic reduction in weight and size, 794 00:45:13,440 --> 00:45:16,719 Speaker 1: and there's an even smaller, lighter, and more capable generation 795 00:45:16,719 --> 00:45:22,040 Speaker 1: of Atlas the debuted in late Boston Dynamics posted a 796 00:45:22,120 --> 00:45:24,520 Speaker 1: video of it showing this robot in action. It could 797 00:45:24,600 --> 00:45:27,680 Speaker 1: jump up on boxes, so it could walk up to 798 00:45:27,920 --> 00:45:31,200 Speaker 1: an obstacle like a box, judge how tall it was, 799 00:45:31,320 --> 00:45:34,719 Speaker 1: and then jump so that it landed and maintained its 800 00:45:34,719 --> 00:45:37,040 Speaker 1: balance on top of the box. It could even do 801 00:45:37,120 --> 00:45:39,640 Speaker 1: a full back flip off a box and land on 802 00:45:39,680 --> 00:45:42,879 Speaker 1: the ground, regaining its balance so it doesn't even tip over, 803 00:45:43,600 --> 00:45:45,839 Speaker 1: which is an incredible thing to see. If you haven't 804 00:45:45,840 --> 00:45:49,239 Speaker 1: watched that video, you should definitely find that one. And 805 00:45:49,320 --> 00:45:52,759 Speaker 1: that brings us to handle a robot that seems to 806 00:45:52,840 --> 00:45:56,080 Speaker 1: look like kind of like a bucking horse on roller skates. 807 00:45:56,080 --> 00:45:57,840 Speaker 1: I've seen other people refer to it as kind of 808 00:45:57,880 --> 00:46:01,280 Speaker 1: like a donkey on roller skates. It's a bipedal robot, 809 00:46:01,640 --> 00:46:05,000 Speaker 1: but it's lower legs don't end in feet, they end 810 00:46:05,040 --> 00:46:07,239 Speaker 1: in wheels, so it has two wheels at its base, 811 00:46:07,320 --> 00:46:09,399 Speaker 1: kind of like a segue, and it uses that same 812 00:46:09,440 --> 00:46:13,879 Speaker 1: sort of technology to maintain its upright position. Has two 813 00:46:13,960 --> 00:46:16,640 Speaker 1: upper limbs that can pick up and carry objects weighing 814 00:46:16,719 --> 00:46:20,680 Speaker 1: up to and the robot is two meters tall, weighs 815 00:46:20,680 --> 00:46:24,120 Speaker 1: a hundred and five kilograms, and has ten actuated joints. 816 00:46:24,640 --> 00:46:27,279 Speaker 1: It uses battery power and a combination of hydraulic and 817 00:46:27,320 --> 00:46:31,600 Speaker 1: electric actuators to actually work those limbs, and has depth 818 00:46:31,640 --> 00:46:35,120 Speaker 1: cameras to provide perception. It can speed along the ground 819 00:46:35,200 --> 00:46:39,279 Speaker 1: and even jump up onto higher surfaces. One demonstration, there's 820 00:46:39,320 --> 00:46:41,080 Speaker 1: a video that you can watch of this as well. 821 00:46:41,560 --> 00:46:44,480 Speaker 1: It will show the robot rolling toward a table and 822 00:46:44,520 --> 00:46:47,400 Speaker 1: then the last second it jumps up and lands on 823 00:46:47,440 --> 00:46:51,080 Speaker 1: the table surface, rolls across the table before landing on 824 00:46:51,080 --> 00:46:54,520 Speaker 1: the other side and continuing on. It's pretty impressive, and 825 00:46:54,600 --> 00:46:57,200 Speaker 1: Boston Dynamics has also pointed out that because it only 826 00:46:57,239 --> 00:47:01,839 Speaker 1: has ten actuated joints, it's far less complex than some 827 00:47:01,920 --> 00:47:05,319 Speaker 1: other models, and that uh that reduction in complexity could 828 00:47:05,320 --> 