1 00:00:04,120 --> 00:00:07,160 Speaker 1: Get in touch with technology with tech Stuff from how 2 00:00:07,200 --> 00:00:13,800 Speaker 1: stuff works dot com. Hey there, and welcome to tech Stuff. 3 00:00:13,800 --> 00:00:16,720 Speaker 1: I'm your host, Jonathan Strickland. I'm an executive producer at 4 00:00:16,720 --> 00:00:20,640 Speaker 1: how stuff Works. My love all things tech and in 5 00:00:20,760 --> 00:00:25,079 Speaker 1: late June, Honda announced it would stopped production on its 6 00:00:25,160 --> 00:00:29,560 Speaker 1: humanoid robot Awesomo, leveraging the work that went into creating 7 00:00:29,560 --> 00:00:33,440 Speaker 1: the bot for other applications. Tech Stuff super fan Cat, 8 00:00:33,479 --> 00:00:37,919 Speaker 1: who loves this robot, requested I do a show about Osomo, 9 00:00:38,280 --> 00:00:41,839 Speaker 1: and I'm happy to do it. Fun fact before we 10 00:00:41,880 --> 00:00:46,480 Speaker 1: get into the Osomo story and what made it special 11 00:00:46,640 --> 00:00:50,520 Speaker 1: and how it worked, the very first article I ever 12 00:00:50,720 --> 00:00:54,000 Speaker 1: wrote for how stuff works dot Com once I was 13 00:00:54,080 --> 00:00:58,360 Speaker 1: hired was how Osomo Works Now. I technically rewrote a 14 00:00:58,400 --> 00:01:02,240 Speaker 1: couple of articles before, at where they wanted to test 15 00:01:02,320 --> 00:01:05,039 Speaker 1: my abilities by giving me an existing article and saying 16 00:01:05,040 --> 00:01:07,920 Speaker 1: could you rewrite this? Could you update it? But the 17 00:01:07,959 --> 00:01:13,199 Speaker 1: first full assignment beginning to end was how Osomo works. Um, 18 00:01:13,280 --> 00:01:15,000 Speaker 1: if you want to read one of my even earlier ones, 19 00:01:15,040 --> 00:01:18,040 Speaker 1: I did a rewrite on electronic voting machines, but Asomo 20 00:01:18,160 --> 00:01:22,160 Speaker 1: was my first full article. The article is still up 21 00:01:22,200 --> 00:01:24,600 Speaker 1: at how stuff works dot com. But it also was 22 00:01:24,720 --> 00:01:28,520 Speaker 1: updated at some point by lee An Aubringer. So I 23 00:01:28,560 --> 00:01:31,640 Speaker 1: didn't I wasn't consulted about that, because that's not the 24 00:01:31,640 --> 00:01:33,680 Speaker 1: way it works. Typically when we write articles for how 25 00:01:33,720 --> 00:01:38,880 Speaker 1: stuff works. As you know, technology evolves very quickly over time, 26 00:01:38,920 --> 00:01:43,440 Speaker 1: and so articles that you write will become outdated rapidly, 27 00:01:44,080 --> 00:01:46,480 Speaker 1: and so we would frequently bring in other people to 28 00:01:46,520 --> 00:01:50,840 Speaker 1: help update articles while our staff writers like myself would 29 00:01:50,840 --> 00:01:55,200 Speaker 1: be on other new, big assignments. So, but this is 30 00:01:55,280 --> 00:01:59,600 Speaker 1: all about AWESOMO. So what was awesomo. The name is 31 00:01:59,640 --> 00:02:05,160 Speaker 1: actually an acronym. It stands for Advanced Step in Innovative Mobility, 32 00:02:05,360 --> 00:02:08,040 Speaker 1: and it's a robot that's intended to interact in a 33 00:02:08,200 --> 00:02:12,400 Speaker 1: human living environment, meaning that Honda was trying to build 34 00:02:12,440 --> 00:02:15,720 Speaker 1: a robot that could integrate into our lives, that could 35 00:02:15,720 --> 00:02:19,680 Speaker 1: move around with human beings in a seamless way, and 36 00:02:19,720 --> 00:02:22,400 Speaker 1: it was meant to be a humanoid robot that could 37 00:02:22,840 --> 00:02:25,880 Speaker 1: interact with our environment similar to the way we humans do, 38 00:02:26,080 --> 00:02:30,079 Speaker 1: so it needed to have human like appendages. And it 39 00:02:30,080 --> 00:02:32,360 Speaker 1: turns out this is a lot easier said than done, 40 00:02:32,400 --> 00:02:34,800 Speaker 1: because a lot of consideration has to go into the 41 00:02:34,840 --> 00:02:39,120 Speaker 1: design of such a robot. On casual glance. Awesome, Oh, 42 00:02:39,200 --> 00:02:43,160 Speaker 1: looked like a child sized astronaut, complete with like a 43 00:02:43,280 --> 00:02:47,440 Speaker 1: space pack and a space helmet. It's humanoid, has arms 44 00:02:47,520 --> 00:02:51,040 Speaker 1: as legs as hands with fingers and thumbs, and like 45 00:02:51,080 --> 00:02:53,200 Speaker 1: I said, the head looks like a like a space 46 00:02:53,240 --> 00:02:55,760 Speaker 1: helmet that you would see an astronaut on the Moon wearing. 47 00:02:55,840 --> 00:02:59,040 Speaker 1: And it had it's batterying a big backpack that sat 48 00:02:59,080 --> 00:03:01,720 Speaker 1: on its back obvious lee which gave it this kind 49 00:03:01,720 --> 00:03:05,600 Speaker 1: of little estronaut looks very cute, and in fact, that 50 00:03:05,639 --> 00:03:08,079 Speaker 1: was done on purpose. They wanted the robot to look 51 00:03:08,120 --> 00:03:12,360 Speaker 1: friendly and approachable because it was meant to interact with people. UH. 52 00:03:12,480 --> 00:03:16,400 Speaker 1: Over time, the design of Osumo evolved a little bit, 53 00:03:16,720 --> 00:03:19,080 Speaker 1: and I'll cover some of that in this episode, but 54 00:03:19,200 --> 00:03:21,560 Speaker 1: I just want to illustrate it here with the earliest 55 00:03:21,680 --> 00:03:26,880 Speaker 1: version of the official Osomo, with the version that was 56 00:03:26,960 --> 00:03:31,560 Speaker 1: around just before they stopped production on it. In the 57 00:03:31,600 --> 00:03:35,360 Speaker 1: first robot that Honda designated as Osomo, because there were 58 00:03:35,400 --> 00:03:41,320 Speaker 1: some predecessors, UH, was born quote unquote on November two thousand. 59 00:03:41,720 --> 00:03:45,680 Speaker 1: It stood one twenty centimeters tall that's just a little 60 00:03:45,680 --> 00:03:49,400 Speaker 1: bit under four feet, and it weighed about fifty or 61 00:03:49,400 --> 00:03:52,680 Speaker 1: a hundred fifteen pounds. But when it was retired in 62 00:03:52,720 --> 00:03:55,840 Speaker 1: two thousand eighteen and Osumo's robots specs were a little different. 63 00:03:55,840 --> 00:03:59,119 Speaker 1: It stood one hundred thirty centimeters tall that's about four 64 00:03:59,160 --> 00:04:01,560 Speaker 1: ft three inches, so it grew a little bit, and 65 00:04:01,560 --> 00:04:04,680 Speaker 1: it weighed forty eight kilograms or about a hundred six pounds, 66 00:04:04,680 --> 00:04:07,520 Speaker 1: so it lost a little weight. At top speed, it 67 00:04:07,560 --> 00:04:10,920 Speaker 1: could dash at a respectable nine kilometers per hour, which 68 00:04:10,920 --> 00:04:13,760 Speaker 1: is about five point six miles per hour. Now, it's 69 00:04:13,760 --> 00:04:15,600 Speaker 1: not like it could go toe to toe with the 70 00:04:15,680 --> 00:04:19,040 Speaker 1: T one thousand from Terminator two, but it was still 71 00:04:19,120 --> 00:04:22,159 Speaker 1: pretty impressive. It was a robot that could run on 72 00:04:22,279 --> 00:04:26,360 Speaker 1: two legs, the first to do so. In fact, Honda 73 00:04:26,480 --> 00:04:28,960 Speaker 1: chose the robot's height carefully. They wanted to design a 74 00:04:29,040 --> 00:04:31,839 Speaker 1: robot that could work in human environments, so it couldn't 75 00:04:31,880 --> 00:04:35,400 Speaker 1: be too small, but it also shouldn't be intimidating. They 76 00:04:35,400 --> 00:04:37,200 Speaker 1: didn't want to make, you know, like a kill bot 77 00:04:37,279 --> 00:04:40,479 Speaker 1: two thousand that would be terrifying. The robot size was 78 00:04:40,520 --> 00:04:43,480 Speaker 1: determined to be people friendly because it'd be large enough 79 00:04:43,760 --> 00:04:48,719 Speaker 1: to operate environmental elements like door knobs or light switches, 80 00:04:49,080 --> 00:04:51,600 Speaker 1: and about four ft three inches meant that it would 81 00:04:51,600 --> 00:04:55,359 Speaker 1: also be around eye level with a seated human adults. 