WEBVTT - What is the future of robotic hands? 0:00:00.160 --> 0:00:07.200 Brought to you by Toyota. Let's go places. Welcome to 0:00:07.360 --> 0:00:15.840 Forward Thinking. Hey they're I'm welcome to Forward Thinking, the 0:00:15.920 --> 0:00:18.759 podcast that looks at the future and says caused the 0:00:18.880 --> 0:00:21.880 be Bop Stork was about to arrive, Mama gave birth 0:00:21.960 --> 0:00:26.880 to the hand Jive. I'm Jonathan Stricklin and I'm Joe McCormick. 0:00:26.920 --> 0:00:29.280 And here on this show where we often talk about 0:00:29.440 --> 0:00:34.160 robotics and the future of machines replacing us, helping us, 0:00:34.200 --> 0:00:39.120 mostly replacing us. Uh yeah, we sometimes have to take 0:00:39.240 --> 0:00:43.680 stock of the things that still fall under the category 0:00:43.880 --> 0:00:47.519 of humans can definitely do them better than machines. Right 0:00:47.600 --> 0:00:50.360 until recently, one of those things was playing the game 0:00:50.440 --> 0:00:56.440 Go not anymore anymore joined chess and checkers in the 0:00:56.480 --> 0:01:00.000 field tik tac toe. Yeah, the computers are so totally 0:01:00.240 --> 0:01:03.720 better than the hermon Nuclear war Yeah probably. Yeah, Well, 0:01:03.760 --> 0:01:07.000 I mean in the movie War Games, it's it's definitive. 0:01:07.319 --> 0:01:08.920 You know, when I said tick tek to, I think 0:01:08.959 --> 0:01:11.560 what I meant was connect for now that is a 0:01:11.600 --> 0:01:14.000 solved game that computers can always beat us in. But 0:01:14.040 --> 0:01:16.400 I'm pretty sure they can at least force a draw 0:01:16.600 --> 0:01:21.920 or tecto. But anyway, Yeah, so always doing better than us. 0:01:21.959 --> 0:01:24.280 Of course, they can drive cars better than us. That's 0:01:24.319 --> 0:01:28.759 been clearly demonstrated now with one crash out of one 0:01:28.800 --> 0:01:31.280 crash caused by a robotic car out of all the 0:01:31.280 --> 0:01:34.919 ones that have actually all the miles and miles millions 0:01:34.920 --> 0:01:37.360 of miles driven by robotic cars. I think even that 0:01:37.400 --> 0:01:39.320 one's kind of question. Well, you're talking about where it went, 0:01:39.400 --> 0:01:41.160 where it got hit by the bus. I was trying 0:01:41.200 --> 0:01:44.440 to merge Google. It's Googled themselves said, you know that 0:01:44.480 --> 0:01:47.160 they accepted responsibility for that, but partial at fall it 0:01:47.200 --> 0:01:49.960 was a little bit of a backheaded acceptance because they said, like, well, 0:01:50.000 --> 0:01:53.080 we just anticipated that the bus driver would slow down there, 0:01:53.120 --> 0:01:57.480 Like well if the bus had been not a shirt, right, Yeah, 0:01:57.600 --> 0:02:00.680 so there's that. You know. One thing I remember coming 0:02:00.680 --> 0:02:02.480 across a couple of years ago, I think is that 0:02:02.560 --> 0:02:05.920 even now now, computers are even better at reading micro 0:02:06.040 --> 0:02:10.520 expressions on human faces than humans are. Yeah, so you know, 0:02:10.560 --> 0:02:13.720 they can tell when someone's disgusted or bored or whatever 0:02:13.840 --> 0:02:17.760 better than we can. Yeah. I definitely I have encountered 0:02:17.840 --> 0:02:21.160 humans that I know clearly lacked some of that ability 0:02:21.200 --> 0:02:23.520 where I'm like, I can't get out of this conversation. 