00:47:09,000 Speaker 1: mean a reduction in cost as well. The wheels work 829 00:47:09,040 --> 00:47:12,560 Speaker 1: on most services until you get into some really difficult terrain, 830 00:47:13,040 --> 00:47:15,080 Speaker 1: and I find it fascinating that a company that had 831 00:47:15,120 --> 00:47:19,440 Speaker 1: made its name from mostly legged robot designs had gone 832 00:47:19,520 --> 00:47:23,239 Speaker 1: with this hybrid approach. But the videos show that the 833 00:47:23,280 --> 00:47:27,040 Speaker 1: design is pretty effective, at least for specific applications. Now, 834 00:47:27,080 --> 00:47:30,400 Speaker 1: one thing I have not really covered in this episode 835 00:47:31,000 --> 00:47:34,920 Speaker 1: is who owns Boston Dynamics these days, because that's actually 836 00:47:35,000 --> 00:47:37,120 Speaker 1: changed a couple of times over the past few years. 837 00:47:37,120 --> 00:47:39,640 Speaker 1: It operated as an independent company for more than a 838 00:47:39,640 --> 00:47:41,719 Speaker 1: couple of well a couple of decades, and then on 839 00:47:41,760 --> 00:47:46,600 Speaker 1: December thirteen, the top secret R and D branch of Google, 840 00:47:47,160 --> 00:47:49,839 Speaker 1: known as Google X and these days just called X 841 00:47:50,120 --> 00:47:54,560 Speaker 1: acquired Boston Dynamics um as X it is now a 842 00:47:54,600 --> 00:47:59,040 Speaker 1: subsidiary company of Google's parent company, alphabet So. At the time, 843 00:47:59,120 --> 00:48:01,560 Speaker 1: it was a division under Google. Now it's its own 844 00:48:01,600 --> 00:48:05,440 Speaker 1: spin off company called x SO. Boston Dynamics was just 845 00:48:05,600 --> 00:48:09,960 Speaker 1: one of nine robotics companies that Google acquired around that time, 846 00:48:10,200 --> 00:48:13,680 Speaker 1: and the co founder of Android, Andy Reuben, was slated 847 00:48:13,680 --> 00:48:16,160 Speaker 1: to take the helm of all nine companies under a 848 00:48:16,160 --> 00:48:21,200 Speaker 1: new robotics division that was called Replicant, which isn't terrifying 849 00:48:21,239 --> 00:48:25,120 Speaker 1: at all thanks a lot Blade Runner well. According to Reuben, 850 00:48:25,920 --> 00:48:28,520 Speaker 1: the initial plan was to have those companies do their 851 00:48:28,520 --> 00:48:31,640 Speaker 1: own thing, working on research the way they normally would 852 00:48:31,719 --> 00:48:34,400 Speaker 1: and advancing the art and science of robotics, so in 853 00:48:34,440 --> 00:48:37,640 Speaker 1: other words, there wouldn't be any pressure to develop commercial 854 00:48:37,640 --> 00:48:40,520 Speaker 1: products right out of the gate. Instead, they would just 855 00:48:40,520 --> 00:48:43,759 Speaker 1: do exploratory research and the advances in the field would 856 00:48:43,800 --> 00:48:47,040 Speaker 1: guide Google's decisions on how to leverage those developments in 857 00:48:47,160 --> 00:48:52,720 Speaker 1: commercial applications. But in October, Andy Reuben would leave Google. 