82 00:04:55,400 --> 00:04:57,719 Speaker 1: So if you were seated down at a desk and 83 00:04:57,760 --> 00:04:59,840 Speaker 1: Osimo walked up to you to let you know that 84 00:05:00,000 --> 00:05:02,960 Speaker 1: there was a visitor at the lobby, then you'd be 85 00:05:03,000 --> 00:05:07,560 Speaker 1: looking pretty much I too camera with Osomo. It could 86 00:05:07,560 --> 00:05:09,680 Speaker 1: also stand behind a desk and it would be at 87 00:05:09,720 --> 00:05:12,640 Speaker 1: about the same height as someone who was seated behind 88 00:05:12,680 --> 00:05:16,040 Speaker 1: a desk, So that made Osumo a potential receptionist, which 89 00:05:16,040 --> 00:05:19,080 Speaker 1: in fact, Honda made use of Osimo as a receptionist 90 00:05:19,120 --> 00:05:24,600 Speaker 1: at their headquarters. The robot also had several degrees of freedom. Now, 91 00:05:24,640 --> 00:05:26,680 Speaker 1: the phrase degrees of freedom actually has a couple of 92 00:05:26,680 --> 00:05:30,760 Speaker 1: different meanings depending upon what industry or context you're looking at, 93 00:05:31,200 --> 00:05:34,400 Speaker 1: uh including statistics. There's a specific meaning for degrees of 94 00:05:34,400 --> 00:05:37,560 Speaker 1: freedom and statistics, But in mechanical systems it refers to 95 00:05:37,600 --> 00:05:42,160 Speaker 1: the number of independent movements a rigid body is capable of. 96 00:05:42,600 --> 00:05:46,599 Speaker 1: So an unrestrained rigid body in space just imagine you've 97 00:05:46,600 --> 00:05:50,120 Speaker 1: got a cube of something and it's magically floating in 98 00:05:50,160 --> 00:05:54,640 Speaker 1: the air. It has six degrees of freedom because it 99 00:05:54,680 --> 00:05:58,119 Speaker 1: can move up or down like it can levitate straight 100 00:05:58,200 --> 00:06:02,279 Speaker 1: up or sink straight down. It could strafe left or 101 00:06:02,440 --> 00:06:06,000 Speaker 1: right uh, or it could move back and forth toward 102 00:06:06,080 --> 00:06:09,320 Speaker 1: you or away from you. Plus it can rotate around 103 00:06:09,360 --> 00:06:12,919 Speaker 1: those axes, which would translate to pitch roll and yaw. 104 00:06:13,640 --> 00:06:17,039 Speaker 1: So that's six different degrees of freedom. The final model 105 00:06:17,080 --> 00:06:20,080 Speaker 1: of Osimo had three degrees of freedom for the head, 106 00:06:20,640 --> 00:06:22,880 Speaker 1: seven degrees of freedom for each arm, so I had 107 00:06:22,920 --> 00:06:26,919 Speaker 1: two arms, thirteen degrees of freedom for each hand, two hands, 108 00:06:27,279 --> 00:06:30,160 Speaker 1: two degrees of freedom at the hip, and six degrees 109 00:06:30,160 --> 00:06:33,320 Speaker 1: of freedom for each of the legs two legs, which 110 00:06:33,360 --> 00:06:37,239 Speaker 1: meant that Osimo ultimately had fifty seven degrees of freedom 111 00:06:37,279 --> 00:06:39,680 Speaker 1: for the full robot, the one that was finally retired 112 00:06:39,720 --> 00:06:43,760 Speaker 1: in that was an improvement of twenty three degrees of 113 00:06:43,800 --> 00:06:47,040 Speaker 1: freedom from its predecessor. Model had twenty three degrees more 114 00:06:47,160 --> 00:06:50,800 Speaker 1: freedom than the previous version of Osomo. I'll go into 115 00:06:50,839 --> 00:06:54,800 Speaker 1: further detail about the different sensors and systems Osumo had, 116 00:06:54,839 --> 00:06:56,719 Speaker 1: but first I'd like to talk about the history of 117 00:06:56,720 --> 00:06:58,960 Speaker 1: the robot itself, and to do that we need to 118 00:06:59,000 --> 00:07:02,760 Speaker 1: travel back to night teen eighties six and what a 119 00:07:02,839 --> 00:07:04,960 Speaker 1: year that was, guys. I mean, you could go to 120 00:07:05,000 --> 00:07:06,960 Speaker 1: the theater and you could go see what came out 121 00:07:06,960 --> 00:07:11,200 Speaker 1: in eight six aliens you could see Ferris Bueller's Day Off, 122 00:07:11,920 --> 00:07:14,640 Speaker 1: or you could see the greatest movie made of all time, 123 00:07:14,960 --> 00:07:18,400 Speaker 1: Big Trouble in Little China. Also, Howard the Duck came 124 00:07:18,400 --> 00:07:23,239 Speaker 1: out in as I recall music in six include stuff 125 00:07:23,240 --> 00:07:27,280 Speaker 1: like Peter Gabriel's Sledgehammer or Peter STA's Glory of Love. 126 00:07:27,840 --> 00:07:33,280 Speaker 1: Quite an amazing diddy there or cameos Immortal, classic word Up. 127 00:07:33,520 --> 00:07:36,800 Speaker 1: But more importantly for our story engineers at Honda, we're 128 00:07:36,800 --> 00:07:39,400 Speaker 1: tackling a very challenging problem. How do you make a 129 00:07:39,480 --> 00:07:44,600 Speaker 1: humanoid robot walk like a person does? So they started 130 00:07:44,600 --> 00:07:48,040 Speaker 1: off slowly. First, they began with designing a basic set 131 00:07:48,360 --> 00:07:51,640 Speaker 1: of robotic legs that could take steps. So their first 132 00:07:51,720 --> 00:07:54,320 Speaker 1: robot model in this phase, which was only meant for 133 00:07:54,440 --> 00:07:56,840 Speaker 1: research and development, this was never going to be something 134 00:07:56,840 --> 00:07:59,760 Speaker 1: that they were going to market, was known as the 135 00:08:00,040 --> 00:08:02,640 Speaker 1: e O and it was essentially nothing more than a 136 00:08:02,640 --> 00:08:05,960 Speaker 1: pair of legs connected to a narrow set of robot hips. 137 00:08:06,760 --> 00:08:08,680 Speaker 1: So just legs and hips and that's it. And it 138 00:08:08,760 --> 00:08:11,440 Speaker 1: was wired up and you had a control system that 139 00:08:11,480 --> 00:08:14,920 Speaker 1: would tell it to walk forward. The robot was connected 140 00:08:15,200 --> 00:08:18,480 Speaker 1: directly to those computer systems no wireless systems at this point, 141 00:08:19,120 --> 00:08:24,120 Speaker 1: and the legs had to be very careful, very deliberate 142 00:08:24,120 --> 00:08:26,840 Speaker 1: in their movements in order to maintain the robot's center 143 00:08:26,840 --> 00:08:30,160 Speaker 1: of gravity on the soles of its feet. So every 144 00:08:30,160 --> 00:08:32,560 Speaker 1: time it takes a step, it's it's removing weight from 145 00:08:32,600 --> 00:08:35,760 Speaker 1: one foot. All of its weight is on its other foot. 146 00:08:35,800 --> 00:08:38,040 Speaker 1: It had to be very careful to move its center 147 00:08:38,080 --> 00:08:40,719 Speaker 1: of gravity over the the foot that was still on 148 00:08:40,760 --> 00:08:44,319 Speaker 1: the ground. That also meant that the robot had to 149 00:08:44,360 --> 00:08:48,199 Speaker 1: continuously make adjustments to its balance as it moved the 150 00:08:48,240 --> 00:08:52,360 