0:02:23.600 --> 0:02:26.160 No matter what I do, you have to actually tell 0:02:26.240 --> 0:02:28.840 them explicitly stopped talking to me. Yeah, I've also been 0:02:28.840 --> 0:02:31.600 guilty of doing that. I just carry around plasmic grenades 0:02:31.600 --> 0:02:33.640 and what I'm done with a conversation and just stick 0:02:33.680 --> 0:02:36.400 it right to their face. I was just going with 0:02:36.400 --> 0:02:39.960 the ninja smoke pomps. Just then run. You know, for 0:02:40.000 --> 0:02:42.640 a while the robots were pretty bad at walking on 0:02:42.760 --> 0:02:45.200 uneven surfaces, but they're even closed in the gap on 0:02:45.320 --> 0:02:49.080 that those Boston Dynamic spots. Although if you watch that 0:02:49.440 --> 0:02:53.560 DARPA Grand Challenge where it had the sequence of tasks 0:02:53.600 --> 0:02:56.200 that the robots had to be able to complete, we 0:02:56.240 --> 0:03:01.240 saw that robots still find certain tasks really challenging, especially bipedales, 0:03:01.800 --> 0:03:04.400 like like opening a door and walking through it. Well, 0:03:04.800 --> 0:03:07.880 bipedal is a terrible system for anyone to use. Even 0:03:07.919 --> 0:03:10.840 our own spines are really meant to be like Crawley spines. No, 0:03:10.919 --> 0:03:13.360 they're meant to sit in front of a computer. I 0:03:13.400 --> 0:03:16.160 sure hope so, because that's kind of what I'm doing. Well, 0:03:16.600 --> 0:03:19.520 it's true. What one of the biggest problems left for 0:03:19.600 --> 0:03:26.440 machines is dexterity diverse object manipulation on the small scale, 0:03:27.639 --> 0:03:30.880 So you deal with this skill every day. You use 0:03:30.919 --> 0:03:35.160 it all the time handling various objects, especially delicate objects 0:03:35.160 --> 0:03:39.840 with precision. One definition that comes from some researchers that 0:03:39.840 --> 0:03:42.680 we're going to cite later in the episode puts dexterity 0:03:42.760 --> 0:03:48.360 as quote many independently controlled degrees of freedom, speed, strength, 0:03:48.520 --> 0:03:52.640 and compliance. I think that pretty well summarizes it. It's 0:03:52.680 --> 0:03:54.680 what you do with your hands to get all the 0:03:54.720 --> 0:03:58.960 different jobs you need to do done without crushing everything 0:03:59.040 --> 0:04:03.360 you touch in the pro or injuring yourself. Yeah. For example, Uh, 0:04:03.440 --> 0:04:06.080 we can use our hands to manipulate all sorts of 0:04:06.120 --> 0:04:09.360 different objects pretty readily, even if we haven't come into 0:04:09.360 --> 0:04:13.320 contact with them before. Just through the experimentation of touching 0:04:13.320 --> 0:04:15.080 the object and then you get a feel for what 0:04:15.160 --> 0:04:17.200 the object is like, you kind of I get an 0:04:17.240 --> 0:04:20.400 idea immediately. How much pressure should you use when picking 0:04:20.480 --> 0:04:22.640 up the object. When it's something squishy like a banana, 0:04:22.680 --> 0:04:24.520 you know not to squeeze so hard that it squirts 0:04:24.520 --> 0:04:28.320 out both ends. But also if it's like a prickly, 0:04:28.800 --> 0:04:32.440 uh sweet gumball or something, you realize those points, like 0:04:32.480 --> 0:04:34.240 if you use too much pressure, you're going to hurt 0:04:34.279 --> 0:04:37.039 yourself and you know to be gentle with it. And 0:04:37.080 --> 0:04:40.200 you also get a feel immediately how heavy that object is, 0:04:40.240 --> 0:04:42.279 so you know how much force to use when lifting. 0:04:42.600 --> 0:04:45.400 These are things that are intuitive to us, but robots 0:04:45.520 --> 0:04:50.760 don't have intuition, except they kind of can if you 0:04:50.800 --> 0:04:53.000 go with machine learning. But we'll get to that. Yes, 0:04:53.080 --> 0:04:55.960 I think that as well, said Jonathan. So think about 0:04:56.000 --> 0:04:58.320 it this way. When you pick up a banana and 0:04:58.360 --> 0:05:01.160 then peel that banana, and then pick up a knife 0:05:01.160 --> 0:05:03.719 with your other hand, and then slice that banana over 0:05:03.760 --> 0:05:06.400 a bowl of cereal. This may be a simple action 0:05:06.440 --> 0:05:10.480 to you, but this is horrifying warlock magic to robots. Right. 