858 00:48:53,280 --> 00:48:56,480 Speaker 1: Now Reuben reportedly left to found a new company to 859 00:48:56,520 --> 00:48:59,400 Speaker 1: act as an incubator for technology hardware products. So, in 860 00:48:59,440 --> 00:49:01,359 Speaker 1: other words, he wanted to build a company that would 861 00:49:01,400 --> 00:49:04,799 Speaker 1: help hardware developers get to market with their ideas. That 862 00:49:04,840 --> 00:49:10,560 Speaker 1: company is called Essential. But later in information would leak 863 00:49:10,600 --> 00:49:14,640 Speaker 1: that Reuben had apparently or allegedly i should say, left 864 00:49:14,680 --> 00:49:18,799 Speaker 1: Google after a one of his employees, a woman, had 865 00:49:18,840 --> 00:49:23,360 Speaker 1: filed a complaint to human resources about Reuben having an 866 00:49:23,400 --> 00:49:28,800 Speaker 1: inappropriate relationship in the workplace. Now, Reuben's spokesperson over it 867 00:49:28,960 --> 00:49:32,600 Speaker 1: Essential says that this report is misleading, that Reuben had 868 00:49:32,640 --> 00:49:36,279 Speaker 1: never engaged in any misconduct at Google, that any relationships 869 00:49:36,280 --> 00:49:39,880 Speaker 1: he had at Google were consensual, uh, and that there 870 00:49:39,920 --> 00:49:43,840 Speaker 1: were no other issues there. But whatever the case, Reuben 871 00:49:43,920 --> 00:49:46,680 Speaker 1: took a leave of absence at Essential to deal with 872 00:49:46,760 --> 00:49:50,319 Speaker 1: quote personal matters in the quote. So back to the 873 00:49:50,400 --> 00:49:55,200 Speaker 1: robotics companies under X. So Reuben leaves the company, and 874 00:49:55,360 --> 00:49:58,640 Speaker 1: there was kind of a vacuuman leadership at that part 875 00:49:58,719 --> 00:50:01,000 Speaker 1: of Google, and the is a lot of confusion about 876 00:50:01,040 --> 00:50:03,040 Speaker 1: what was going to happen after all these companies had 877 00:50:03,080 --> 00:50:05,799 Speaker 1: been told that they were going to continue operating as normal, 878 00:50:05,960 --> 00:50:11,960 Speaker 1: really doing this exploratory research. But in Google management attempted 879 00:50:11,960 --> 00:50:14,960 Speaker 1: to put some sort of leadership and structure in place, 880 00:50:15,200 --> 00:50:19,200 Speaker 1: and reportedly the change came with an intent to make 881 00:50:19,360 --> 00:50:23,960 Speaker 1: actual commercial products, so the divisions would have to dedicate 882 00:50:24,000 --> 00:50:27,320 Speaker 1: themselves to building stuff that could be sold. The mission 883 00:50:27,320 --> 00:50:31,080 Speaker 1: statement was effectively being changed from do exploratory research to 884 00:50:31,160 --> 00:50:34,600 Speaker 1: build us something that we can market now. According to 885 00:50:34,680 --> 00:50:38,719 Speaker 1: Business Insider, Boston Dynamics was told to start working on 886 00:50:38,800 --> 00:50:42,160 Speaker 1: designing consumer robots. Now. These would have to be quiet, 887 00:50:42,280 --> 00:50:44,479 Speaker 1: and they'd have to be suitable for operating around human 888 00:50:44,520 --> 00:50:46,800 Speaker 1: beings of all ages. And this was a really narrow 889 00:50:46,880 --> 00:50:49,680 Speaker 1: focus and something that many people at Boston Dynamics were 890 00:50:49,680 --> 00:50:54,759 Speaker 1: not necessarily keen to work on. For one thing, they 891 00:50:54,760 --> 00:50:57,799 Speaker 1: wanted to continue their work in pushing the boundaries and 892 00:50:57,800 --> 00:51:01,560 Speaker 1: looking into all sorts of mechanisms to operate robots, including 893 00:51:01,600 --> 00:51:04,200 Speaker 1: ones that would not be suitable for a consumer robot. 