Speaker 1: free leg forward to take a step. That made taking 151 00:08:52,400 --> 00:08:55,760 Speaker 1: a single step a slow process, sometimes taking as long 152 00:08:55,840 --> 00:08:59,960 Speaker 1: as twenty seconds for one step. Now, clearly that's way 153 00:09:00,120 --> 00:09:02,720 Speaker 1: too slow for a robot to move around in a 154 00:09:02,800 --> 00:09:04,920 Speaker 1: human environment, but it was important for the research and 155 00:09:04,960 --> 00:09:09,480 Speaker 1: development phase. The robot and all future robots i'll cover 156 00:09:09,760 --> 00:09:13,840 Speaker 1: used servo motors for locomotion. Now, those are actuators. They're 157 00:09:13,840 --> 00:09:17,400 Speaker 1: either linear or their rotary, which tells you the type 158 00:09:17,480 --> 00:09:21,520 Speaker 1: of motion they control, and they allow for precise control 159 00:09:21,559 --> 00:09:26,040 Speaker 1: of motion, including velocity and acceleration. A servo motor uses 160 00:09:26,080 --> 00:09:29,040 Speaker 1: position feedback, which means the servo motor has to quote 161 00:09:29,120 --> 00:09:32,360 Speaker 1: unquote no, it's position so that when it receives an 162 00:09:32,400 --> 00:09:35,599 Speaker 1: incoming command to change positions, it can do so accurately. 163 00:09:35,880 --> 00:09:38,800 Speaker 1: So if I translate this into human terms, it's like saying, 164 00:09:39,200 --> 00:09:41,760 Speaker 1: you put your right leg in, you take your right 165 00:09:41,840 --> 00:09:44,679 Speaker 1: leg out, you put your right leg in, and then 166 00:09:44,720 --> 00:09:48,160 Speaker 1: you shake it all about. Humans typically possess a sense 167 00:09:48,200 --> 00:09:51,600 Speaker 1: that we call a proprioception. That's describes how our brains 168 00:09:51,840 --> 00:09:54,679 Speaker 1: sense our bodies, how our brains know where our limbs are. 169 00:09:54,720 --> 00:09:57,240 Speaker 1: This is why if you close your eyes, you can 170 00:09:57,280 --> 00:10:00,000 Speaker 1: touch your finger to your nose, assuming you're not under 171 00:10:00,000 --> 00:10:02,920 Speaker 1: of the influence of something. Because we have appropriate exception, 172 00:10:03,360 --> 00:10:05,040 Speaker 1: we know the location of each of our limbs, and 173 00:10:05,040 --> 00:10:07,240 Speaker 1: therefore we can move those limbs from where they were 174 00:10:07,360 --> 00:10:09,319 Speaker 1: to where they need to be. And that's what it's 175 00:10:09,320 --> 00:10:13,360 Speaker 1: all about. But machines do not innately possess this ability. 176 00:10:13,679 --> 00:10:16,439 Speaker 1: Engineers and computer scientists had to come up with ways 177 00:10:16,480 --> 00:10:19,320 Speaker 1: to mimic it, and servo motors are part of systems 178 00:10:19,320 --> 00:10:22,600 Speaker 1: that keep track of positions so that the overall system 179 00:10:22,640 --> 00:10:25,040 Speaker 1: can behave as directed. So if you tell your robot 180 00:10:25,440 --> 00:10:28,680 Speaker 1: walk forward three steps, the robot can take that command 181 00:10:28,760 --> 00:10:32,400 Speaker 1: and translate it into a series of smaller commands. If 182 00:10:32,400 --> 00:10:35,160 Speaker 1: one leg is already placed forward, then the back leg 183 00:10:35,240 --> 00:10:36,520 Speaker 1: is the one that's going to have to come up 184 00:10:36,520 --> 00:10:39,160 Speaker 1: and take a step for locomotion to begin. For example, 185 00:10:40,000 --> 00:10:43,520 Speaker 1: pretty much every degree of freedom has an associated servo motor. 186 00:10:43,840 --> 00:10:47,319 Speaker 1: So as ASUMO or osimo, I should say, increased in complexity, 187 00:10:47,800 --> 00:10:51,480 Speaker 1: it got more servo motors. Back to EO, The type 188 00:10:51,480 --> 00:10:55,560 Speaker 1: of walking EO could do was called static walking. Before 189 00:10:55,559 --> 00:10:57,800 Speaker 1: it could take a second step, the robot had to 190 00:10:57,800 --> 00:11:00,120 Speaker 1: be certain it's center of weight had shifted over the 191 00:11:00,200 --> 00:11:03,080 Speaker 1: soul of the foot it had just placed down on 192 00:11:03,120 --> 00:11:06,000 Speaker 1: the ground. This was a good start to working out 193 00:11:06,040 --> 00:11:08,360 Speaker 1: the actual mechanics of limb motion, but it was a 194 00:11:08,400 --> 00:11:11,280 Speaker 1: far cry from the way organic critters move around. The 195 00:11:11,280 --> 00:11:13,600 Speaker 1: team was gonna have to do a lot more research, 196 00:11:13,960 --> 00:11:16,600 Speaker 1: and I'll talk about that in just a second, but 197 00:11:16,800 --> 00:11:20,440 Speaker 1: first let's take a quick break to thank our sponsor. 198 00:11:28,040 --> 00:11:31,840 Speaker 1: Engineers got to work analyzing the way humans, animals, and 199 00:11:31,880 --> 00:11:35,920 Speaker 1: even insects walk. They studied hours of videos to understand 200 00:11:36,000 --> 00:11:38,760 Speaker 1: what is going on from a physical or mechanical side 201 00:11:38,760 --> 00:11:42,000 Speaker 1: of things. You know, physiologically, humans don't maintain their center 202 00:11:42,080 --> 00:11:44,240 Speaker 1: of gravity directly over the souls of their feet as 203 00:11:44,240 --> 00:11:46,800 Speaker 1: they walk. The center of gravity as we walk can 204 00:11:46,840 --> 00:11:49,280 Speaker 1: move around quite a bit, so designing a robot that 205 00:11:49,280 --> 00:11:51,760 Speaker 1: could do this too was gonna be pretty tricky. The 206 00:11:51,840 --> 00:11:53,600 Speaker 1: robot would have to take into account a lot of 207 00:11:53,600 --> 00:11:56,560 Speaker 1: different factors that we kind of grasp innately once we 208 00:11:56,640 --> 00:11:59,280 Speaker 1: learn how to walk. That would include stuff like the 209 00:11:59,360 --> 00:12:02,760 Speaker 1: robots fed its momentum, the ground it was walking on, 210 00:12:02,840 --> 00:12:05,440 Speaker 1: whether it was level, whether it was flat, that kind 211 00:12:05,440 --> 00:12:09,600 Speaker 1: of thing. One thing the team did was study human skeletons. 212 00:12:09,640 --> 00:12:12,520 Speaker 1: They noted the location of the joints in the human 213 00:12:12,600 --> 00:12:15,520 Speaker 1: legs and determined that the toes play an important part 214 00:12:15,520 --> 00:12:17,760 Speaker 1: of walking as they helped guide us in the way 215 00:12:17,760 --> 00:12:20,840 Speaker 1: we support our weight from step to step, so their 216 00:12:20,920 --> 00:12:23,400 Speaker 1: robots would need to be able to do a similar thing. 