0:05:10.640 --> 0:05:13.800 We we have nothing even close to being able to 0:05:13.880 --> 0:05:18.040 do this unless there's a robot that has been painstakingly 0:05:18.080 --> 0:05:22.800 explicitly programmed specifically for this task, which I haven't heard of. 0:05:22.839 --> 0:05:25.520 If there is, and if there is, it makes you 0:05:25.560 --> 0:05:28.720 wonder what better things people could have done with their 0:05:28.760 --> 0:05:31.640 time rather than training a robot to slice a banana 0:05:31.680 --> 0:05:36.359 into cereals robot. But yes, that's true. That's that's a 0:05:36.440 --> 0:05:39.160 kind of good example of the the idea of explicit 0:05:39.160 --> 0:05:44.360 programming versus a sort of generalist adaptable programming. But yet 0:05:44.520 --> 0:05:49.280 gentle variable, dextrous and deft control of objects in the hands. 0:05:49.440 --> 0:05:52.880 That's just something that robots are not at yet. And 0:05:52.920 --> 0:05:55.839 another thing is that watching video of robots trying to 0:05:55.880 --> 0:05:59.960 achieve this is often really funny. Now. I watched one 0:06:00.000 --> 0:06:03.880 recently where they weren't very um advanced robotic hands. They 0:06:03.920 --> 0:06:07.960 were actually pretty simple clamps. Clamps there, clamps. Yeah, it 0:06:08.040 --> 0:06:10.200 was essentially it was a two pronged kind of hand, right, 0:06:10.240 --> 0:06:13.159 like like just just clamp. Yeah, and uh, but it 0:06:13.200 --> 0:06:17.120 had a box full of different objects, and the robotic 0:06:17.240 --> 0:06:20.480 arms or the hands rather, we're quote unquote learning how 0:06:20.520 --> 0:06:22.520 to pick up those different objects. And they would attempt 0:06:22.600 --> 0:06:25.560 to pick up an object and start to learn from uh, 0:06:26.000 --> 0:06:30.200 trial and error how that was not working properly or 0:06:30.240 --> 0:06:32.360 what did work properly, and it would try to replicate 0:06:32.400 --> 0:06:34.480 that in the future attempt. And it was a box 0:06:34.520 --> 0:06:37.440 full of tons of different stuff, like, uh, staplers, and 0:06:37.480 --> 0:06:40.360 it was mostly office equipment type stuff staplers and and 0:06:40.360 --> 0:06:42.520 and tape and that kind of thing. And they were 0:06:42.560 --> 0:06:45.640 supposed to pick up the staplers and so for a while, 0:06:46.040 --> 0:06:48.719 you know, they're just kind of fumbling around trying to 0:06:48.760 --> 0:06:50.760 find the best way to get a grip on the 0:06:51.080 --> 0:06:53.479 staplers and eventually started to figure it out pretty quickly. 0:06:53.760 --> 0:06:57.960 Then they started to go with more advanced approaches, uh, 0:06:58.040 --> 0:07:00.640 which the hands were learning on their own, direct developing 0:07:00.640 --> 0:07:03.599 these techniques on their own where they would knock an 0:07:03.600 --> 0:07:06.719 object that they didn't want out of the way in 0:07:06.839 --> 0:07:08.240 order to be able to get a grip on the 0:07:08.240 --> 0:07:10.800 object they did want. So it's kind of like you 0:07:10.800 --> 0:07:13.280 know a human where you see a pile of stuff 0:07:13.280 --> 0:07:14.880 and the thing you want is under something else. You 0:07:14.920 --> 0:07:16.600 just kind of move the thing on top in order 0:07:16.600 --> 0:07:18.360 to get it to the thing that you actually want. 0:07:19.000 --> 0:07:22.760 And it was fascinating to watch. It was very funny 0:07:22.800 --> 0:07:27.000 because at the beginning it was just chaos. Right, It's 0:07:27.040 --> 0:07:29.600 just robot hands going everywhere in this box of stuff. 