894 00:51:04,719 --> 00:51:07,759 Speaker 1: For example, hydraulic actuators. You know, they're much louder than 895 00:51:07,760 --> 00:51:10,960 Speaker 1: electronic ones, but more powerful, and therefore they can give 896 00:51:11,000 --> 00:51:14,960 Speaker 1: a lot more uh uses applications for robots, but ones 897 00:51:15,000 --> 00:51:19,560 Speaker 1: that wouldn't necessarily be appropriate for a consumer product. Jonathan Rosenberg, 898 00:51:19,600 --> 00:51:22,880 Speaker 1: who had taken the helm of the replicant division, stated, quote, we, 899 00:51:23,040 --> 00:51:26,160 Speaker 1: as a startup of our size, cannot spend thirty plus 900 00:51:26,200 --> 00:51:29,320 Speaker 1: percent of our resources on things that take ten years 901 00:51:29,480 --> 00:51:33,720 Speaker 1: end quote. So he was saying, we can't really afford 902 00:51:33,760 --> 00:51:37,480 Speaker 1: the luxury of doing exploratory research with the idea that 903 00:51:37,600 --> 00:51:39,840 Speaker 1: sometime down the line it's going to pay off. We 904 00:51:39,920 --> 00:51:42,560 Speaker 1: have to start making things that we can actually sell. 905 00:51:43,000 --> 00:51:44,960 Speaker 1: So from a business perspective, you can kind of see 906 00:51:45,000 --> 00:51:47,799 Speaker 1: where he's coming from. But from the perspective of people 907 00:51:47,800 --> 00:51:52,120 Speaker 1: at Boston Dynamics, this was very antithetical to what they 908 00:51:52,480 --> 00:51:56,880 Speaker 1: had been doing up to that point. So Boston Dynamics 909 00:51:57,200 --> 00:52:00,120 Speaker 1: would eventually go on to release a video of Atlas 910 00:52:00,200 --> 00:52:03,319 Speaker 1: stacking boxes in an application that might be suitable for 911 00:52:03,360 --> 00:52:07,799 Speaker 1: something like a warehouse, But apparently that quote soured the 912 00:52:07,880 --> 00:52:12,120 Speaker 1: soup end quote with Google. Google management wanted a consumer 913 00:52:12,160 --> 00:52:14,719 Speaker 1: facing product, not something that was going to have applications 914 00:52:14,719 --> 00:52:18,120 Speaker 1: in industrial settings, and so at that point Google began 915 00:52:18,160 --> 00:52:22,719 Speaker 1: to shop around for another suitor to take over Boston Dynamics. Reportedly, 916 00:52:23,719 --> 00:52:27,040 Speaker 1: one of those potential parties was the Toyota Research Institute, 917 00:52:27,160 --> 00:52:31,640 Speaker 1: but ultimately Google found a buyer in a Japanese company 918 00:52:31,680 --> 00:52:36,000 Speaker 1: called soft Bank. Soft Bank also bought another robotics company 919 00:52:36,040 --> 00:52:40,000 Speaker 1: off of Google slash alphabet called Shaft, which had competed 920 00:52:40,040 --> 00:52:44,080 Speaker 1: in that DARPA robotics challenge I mentioned earlier. Masa Yoshi, 921 00:52:44,120 --> 00:52:48,040 Speaker 1: san CEO of soft Bank, released this statement when news 922 00:52:48,080 --> 00:52:51,720 Speaker 1: broke that the company was acquiring Boston Dynamics. Quote, today, 923 00:52:51,760 --> 00:52:54,839 Speaker 1: there are many issues we still cannot solve by ourselves 924 00:52:54,840 --> 00:52:58,120 Speaker 1: with human capabilities. Smart robotics are going to be a 925 00:52:58,239 --> 00:53:01,680 Speaker 1: key driver of the next stage of information revolution, and 926 00:53:01,719 --> 