217 00:12:23,720 --> 00:12:25,760 Speaker 1: The robots were also going to need degrees of freedom 218 00:12:25,800 --> 00:12:28,840 Speaker 1: similar to what you could find in a human ankle, knee, 219 00:12:28,840 --> 00:12:32,439 Speaker 1: and hip joint, so the engineers also studied walking humans 220 00:12:32,480 --> 00:12:35,600 Speaker 1: to determine stuff like the range of motion every joint 221 00:12:35,640 --> 00:12:38,120 Speaker 1: should be able to replicate, where the center of gravity 222 00:12:38,160 --> 00:12:41,640 Speaker 1: should be for every leg, how much torque should be 223 00:12:41,640 --> 00:12:44,920 Speaker 1: exerted on leg joints, and also the sensors that would 224 00:12:44,920 --> 00:12:48,559 Speaker 1: be needed to replicate how we humans sense, stuff like 225 00:12:49,120 --> 00:12:51,640 Speaker 1: the speed of motion and the impact of our foot 226 00:12:51,720 --> 00:12:54,200 Speaker 1: hitting the ground. All that would be very important so 227 00:12:54,240 --> 00:12:56,000 Speaker 1: that the robot would be able to walk in a 228 00:12:56,160 --> 00:12:59,680 Speaker 1: stable way and not just hop around or fall off 229 00:12:59,679 --> 00:13:03,400 Speaker 1: of its feet or otherwise have some disaster occur in 230 00:13:04,880 --> 00:13:08,440 Speaker 1: the engineers designed E one. This was the first of 231 00:13:08,600 --> 00:13:12,000 Speaker 1: three robots, the others being E two and E three 232 00:13:12,200 --> 00:13:14,880 Speaker 1: that the engineers designed in an effort to move from 233 00:13:15,000 --> 00:13:19,360 Speaker 1: the static walking model to what they called fast walking 234 00:13:20,120 --> 00:13:25,200 Speaker 1: Like EO. These robots were essentially legs attached to robotic hips, 235 00:13:25,240 --> 00:13:27,240 Speaker 1: and maybe you could argue as also sort of a 236 00:13:27,320 --> 00:13:30,600 Speaker 1: rudimentary torso, but there were no arms, there's no head, 237 00:13:31,040 --> 00:13:34,640 Speaker 1: so I was about it. Uh. Each design looked a 238 00:13:34,679 --> 00:13:37,679 Speaker 1: little bit more sophisticated than its predecessor did, but they 239 00:13:37,679 --> 00:13:42,199 Speaker 1: were all very almost industrial looking kind of robots. They 240 00:13:42,240 --> 00:13:46,360 Speaker 1: would would not take steps as painstakingly slowly as EO did. 241 00:13:46,640 --> 00:13:49,599 Speaker 1: They moved a bit more naturally, which for humans involves 242 00:13:49,600 --> 00:13:53,200 Speaker 1: shifting our weight forward and leaning into a step. It's 243 00:13:53,240 --> 00:13:55,400 Speaker 1: almost like we're about to fall, right like when we 244 00:13:55,440 --> 00:13:57,240 Speaker 1: take a step. It's almost as if we're leaning forward 245 00:13:57,240 --> 00:13:59,440 Speaker 1: and we're gonna fall if we don't catch ourselves, and 246 00:13:59,440 --> 00:14:01,960 Speaker 1: then we move a foot forward and we do catch 247 00:14:01,960 --> 00:14:04,920 Speaker 1: ourselves with our foot, and then we keep leaning forward 248 00:14:04,960 --> 00:14:06,960 Speaker 1: and we catch ourselves again with our foot. So you 249 00:14:06,960 --> 00:14:10,760 Speaker 1: can almost think of walking as consistently nearly falling over 250 00:14:11,360 --> 00:14:12,880 Speaker 1: in a way if you're looking at it from a 251 00:14:13,040 --> 00:14:14,679 Speaker 1: robotics way where you're trying to figure out how to 252 00:14:14,720 --> 00:14:17,880 Speaker 1: design a robot to do a similar thing, So it 253 00:14:17,920 --> 00:14:21,880 Speaker 1: involves a lot of almost falling and catching yourself. While 254 00:14:21,960 --> 00:14:25,720 Speaker 1: the robots in this phase were considered fast walkers, fast 255 00:14:25,840 --> 00:14:28,880 Speaker 1: is a relative term. The E two, for example, the 256 00:14:29,240 --> 00:14:32,880 Speaker 1: middle of the three models, was clocked at a top 257 00:14:32,920 --> 00:14:37,400 Speaker 1: speed of one point two kilometers per hour on flat surfaces, 258 00:14:37,600 --> 00:14:40,440 Speaker 1: which is about three quarters of a mile per hour. 259 00:14:40,800 --> 00:14:45,040 Speaker 1: By comparison, your average humans walking speed is right around 260 00:14:45,080 --> 00:14:48,480 Speaker 1: five kilometers per hour or three point one miles per hour. 261 00:14:48,840 --> 00:14:52,280 Speaker 1: So these robots weren't exactly burning up the track walking 262 00:14:52,280 --> 00:14:55,400 Speaker 1: around the work on E one through E three stretch 263 00:14:55,440 --> 00:14:59,560 Speaker 1: from seven to and then the team moved into the 264 00:14:59,600 --> 00:15:02,600 Speaker 1: next as new wave dance craze. Anyways, it's still rock 265 00:15:02,640 --> 00:15:05,440 Speaker 1: and roll to me. I don't know what happened there, 266 00:15:05,440 --> 00:15:07,200 Speaker 1: it's all on my notes to What I meant to 267 00:15:07,240 --> 00:15:09,560 Speaker 1: say is that they began developing the next generation of 268 00:15:09,600 --> 00:15:12,320 Speaker 1: walking robots that would be E four, E five, and 269 00:15:12,440 --> 00:15:15,920 Speaker 1: E six, and this wasn't an effort to create stabilized walking, 270 00:15:16,400 --> 00:15:18,640 Speaker 1: which meant the engineers wanted to create robots that could 271 00:15:18,680 --> 00:15:22,400 Speaker 1: remain stable while walking on a variety of surfaces, including 272 00:15:22,400 --> 00:15:25,160 Speaker 1: stuff like slopes or stairs, that they would be able 273 00:15:25,160 --> 00:15:27,920 Speaker 1: to adjust their steps and be able to keep their 274 00:15:27,920 --> 00:15:31,360 Speaker 1: weight centered so that they didn't fall over. These robots, 275 00:15:31,400 --> 00:15:33,840 Speaker 1: like the one through E three also looked like an 276 00:15:33,960 --> 00:15:37,280 Speaker 1: armless torso attached to legs, so still didn't look very 277 00:15:37,360 --> 00:15:39,880 Speaker 1: much like Awesomo. At this point, the E four through 278 00:15:39,920 --> 00:15:43,320 Speaker 1: E six models began to incorporate three areas of control 279 00:15:43,440 --> 00:15:47,440 Speaker 1: to achieve stabilized walking, and they are floor reaction control 280 00:15:48,040 --> 00:15:50,400 Speaker 1: that refers to the ability of the robot to absorb 281 00:15:50,560 --> 00:15:53,040 Speaker 1: floor unevenness through the soles of its feet in an 282 00:15:53,040 --> 00:15:55,840 Speaker 1: effort to maintain a firm stance. So the engineers had 283 00:15:55,880 --> 00:15:58,400 Speaker 1: to build sensors into the feet of the robots so 284 00:15:58,480 --> 00:16:00,720 Speaker 1: that the robot could gather information and about the floor 285 00:16:01,280 --> 00:16:03,680 Speaker 1: and then process that information in a way that was 286 00:16:03,720 --> 00:16:06,280 Speaker 1: meaningful and then adjust its stance to give the robot 287 00:16:06,360 --> 00:16:09,840 Speaker 1: the best chance of not pitching over. Next would be 288 00:16:09,880 --> 00:16:13,040 Speaker 1: the target zero moment point control, which is a fancy 289 00:16:13,040 --> 00:16:14,680 Speaker 1: way of saying the robot needs to be able to 290 00:16:14,720 --> 00:16:19,280 Speaker 1: balance itself, so the zero moment point refers to balancing 291 00:16:19,320 --> 00:16:23,640 Speaker 1: different forces in order to maintain posture. Those forces include 292 00:16:23,640 --> 00:16:26,760 Speaker 1: stuff like gravity and walking speed that falls into a 293 00:16:26,840 --> 00:16:31,200 Speaker 1: category called total inertial force. The other force that occurs 294 00:16:31,320 --> 00:16:33,720 Speaker 1: is when the robot's foot connects with the ground. That's 295 00:16:33,720 --> 00:16:36,760 Speaker 1: called the ground reaction force. And you want those two 296 00:16:36,800 --> 00:16:41,920 Speaker 1: forces to cancel one another out in order to maintain posture. Also, 297 00:16:42,120 --> 00:16:44,320 Speaker 1: the robot has to be able to detect when it 298 00:16:44,480 --> 00:16:47,760 Speaker 1: is unable to stand firmly. So first you have to 299 00:16:47,760 --> 00:16:51,240 Speaker 1: incorporate sensors that can detect and imbalance in the robot. 