0:07:29.640 --> 0:07:32.480 It's like the hands in that office simulator game talk 0:07:32.600 --> 0:07:37.560 about just throwing things randomly in different directions. Exactly. Yeah, So, okay, 0:07:37.560 --> 0:07:43.600 now we're we're criticizing the current dexterous nous, dexterousness, dexterity, 0:07:44.080 --> 0:07:47.600 dexterity of robot hands. But you might be thinking, wait 0:07:47.640 --> 0:07:51.840 a second, now, I thought that there were industrial robot 0:07:51.920 --> 0:07:54.680 arms and so forth, all kinds of robotic effectors that 0:07:54.760 --> 0:07:59.600 carry out very precise jobs like machining, assembly, and manipulation. 0:07:59.680 --> 0:08:03.320 Job and they do, yeah, they certainly do. But these 0:08:03.440 --> 0:08:07.520 robots have been explicitly programmed for a very specific job 0:08:07.600 --> 0:08:11.080 that they do over and over again. It's worth the 0:08:11.120 --> 0:08:14.720 time for somebody to figure out exactly what motions they 0:08:14.800 --> 0:08:17.400 need to make and teach them how to do that, 0:08:17.880 --> 0:08:20.080 because they're going to be doing the same thing maybe 0:08:20.120 --> 0:08:21.840 for the rest of their lives or at least for 0:08:21.880 --> 0:08:25.280 one long production run. Right. It's it's the idea of 0:08:25.280 --> 0:08:28.600 of using a robot instead of a human for those 0:08:28.640 --> 0:08:34.959 those uh jobs that have really tedious, replicated work that 0:08:35.080 --> 0:08:37.880 you have to do adding a single part to a 0:08:37.960 --> 0:08:41.959 larger thing over and over and over, right, or welding 0:08:42.880 --> 0:08:45.480 a door to a car frame, that kind of stuff 0:08:46.240 --> 0:08:48.760 where you know it's something that would be dull or 0:08:48.840 --> 0:08:51.920 dangerous or dirty. That's the three DS that we often 0:08:51.920 --> 0:08:53.800 get for the kind of jobs that robots are best 0:08:53.800 --> 0:08:57.480 suited for. And as long as it is a task 0:08:57.520 --> 0:09:00.400 that is going to be exactly the same time after time, 0:09:01.040 --> 0:09:04.480 you don't have to worry as much about the um 0:09:05.320 --> 0:09:07.840 the general dexterity. You just have to you just have 0:09:07.920 --> 0:09:11.200 to make it really good at specific movements. Yeah, one 0:09:11.400 --> 0:09:14.560 thing that it does. And so industrial robot might be 0:09:14.600 --> 0:09:17.200 able to, for example, place a ball bearing in a 0:09:17.280 --> 0:09:20.400 machine slot with precision over and over again. But if 0:09:20.440 --> 0:09:22.720 you try to get that same robot arm with its 0:09:22.720 --> 0:09:26.080 same end defect or to open up a ketchup packet 0:09:26.160 --> 0:09:29.680 or something, well, let's see how well it does. End defector, 0:09:29.720 --> 0:09:32.160 by the way, is the industry term for thing at 0:09:32.200 --> 0:09:34.839 the end of a robot arm. Yeah, it's the part 0:09:34.880 --> 0:09:37.280 of the robot that the robot uses to interact with 0:09:37.360 --> 0:09:39.720 the world. So this could be like a welding tor 0:09:39.880 --> 0:09:43.720 to drill, a puncturing style, as a suction cup, a claw, 0:09:43.840 --> 0:09:47.080 a clamp, or maybe a hand, maybe a humanoid hand 0:09:47.080 --> 0:09:51.040 with five fingers. But hey, you might also be thinking, now, 0:09:51.120 --> 0:09:54.360 didn't you guys talk in the past about