00:53:04,240 Speaker 1: Mark and his team at Boston Dynamics are the clear 927 00:53:04,320 --> 00:53:08,920 Speaker 1: technology technology leaders in advanced dynamic robots. I am thrilled 928 00:53:09,040 --> 00:53:11,919 Speaker 1: to welcome them to the soft Bank family and look 929 00:53:11,960 --> 00:53:14,440 Speaker 1: forward to supporting them as they continue to advance the 930 00:53:14,440 --> 00:53:17,600 Speaker 1: field of robotics and explore applications that can help make 931 00:53:17,640 --> 00:53:22,560 Speaker 1: life easier, safer, and more fulfilling. So that's the story 932 00:53:22,640 --> 00:53:25,600 Speaker 1: as it stands now with Boston Dynamics. There's a lot 933 00:53:25,640 --> 00:53:28,080 Speaker 1: of hope over at Boston Dynamics that they will be 934 00:53:28,120 --> 00:53:31,600 Speaker 1: able to continue their mission that they had been following 935 00:53:31,640 --> 00:53:35,920 Speaker 1: for years before the Google acquisition. There's also obviously some 936 00:53:36,000 --> 00:53:38,840 Speaker 1: concern that maybe soft Bank will follow suit with Google 937 00:53:38,840 --> 00:53:40,719 Speaker 1: and say, hey, can you make us a robot that 938 00:53:41,560 --> 00:53:44,359 Speaker 1: you know, we'll bake cheesecake and greet people at the door. 939 00:53:45,280 --> 00:53:48,840 Speaker 1: And I totally want a cheesecake robot now. And I 940 00:53:48,880 --> 00:53:50,719 Speaker 1: need to think these things through before I say them, 941 00:53:50,760 --> 00:53:52,319 Speaker 1: because this was not in my notes. It was just 942 00:53:53,160 --> 00:53:56,399 Speaker 1: kind of off the cuff. And Okay, guys, we're gonna 943 00:53:56,400 --> 00:53:59,400 Speaker 1: wrap this up. So that's the story of Boston Dynamics 944 00:53:59,440 --> 00:54:01,319 Speaker 1: so far. I can't wait to see what comes out 945 00:54:01,360 --> 00:54:05,799 Speaker 1: of that story next. But if you guys have suggestions 946 00:54:05,840 --> 00:54:08,080 Speaker 1: for things I should cover in future episodes of tech Stuff, 947 00:54:08,080 --> 00:54:10,919 Speaker 1: you should write me the email addresses tech Stuff at 948 00:54:10,920 --> 00:54:13,400 Speaker 1: how stuff works dot com, or you can drop me 949 00:54:13,400 --> 00:54:15,399 Speaker 1: a line on Facebook or Twitter. The handle of both 950 00:54:15,440 --> 00:54:18,120 Speaker 1: of those is tech Stuff hs W. Remember we also 951 00:54:18,160 --> 00:54:21,360 Speaker 1: have an Instagram account, and you can watch me record 952 00:54:21,400 --> 00:54:24,920 Speaker 1: shows live on Wednesdays and Fridays over at twitch dot 953 00:54:25,000 --> 00:54:28,400 Speaker 1: tv slash tech Stuff. There's a schedule right there. You 954 00:54:28,400 --> 00:54:30,279 Speaker 1: can check that out and chat with me in the 955 00:54:30,320 --> 00:54:34,200 Speaker 1: chat room if you like. I'll enjoy that interaction. I'm 956 00:54:34,200 --> 00:54:36,880 Speaker 1: wrapping this up because there's a cheesecake out there with 957 00:54:36,920 --> 00:54:39,160 Speaker 1: my name on it. I gotta go find it. And 958 00:54:39,320 --> 00:54:42,880 Speaker 1: you guys, well, I'll talk to you again really soon 959 00:54:48,680 --> 00:54:51,160 Speaker 1: for more on this and thousands of other topics because 960 00:54:51,160 --> 00:55:00,319 Speaker 1: it has staff works dot com zo