300 00:16:51,720 --> 00:16:53,720 Speaker 1: Then you have to figure out what to do without 301 00:16:54,000 --> 00:16:57,200 Speaker 1: you know, how do you address and imbalance. So with 302 00:16:57,280 --> 00:17:00,680 Speaker 1: these robots, the engineers designed a system which will allow 303 00:17:00,720 --> 00:17:03,560 Speaker 1: the robots to make adjustments to its upper body and 304 00:17:03,600 --> 00:17:06,280 Speaker 1: they would shift their upper body around to act as 305 00:17:06,280 --> 00:17:08,600 Speaker 1: a counter balance. So if it's since it was going 306 00:17:08,640 --> 00:17:11,240 Speaker 1: to fall forward, it might shift its upper body backward 307 00:17:11,480 --> 00:17:14,800 Speaker 1: to counter that action and hopefully remain upright as a result. 308 00:17:15,359 --> 00:17:19,840 Speaker 1: The third area of control was called foot planting location control. 309 00:17:20,160 --> 00:17:24,199 Speaker 1: This system engaged once the ZMP control had activated, and 310 00:17:24,280 --> 00:17:27,359 Speaker 1: this system would determine the length of the robot step 311 00:17:27,520 --> 00:17:30,040 Speaker 1: to catch the robot and make certain it remains upright. 312 00:17:30,320 --> 00:17:32,760 Speaker 1: So it's all about maintaining the proper relationship between the 313 00:17:32,760 --> 00:17:36,119 Speaker 1: position and speed of the robot body with the length 314 00:17:36,240 --> 00:17:39,320 Speaker 1: of the steps it takes. Now, up to this point, 315 00:17:39,359 --> 00:17:42,200 Speaker 1: all the robots have been prototypes to help engineers understand 316 00:17:42,200 --> 00:17:46,440 Speaker 1: the fundamentals necessary that would be needed for basic walking. 317 00:17:46,800 --> 00:17:50,840 Speaker 1: The next stage involved building robots that had arms, hands, 318 00:17:50,840 --> 00:17:53,920 Speaker 1: and a head, and that wasn't just for aesthetics, although 319 00:17:53,960 --> 00:17:56,280 Speaker 1: that did play a part in it, but it's also 320 00:17:56,320 --> 00:17:58,720 Speaker 1: for locomotion. We use our arms and our body in 321 00:17:58,720 --> 00:18:01,199 Speaker 1: our head while we're walking. If the engineers wanted their 322 00:18:01,280 --> 00:18:03,159 Speaker 1: robot to move like a human, they were going to 323 00:18:03,240 --> 00:18:06,040 Speaker 1: need to incorporate those elements as well. Plus, if they 324 00:18:06,040 --> 00:18:08,320 Speaker 1: wanted it to interact in human environments, they wanted it 325 00:18:08,400 --> 00:18:12,520 Speaker 1: to look not terrifying, So giving it arms in the 326 00:18:12,600 --> 00:18:15,280 Speaker 1: head was probably a step in the right direction. From 327 00:18:17,400 --> 00:18:20,159 Speaker 1: the team built another series of three robots. These were 328 00:18:20,200 --> 00:18:23,040 Speaker 1: designated P one, P two, and P three, and all 329 00:18:23,080 --> 00:18:26,120 Speaker 1: of them were humanoid. They were all taller and heavier 330 00:18:26,160 --> 00:18:29,160 Speaker 1: than Osumo would be, and this was when the team 331 00:18:29,200 --> 00:18:32,120 Speaker 1: was still working out the physics and mechanics of humanoid walking, 332 00:18:32,400 --> 00:18:34,680 Speaker 1: so they were more concerned with getting those elements right 333 00:18:34,760 --> 00:18:36,760 Speaker 1: rather than producing a robot that would be suitable for 334 00:18:36,880 --> 00:18:39,439 Speaker 1: human use, so again these were never intended to go 335 00:18:39,480 --> 00:18:43,040 Speaker 1: into the workplace. The P one humanoid robot was one 336 00:18:44,320 --> 00:18:48,040 Speaker 1: five centimeters tall that's about six ft three inches. It's 337 00:18:48,040 --> 00:18:51,720 Speaker 1: a big robot and also weighed into a hundred which 338 00:18:51,720 --> 00:18:54,960 Speaker 1: is about three eighty six pounds, so stats like that. 339 00:18:55,000 --> 00:18:57,440 Speaker 1: It could have wrestled for the w w E. Now, 340 00:18:57,480 --> 00:18:59,639 Speaker 1: clearly that type of robot would be too big and 341 00:18:59,680 --> 00:19:02,520 Speaker 1: heavy and potentially dangerous for a human environment. If it 342 00:19:02,560 --> 00:19:06,640 Speaker 1: lost its balance and fell, it could cause a serious injury. 343 00:19:06,680 --> 00:19:09,480 Speaker 1: But it was one of the first of Honda's robots 344 00:19:09,520 --> 00:19:12,240 Speaker 1: in this line to have arms and sort of claw 345 00:19:12,359 --> 00:19:15,600 Speaker 1: like hands, and engineers worked on coordinating arm and leg 346 00:19:15,680 --> 00:19:18,119 Speaker 1: movements and programmed the robots so it could operate simple 347 00:19:18,160 --> 00:19:20,600 Speaker 1: things like light switches and door knobs. And pick up 348 00:19:20,720 --> 00:19:23,840 Speaker 1: various objects, and coordinating all of that was also another 349 00:19:23,880 --> 00:19:26,960 Speaker 1: big challenge, although to be fair, the walking and running 350 00:19:26,960 --> 00:19:29,159 Speaker 1: was probably the biggest of the challenges they faced at 351 00:19:29,160 --> 00:19:32,359 Speaker 1: that point. The P two robot was the first self regulating, 352 00:19:32,600 --> 00:19:36,280 Speaker 1: two legged humanoid robot walking robot, i should say, and 353 00:19:36,320 --> 00:19:39,879 Speaker 1: it first started strutting its stuff in December n This 354 00:19:40,040 --> 00:19:42,560 Speaker 1: robot was the first to have a computer system incorporated 355 00:19:42,560 --> 00:19:46,040 Speaker 1: directly into the robots design. Previous robots had been controlled 356 00:19:46,080 --> 00:19:49,760 Speaker 1: by computers through a wired connection. This one was completely wireless. 357 00:19:50,200 --> 00:19:52,600 Speaker 1: The P two had a battery, It also had a 358 00:19:52,600 --> 00:19:56,560 Speaker 1: wireless radio, had motor drives, it had its control computer, 359 00:19:56,960 --> 00:20:00,679 Speaker 1: and more systems on board. Operators would say commands to 360 00:20:00,760 --> 00:20:03,320 Speaker 1: the robot through a computer, it would beam the commands over. 361 00:20:03,359 --> 00:20:07,119 Speaker 1: The robot would receive these commands wirelessly, and then they 362 00:20:07,119 --> 00:20:09,439 Speaker 1: would process the commands and the robot would then do 363 00:20:09,480 --> 00:20:13,159 Speaker 1: whatever it was supposed to do, including pushing carts or 364 00:20:13,240 --> 00:20:16,480 Speaker 1: climbing stairs. This robot was a little bit shorter than 365 00:20:16,520 --> 00:20:19,000 Speaker 1: the P one. It measured a hundred a D two centimeters, 366 00:20:19,040 --> 00:20:21,000 Speaker 1: so it was just a hair under six ft tall. 367 00:20:21,840 --> 00:20:26,040 Speaker 1: But putting all that on board processing capability onto the 368 00:20:26,119 --> 00:20:29,399 Speaker 1: robots uh skeleton meant that they added a lot of 369 00:20:29,440 --> 00:20:33,159 Speaker 1: weight to it, so it was a hefty two kilograms 370 00:20:33,280 --> 00:20:36,520 Speaker 1: or nearly four hundred sixty three pounds. I stumbled there 371 00:20:36,520 --> 00:20:38,840 Speaker 1: for a second because in my notes, just a glance 372 00:20:38,880 --> 00:20:42,720 Speaker 1: point the curtain, I wrote two ten kb. That's two bites, 373 00:20:43,440 --> 00:20:48,600 Speaker 1: No two kilograms, silly typo. The P three, which was 374 00:20:48,600 --> 00:20:53,320 Speaker 1: created in September, was much shorter and lighter than its predecessors. 