robotic surgeons 0:09:54.440 --> 0:09:59.120 and how these robotic these surgical robots could give surgeons 0:09:59.200 --> 0:10:03.480 really incredible precision in laparoscopic procedures reaching into the body, 0:10:03.640 --> 0:10:07.839 And yeah, that is also true with precision cutting sutures 0:10:07.920 --> 0:10:11.040 and a tiny, tiny little movements. There is precision in 0:10:11.040 --> 0:10:16.080 these robotic tools. But a couple of things. Number One, 0:10:16.400 --> 0:10:20.880 they're controlled by humans, So What they do is they 0:10:20.920 --> 0:10:25.679 translate large movements of a controller's hands into very tiny, 0:10:25.800 --> 0:10:29.520 precise movements in reality. And this is something that anyone 0:10:29.559 --> 0:10:32.880 who has used a computer in the last thirty years 0:10:32.920 --> 0:10:35.520 has had experience with, because your mouse does the same thing. 0:10:36.080 --> 0:10:39.760 When you move a mouse, you're moving it, uh a 0:10:39.880 --> 0:10:42.400 small Well, it's it's kind of the opposite side. That's 0:10:42.440 --> 0:10:45.959 the same principle, but in opposite right in scale of me. Exactly, 0:10:46.000 --> 0:10:48.920 you move the mouse a little bit and the the 0:10:49.080 --> 0:10:51.680 arrow will move a greater amount on your screen. Because 0:10:52.120 --> 0:10:53.960 if you had to move the mouse exactly the same 0:10:54.000 --> 0:10:57.840 distance as the cursor needs to travel, you would need 0:10:58.320 --> 0:11:01.280 a really big desk. But yeah, so it's the it's 0:11:01.320 --> 0:11:04.480 the same principle, just in reverse, right. Uh. And also 0:11:04.559 --> 0:11:07.120 the robots in these cases, they have a limited number 0:11:07.160 --> 0:11:10.600 of objects and tools they're designed to interact with. Uh. 0:11:10.600 --> 0:11:13.440 And as we said, it's directly controlled by human operator. 0:11:13.520 --> 0:11:15.840 So it's not the same thing as the needing a 0:11:15.920 --> 0:11:19.880 robot that has the kind of dexterity of a human hand. Now, 0:11:19.920 --> 0:11:23.920 some robots can achieve more variable motion, uh, not by 0:11:23.960 --> 0:11:28.319 having a really really powerfully generalist in defector like a 0:11:28.400 --> 0:11:31.839 human hand, but just being able to like switch out tools, right, 0:11:31.920 --> 0:11:33.880 Oh sure, yeah, just sort of like a plug and 0:11:33.920 --> 0:11:35.720 play kind of option of like, well, I need a 0:11:35.720 --> 0:11:37.400 clamp for this, and I need a blowtorch for that. 0:11:37.400 --> 0:11:39.680 I'm not going to put a blowtorch on the clamp. 0:11:39.720 --> 0:11:43.560 I'm going to switch out my robot in defector hand. 0:11:43.720 --> 0:11:47.040 So clearly it's like Dustin Hoffman's hook in the movie 0:11:47.040 --> 0:11:49.680 of the same name, where he could switch the hook 0:11:49.720 --> 0:11:52.400 out his other hook he has. He has one that 0:11:52.600 --> 0:11:57.439 is a double cigarette hold, right, and he has one 0:11:57.480 --> 0:12:01.480 that's a magnifying glass. Yeah. Does he have one that's 0:12:01.480 --> 0:12:04.080 a hand mirror? He might have. Yeah, he did have 0:12:04.120 --> 0:12:05.719 one that was a mirror. You're right. Yeah, it's when 0:12:05.720 --> 0:12:10.480 he's being um attended to by smee. Yeah, where his 0:12:10.480 --> 0:12:15.080 his mustache is being perfectly curled. I can I can 0:12:15.080 --> 0:12:19.320 respect that. Yeah, yeah, but the same sort of thing. Well, anyway, 0:12:19.360 --> 0:12:22.080 so we've discussed all the reasons that robots generally just 0:12:22.160 --> 0:12:25.160 have a very specific type of end defectory that's