375 00:20:53,680 --> 00:20:56,800 Speaker 1: It stood one sixty centimeters tall that's about five ft 376 00:20:56,840 --> 00:20:59,720 Speaker 1: three inches, and it weighed in a relatively sevelled hundred 377 00:20:59,800 --> 00:21:03,359 Speaker 1: third kilograms or two seven pounds. The team was able 378 00:21:03,400 --> 00:21:05,880 Speaker 1: to decentralize the control system for the P three, which 379 00:21:05,880 --> 00:21:08,280 Speaker 1: helped remove a lot of that weight that it was 380 00:21:08,359 --> 00:21:11,280 Speaker 1: carried around by the P two. Now, those robots helped 381 00:21:11,280 --> 00:21:13,919 Speaker 1: the engineers put together the information they needed to create 382 00:21:13,960 --> 00:21:17,640 Speaker 1: the first robot to be called Asomo. This robot would 383 00:21:17,640 --> 00:21:19,919 Speaker 1: be smaller, it would be lighter, and it would feature 384 00:21:19,920 --> 00:21:21,679 Speaker 1: a design that was meant to make it look friendly 385 00:21:21,720 --> 00:21:24,560 Speaker 1: and playful. So next I'll talk about some of the 386 00:21:24,600 --> 00:21:26,840 Speaker 1: tech that was used to make Osumo work and how 387 00:21:26,880 --> 00:21:30,040 Speaker 1: it became the first humanoid two legged robot. To run. 388 00:21:30,640 --> 00:21:34,120 Speaker 1: But first, let's take another quick break and thank our sponsor. 389 00:21:41,800 --> 00:21:45,439 Speaker 1: Asomo represented a big breakthrough in creating a robot that 390 00:21:45,520 --> 00:21:48,480 Speaker 1: can walk like a human can. For one thing, the 391 00:21:48,520 --> 00:21:52,520 Speaker 1: engineers developed what they called intelligent real time flexible walking 392 00:21:52,880 --> 00:21:56,360 Speaker 1: or I walk for Osomo. So Osmo can shift its 393 00:21:56,359 --> 00:21:58,879 Speaker 1: center of gravity while going through a turn, and that 394 00:21:58,920 --> 00:22:01,159 Speaker 1: allows it to make a turn in a gradual curve. 395 00:22:01,200 --> 00:22:04,639 Speaker 1: It's like it's it's leaning into the curve, which is 396 00:22:04,640 --> 00:22:07,639 Speaker 1: a big deal because earlier robots the only way they 397 00:22:07,680 --> 00:22:09,719 Speaker 1: could turn is they would actually have to stop moving 398 00:22:10,080 --> 00:22:12,399 Speaker 1: and then they would sort of shuffle in place. They 399 00:22:12,400 --> 00:22:16,200 Speaker 1: would stand on one foot, lift their other foot, turn 400 00:22:16,240 --> 00:22:19,439 Speaker 1: it slightly, put their other foot down, lift up their 401 00:22:19,480 --> 00:22:22,280 Speaker 1: first foot, and put it in parallel with the second foot, 402 00:22:22,480 --> 00:22:24,080 Speaker 1: and then they have to keep doing that over and 403 00:22:24,080 --> 00:22:26,520 Speaker 1: over and over again until they gradually we're facing in 404 00:22:26,560 --> 00:22:29,719 Speaker 1: the correct direction. So Osomo was able to do this 405 00:22:29,960 --> 00:22:32,640 Speaker 1: in a much more fluid way, one that was much 406 00:22:32,680 --> 00:22:34,679 Speaker 1: more human, which is very important if you're gonna have 407 00:22:34,720 --> 00:22:38,080 Speaker 1: it moving through human environments. It can actually calculate the 408 00:22:38,080 --> 00:22:41,360 Speaker 1: amount of momentum it will need to get through a 409 00:22:41,480 --> 00:22:44,480 Speaker 1: turn and shift its weight to help compensate for all 410 00:22:44,520 --> 00:22:47,680 Speaker 1: of that. Osomo is also the first humanoid two legged 411 00:22:47,760 --> 00:22:50,280 Speaker 1: robot to run. I mentioned that earlier, and by run, 412 00:22:50,840 --> 00:22:53,600 Speaker 1: I mean Osomo can move forward at a slightly accelerated 413 00:22:53,640 --> 00:22:56,399 Speaker 1: pace and in such a way that at some point 414 00:22:56,480 --> 00:22:59,399 Speaker 1: during its gait both of its feet are off the 415 00:22:59,440 --> 00:23:01,840 Speaker 1: ground at the same time. That's how they define running. 416 00:23:01,840 --> 00:23:05,280 Speaker 1: It's not by super top speed, but rather that at 417 00:23:05,320 --> 00:23:08,560 Speaker 1: some point in the in its stride, both feet are 418 00:23:08,640 --> 00:23:12,399 Speaker 1: actually off the ground. It happens for less than a second, 419 00:23:12,880 --> 00:23:15,439 Speaker 1: but it means that Osumo moves around with a kind 420 00:23:15,440 --> 00:23:17,760 Speaker 1: of like a little hopping motion. Some people have said 421 00:23:17,800 --> 00:23:20,120 Speaker 1: that it looks like whoever is inside the space suit 422 00:23:20,160 --> 00:23:23,119 Speaker 1: needs to go to the bathroom, but it is running. 423 00:23:23,160 --> 00:23:25,920 Speaker 1: And it's actually a big deal for robots because when 424 00:23:25,960 --> 00:23:27,760 Speaker 1: the robot has both of its feet off the ground, 425 00:23:27,800 --> 00:23:30,439 Speaker 1: it no longer has any information about its balance with 426 00:23:30,520 --> 00:23:33,280 Speaker 1: regard to the ground. Right it's actually completely clear of 427 00:23:33,320 --> 00:23:36,080 Speaker 1: the ground for even a split second. Osumo has to 428 00:23:36,119 --> 00:23:39,320 Speaker 1: be able to maintain it's balance and it's weight so 429 00:23:39,359 --> 00:23:42,560 Speaker 1: that doesn't spin when it goes off the ground. It 430 00:23:42,560 --> 00:23:44,240 Speaker 1: has to be able to plan a foot down for 431 00:23:44,280 --> 00:23:47,560 Speaker 1: the next step and be uh, firm enough so it 432 00:23:47,560 --> 00:23:49,840 Speaker 1: can continue it's run. And it has to do all 433 00:23:49,840 --> 00:23:52,600 Speaker 1: of this without having any touch to the ground at all. 434 00:23:53,000 --> 00:23:55,159 Speaker 1: And it's a pretty complicated problem to solve from an 435 00:23:55,200 --> 00:23:57,680 Speaker 1: engineering perspective. How do you get a robot so that 436 00:23:58,080 --> 00:24:01,200 Speaker 1: when it leaves the ground, it maintain aims it's orientation 437 00:24:01,320 --> 00:24:03,840 Speaker 1: so that it doesn't just twist out of the way 438 00:24:03,880 --> 00:24:06,600 Speaker 1: and then come crashing down. One thing I think is 439 00:24:06,640 --> 00:24:11,000 Speaker 1: interesting about Osmo is that it combines pregenerated environmental models 440 00:24:11,119 --> 00:24:14,200 Speaker 1: and the ability to recognize things within an environment. Now 441 00:24:14,240 --> 00:24:17,800 Speaker 1: by that, I mean you can't just PLoP Osomo down 442 00:24:17,920 --> 00:24:20,560 Speaker 1: into a brand new space and expect the robot to 443 00:24:20,640 --> 00:24:24,920 Speaker 1: seamlessly navigate through various interactions. Osumo is not autonomous. It's 444 00:24:24,960 --> 00:24:29,720 Speaker 1: not an autonomous robot. It cannot operate all on its own. Instead, 445 00:24:29,720 --> 00:24:33,400 Speaker 1: it relies on a combination of programming, an operator who 446 00:24:33,400 --> 00:24:36,280 Speaker 1: can run Osomo from a computer sort of like a 447 00:24:36,320 --> 00:24:40,439 Speaker 1: remote controlled vehicle or both, in addition to its own ability, 448 00:24:40,560 --> 00:24:43,240 Speaker 1: so it can respond to things like verbal commands and 449 00:24:43,359 --> 00:24:46,520 Speaker 1: gesture commands. So it could do all those, but it 450 00:24:46,560 --> 00:24:49,440 Speaker 1: can't do stuff on its own. This does not mean 451 00:24:49,480 --> 00:24:52,239 Speaker 1: it's not an impressive technology. This is very impressive. If 452 00:24:52,240 --> 00:24:54,280 Speaker 1: you were to take control of Osumo and make it 453 00:24:54,320 --> 00:24:57,720 Speaker 1: walk forward, it could actually adjust its own steps to 454 00:24:57,760 --> 00:25:00,640 Speaker 1: meet your commands, which might include move having a leg 455 00:25:00,680 --> 00:25:03,560 Speaker 1: a certain way to avoid an obstacle while still navigating 456 00:25:03,640 --> 00:25:06,920 Speaker 1: to the location you guide it to. So, in other words, 457 00:25:06,920 --> 00:25:09,960 Speaker 1: Osmo can make small decisions like where it needs to 458 00:25:10,000 --> 00:25:14,399 Speaker 1: move body parts in order to continue to fulfill whatever 459 00:25:14,400 --> 00:25:16,760 Speaker 1: the command was that was given to it. But it 460 00:25:16,880 --> 00:25:20,200 Speaker 1: can't decide to do something all on its own. It's 461 00:25:20,200 --> 00:25:24,840 Speaker 1: making these smaller decisions about minute stuff within the context 462 00:25:24,840 --> 00:25:28,119 Speaker 1: of a larger command. To do things like navigate stairways. 463 00:25:28,160 --> 00:25:31,640 Speaker 1: You typically would program Osumo to have a working knowledge 464 00:25:31,720 --> 00:25:35,160 Speaker 1: of the environment before having it moved through the room. 465 00:25:35,200 --> 00:25:38,400 Speaker 1: That helps Osumo's navigational systems as well as it means 466 00:25:38,400 --> 00:25:41,359 Speaker 1: the robot knows where things are supposed to be and 467 00:25:41,440 --> 00:25:45,040 Speaker 1: can compare where things really are against that model in 468 00:25:45,119 --> 00:25:47,600 Speaker 1: order to make decisions. So, for example, let's say you 469 00:25:47,640 --> 00:25:51,199 Speaker 1: have Osumo walking around a hotel lobby greeting guests, and 470 00:25:51,320 --> 00:25:53,119 Speaker 1: one morning, some guests have moved a few of the 471 00:25:53,200 --> 00:25:55,919 Speaker 1: chairs so that they can sit together. But the chairs 472 00:25:56,160 --> 00:25:58,360 Speaker 1: are no longer where they used to be in Osomo's 473 00:25:58,400 --> 00:26:02,600 Speaker 1: little preplanned man model of the hotel lobby, and if 474 00:26:02,640 --> 00:26:05,760 Speaker 1: it only worked off of that pre program model, then 475 00:26:05,840 --> 00:26:08,359 Speaker 1: it would try and walk around chairs that weren't actually there, 476 00:26:08,440 --> 00:26:10,520 Speaker 1: or it would bump into chairs that had been moved. 477 00:26:10,760 --> 00:26:13,000 Speaker 1: But Osomo can also use its sensors, which I'm going 478 00:26:13,040 --> 00:26:15,440 Speaker 1: to talk about in the second, to locate obstacles and 479 00:26:15,560 --> 00:26:18,240 Speaker 1: navigate around them within the context of the room, and 480 00:26:18,320 --> 00:26:20,159 Speaker 1: knows how far out the way it can go in 481 00:26:20,200 --> 00:26:23,159 Speaker 1: any given direction. To plot out its path, typically you 482 00:26:23,160 --> 00:26:25,880 Speaker 1: would also mark a room with certain types of markers 483 00:26:25,880 --> 00:26:29,440 Speaker 1: that Osomo can detect, and that gives Osumo more precise 484 00:26:29,760 --> 00:26:32,680 Speaker 1: information about where it is in context of other things 485 00:26:32,760 --> 00:26:36,159 Speaker 1: inside the room. Generally speaking, Osumo tries to take the 486 00:26:36,160 --> 00:26:39,000 Speaker 1: pathway that will require the fewest number of steps or 487 00:26:39,040 --> 00:26:41,360 Speaker 1: the least amount of work. And it's not that Osmo 488 00:26:41,480 --> 00:26:44,000 Speaker 1: is lazy, but rather it has a fifty one point 489 00:26:44,080 --> 00:26:46,919 Speaker 1: eight volt lithium ion battery and that's good for about 490 00:26:47,000 --> 00:26:50,000 Speaker 1: one hour of operation, so it takes about three hours 491 00:26:50,000 --> 00:26:53,160 Speaker 1: to recharge it. So Osumo wants to make the most 492 00:26:53,200 --> 00:26:56,200 Speaker 1: out of its brief moments of wakefulness. So note to self, 493 00:26:56,240 --> 00:26:57,920 Speaker 1: if I ever get an Osumo, make sure I also 494 00:26:57,920 --> 00:26:59,800 Speaker 1: pick up a couple of extra batteries, so I always 495 00:26:59,840 --> 00:27:02,639 Speaker 1: have one charged. Those batteries, by the way, account for 496 00:27:02,720 --> 00:27:06,359 Speaker 1: nearly six kilograms of Ozumo's weight, or about thirteen pounds. 497 00:27:06,880 --> 00:27:09,040 Speaker 1: So let's talk about some of the sensors Ozomo has 498 00:27:09,080 --> 00:27:12,439 Speaker 1: to help it navigate an environment. In order to maintain 499 00:27:12,440 --> 00:27:15,760 Speaker 1: its balance, Osmo has a gyroscope, and gyroscopes are devices 500 00:27:15,800 --> 00:27:18,679 Speaker 1: that measure or maintain rotational motion, and they have some 501 00:27:18,760 --> 00:27:22,040 Speaker 1: interesting properties that, on a casual glance, seemed to defy 502 00:27:22,160 --> 00:27:25,200 Speaker 1: common sense. A gyroscope works on the principle of angular 503 00:27:25,280 --> 00:27:28,520 Speaker 1: velocity around an axis of rotation. So imagine you've got 504 00:27:28,560 --> 00:27:31,679 Speaker 1: a bike wheel and you have an axle, and the 505 00:27:31,680 --> 00:27:34,320 Speaker 1: bike wheel can rotate freely around the axle. So you're 506 00:27:34,320 --> 00:27:36,680 Speaker 1: holding it up right with one hand on either side 507 00:27:36,720 --> 00:27:38,480 Speaker 1: of this axle. You're holding it up in front of you, 508 00:27:38,960 --> 00:27:42,240 Speaker 1: and your friend gives the wheel a really good spin. 509 00:27:42,359 --> 00:27:45,960 Speaker 1: You're holding it vertically. You then attach a string to 510 00:27:46,040 --> 00:27:48,520 Speaker 1: one side of the axle of this upright wheel, and 511 00:27:48,600 --> 00:27:51,080 Speaker 1: your hand is still on the other side. Now, what 512 00:27:51,119 --> 00:27:53,760 Speaker 1: would happen if you were to let go of the axle? 513 00:27:54,480 --> 00:27:56,119 Speaker 1: And you might think that, well, the wheel is just 514 00:27:56,119 --> 00:27:59,479 Speaker 1: gonna flop over horizontally, but it doesn't. The wheel, as 515 00:27:59,520 --> 00:28:02,440 Speaker 1: long as it's rotating rounds axle will remain upright, and 516 00:28:02,520 --> 00:28:07,760 Speaker 1: that effect is called precession. A spinning gyroscope will uh 517 00:28:08,000 --> 00:28:11,000 Speaker 1: is stable, and once the gyroscope is spinning, it has 518 00:28:11,040 --> 00:28:13,800 Speaker 1: a tendency to remain in the same orientation, and any 519 00:28:13,840 --> 00:28:17,040 Speaker 1: force applied to change the orientation of the spin axis 520 00:28:17,480 --> 00:28:20,480 Speaker 1: is met with a resistive force. So let's say you're 521 00:28:20,480 --> 00:28:22,960 Speaker 1: still holding both sides of the axle of this bicycle 522 00:28:22,960 --> 00:28:24,920 Speaker 1: wheel and your friend gives it a really good spin. 523 00:28:24,960 --> 00:28:27,600 Speaker 1: You're holding it vertically, and then you try to turn 524 00:28:27,640 --> 00:28:31,760 Speaker 1: the wheel horizontally while it's still spinning. You would actually 525 00:28:31,760 --> 00:28:34,439 Speaker 1: feel resistance to this. This is what actually makes it 526 00:28:34,440 --> 00:28:36,520 Speaker 1: easy to ride a bicycle. Once you really get going, 527 00:28:36,560 --> 00:28:39,120 Speaker 1: the rotation of the wheels along their spin access will 528 00:28:39,200 --> 00:28:43,120 Speaker 1: help keep you upright. In addition to the gyroscope, Osomo 529 00:28:43,200 --> 00:28:47,080 Speaker 1: has an accelerometer which measures acceleration. Acceleration refers to a 530 00:28:47,160 --> 00:28:49,959 Speaker 1: change in velocity, so that could be either in speed 531 00:28:50,080 --> 00:28:54,280 Speaker 1: or direction or both. Osamo also has a six as 532 00:28:54,360 --> 00:28:58,160 Speaker 1: to six access force sensors that's for detecting the direction 533 00:28:58,200 --> 00:29:02,040 Speaker 1: and amount of force that the hand ends encounter. Uh. 534 00:29:02,240 --> 00:29:04,440 Speaker 1: They also actually have two more for the feet. I 535 00:29:04,440 --> 00:29:07,160 Speaker 1: forgot about that, So technically it's got four six access 536 00:29:07,160 --> 00:29:11,240 Speaker 1: force sensors. Osmo also has cameras to provide a stereoscopic 537 00:29:11,320 --> 00:29:14,400 Speaker 1: view of its surroundings that allows Ozomo to judge the 538 00:29:14,480 --> 00:29:17,520 Speaker 1: depth of a scene and determine which objects are close 539 00:29:17,560 --> 00:29:19,720 Speaker 1: to it, which one versus which ones are further away. 540 00:29:20,200 --> 00:29:23,680 Speaker 1: The systems on board Osmo also have facial recognition capabilities 541 00:29:23,680 --> 00:29:26,240 Speaker 1: and allow the robot to recognize objects that are in motion, 542 00:29:26,600 --> 00:29:29,520 Speaker 1: and it can also respond to gesture commands like waving. 543 00:29:29,600 --> 00:29:31,400 Speaker 1: So if you wave at Osmo, it can wave back 544 00:29:31,440 --> 00:29:34,720 Speaker 1: at you and recognizes that as a gesture command. Osmo 545 00:29:34,840 --> 00:29:39,480 Speaker 1: also has environment identifying sensors, including ultrasonic sensors that can 546 00:29:39,560 --> 00:29:42,000 Speaker 1: detect obstacles that are up to three meters ahead of it, 547 00:29:42,400 --> 00:29:45,880 Speaker 1: including glass. Because you know, it works off echolocation, it's 548 00:29:45,920 --> 00:29:49,960 Speaker 1: not optical. They are also laser sensors and infrared sensors 549 00:29:50,000 --> 00:29:53,840 Speaker 1: to help the robot detect the ground. Osimo frequently navigates 550 00:29:53,840 --> 00:29:56,200 Speaker 1: by referencing those markings on the ground I mentioned before. 551 00:29:56,200 --> 00:29:59,480 Speaker 1: The infrared sensor can detect those and that tells Ozomo 552 00:29:59,560 --> 00:30:01,840 Speaker 1: it's in the place, or that it needs to move 553 00:30:01,880 --> 00:30:03,800 Speaker 1: in a certain area that may not be in the 554 00:30:03,920 --> 00:30:07,200 Speaker 1: location it thought it was. The ground sensor is located 555 00:30:07,200 --> 00:30:10,560 Speaker 1: at the base of Osomo's torso, and there are ultrasonic 556 00:30:10,600 --> 00:30:12,840 Speaker 1: sensors in an array both on the front and the 557 00:30:12,880 --> 00:30:16,680 Speaker 1: backside of Osumo. Osmo also has three microphones that allow 558 00:30:16,720 --> 00:30:19,320 Speaker 1: it to detect noise and determine the origin of the noise, 559 00:30:19,360 --> 00:30:22,480 Speaker 1: as well as to receive voice commands. And for a while, 560 00:30:22,680 --> 00:30:26,360 Speaker 1: you could meet Osumo at Disneyland Interventions in an attraction 561 00:30:26,440 --> 00:30:30,440 Speaker 1: titled Say Hello to Honda's Asumo, and I did. I 562 00:30:30,480 --> 00:30:33,240 Speaker 1: got to meet Osmo. I just went to watch the 563 00:30:33,400 --> 00:30:35,400 Speaker 1: live show and that's all I was gonna do, is was 564 00:30:35,360 --> 00:30:37,640 Speaker 1: was just watch it and walk away. But a cast 565 00:30:37,680 --> 00:30:39,400 Speaker 1: member was talking to me and I mentioned that I 566 00:30:39,440 --> 00:30:42,840 Speaker 1: had written an article about how Ozomo works, and they said, 567 00:30:42,880 --> 00:30:45,040 Speaker 1: would you like to meet Osumo? And I said yes, 568 00:30:45,040 --> 00:30:47,800 Speaker 1: I would please, and so they introduced me and I 569 00:30:47,840 --> 00:30:50,240 Speaker 1: got to meet Osumo. Osmo, by the way, you can 570 00:30:50,240 --> 00:30:54,520 Speaker 1: still be seen at Disneyland's Autopia, but Osumo itself is 571 00:30:54,600 --> 00:30:57,520 Speaker 1: no longer in production. Instead, Honda plans to incorporate the 572 00:30:57,520 --> 00:31:01,280 Speaker 1: technology and discoveries made during developing the robot into stuff 573 00:31:01,320 --> 00:31:04,960 Speaker 1: like the unicub device or the walking assist Harness and 574 00:31:05,000 --> 00:31:09,360 Speaker 1: Honda introduced four new helper robot concepts at CE, and 575 00:31:09,400 --> 00:31:13,200 Speaker 1: all of them seem to incorporate some elements of Osmo's 576 00:31:13,240 --> 00:31:16,280 Speaker 1: design in them. So Osumo lives on in a way, 577 00:31:16,680 --> 00:31:19,440 Speaker 1: but in new products, though the form factor of the 578 00:31:19,480 --> 00:31:22,440 Speaker 1: childlike space man appears to be a thing in the past. 579 00:31:22,520 --> 00:31:26,360 Speaker 1: So farewell, Osumo. You're really good to me. I appreciate 580 00:31:26,440 --> 00:31:28,600 Speaker 1: you giving me my first writing assignment here at how 581 00:31:28,640 --> 00:31:32,040 Speaker 1: Stuff Works, and thank you Cat for suggesting the topic. 582 00:31:32,080 --> 00:31:33,479 Speaker 1: It was a lot of fun to go back and 583 00:31:33,520 --> 00:31:37,160 Speaker 1: revisit this and watch videos of Osimo running around and 584 00:31:37,480 --> 00:31:40,400 Speaker 1: sometimes toppling over. It's sad to see, but you know, 585 00:31:40,560 --> 00:31:42,880 Speaker 1: technology doesn't always work the way you wanted to, and 586 00:31:42,960 --> 00:31:45,240 Speaker 1: it's good to remind ourselves of that from time to time. 587 00:31:45,560 --> 00:31:48,200 Speaker 1: If you have any suggestions for topics I should tackle 588 00:31:48,240 --> 00:31:50,760 Speaker 1: in future episodes of tech Stuff, write me and let 589 00:31:50,800 --> 00:31:53,600 Speaker 1: me know. The email addresses tech Stuff at how stuff 590 00:31:53,600 --> 00:31:55,520 Speaker 1: Works dot com, or you can drop me a line 591 00:31:55,520 --> 00:31:57,840 Speaker 1: on Facebook or Twitter. The handle of both of those 592 00:31:57,920 --> 00:32:01,160 Speaker 1: is tech Stuff H. S W. Don't forget to follow 593 00:32:01,160 --> 00:32:03,719 Speaker 1: the show over on Instagram and I'll talk to you 594 00:32:03,760 --> 00:32:12,360 Speaker 1: again really soon. For more on this and thousands of 595 00:32:12,360 --> 00:32:24,520 Speaker 1: other topics. Is that how stuff Works dot com