for 0:12:25.280 --> 0:12:28.959 the job they do. But why would you maybe need 0:12:29.000 --> 0:12:32.600 a robot to have something more like a human hand. Joe, 0:12:32.679 --> 0:12:34.920 let me let me run this by you. We humans 0:12:34.920 --> 0:12:38.400 we've been around. The stuff that we've built, we've built 0:12:38.400 --> 0:12:41.760 for ourselves for the most part, and and we humans, 0:12:42.080 --> 0:12:44.839 uh many of us, have hands, and so the stuff 0:12:44.880 --> 0:12:48.760 we have built has been meant for creatures that have hands. 0:12:49.280 --> 0:12:52.320 And therefore the world that we interact with, particularly the 0:12:52.320 --> 0:12:57.600 the man made portion of that world, is really hand centric. 0:12:57.760 --> 0:12:59.960 So if you want to interact with that world, hand 0:13:00.200 --> 0:13:02.520 are pretty much where it's at. So what you're saying 0:13:02.600 --> 0:13:05.560 is that asking a robot to move around in the 0:13:05.640 --> 0:13:10.319 human world is unbearably cruel and frustrating if that robot 0:13:10.320 --> 0:13:12.240 does not have hands and you need it to do 0:13:12.320 --> 0:13:16.920 things that hands would come in really handy. Yes, I like, 0:13:16.960 --> 0:13:20.840 if you ask a rumba to open a door, I'm sorry, 0:13:20.960 --> 0:13:23.400 that's just not happening. Trying to get clamps to break 0:13:23.480 --> 0:13:26.079 the yard for you, It's gonna have a hard time. Yeah, 0:13:26.200 --> 0:13:29.200 I mean what I Here's the funny thing is that 0:13:29.280 --> 0:13:33.640 as a left hander, UM, I encounter a tiny percentage 0:13:33.880 --> 0:13:36.640 of the frustration that if a robot could feel it, 0:13:36.640 --> 0:13:40.000 it's certainly would feel in a world that is made 0:13:40.080 --> 0:13:44.520 for people who are predominantly right hand dominant. Right, So 0:13:44.840 --> 0:13:47.080 I have this little experience where when I want to 0:13:47.120 --> 0:13:50.559 interact with tools. I need to use certain things. Sometimes 0:13:50.640 --> 0:13:54.160 it is counterintuitive and difficult, and sometimes even in the 0:13:54.200 --> 0:13:57.400 case of chainsaws, dangerous for a left hander to wield. 0:13:58.200 --> 0:14:01.440 And so uh it there and think, well, I actually 0:14:01.480 --> 0:14:04.000 have hands. It's just that I'm dominant with my left hand, 0:14:04.000 --> 0:14:06.600 not my right hand. For robots that don't have hands, 0:14:07.000 --> 0:14:11.520 then these problems are magnified by orders of magnitude, and 0:14:11.559 --> 0:14:14.800 so so the robots that exist today with very specific 0:14:14.840 --> 0:14:17.080 in defectors are not really a problem because they don't 0:14:17.120 --> 0:14:19.520 come into our world. They're not supposed to be able 0:14:19.560 --> 0:14:23.080 to do all the things we do. But people project 0:14:23.280 --> 0:14:25.920 that in the future, maybe in the next couple of decades, 0:14:25.960 --> 0:14:29.680 we're going to start introducing robots that are not the 0:14:29.760 --> 0:14:34.120 specialists that robots have always been, but their generalists. And 0:14:34.160 --> 0:14:37.120 they're not segregated from us in the way robots have 0:14:37.160 --> 0:14:40.800 always been. They're integrated into our lives. They live with us, 0:14:40.800 --> 0:14:43.320 and they do the same kinds of jobs we do. 0:14:43.920 --> 0:14:47.040 And so, for exactly the reason you point out, these 0:14:47.120 --> 0:14:50.120 robots need to be able to touch the same things 0:14:50.120 --> 0:14:52.920 we touch, and touch them in the same way, manipulate 0:14:52.960 --> 0:14:55.440 them in the same way to to be able to 0:14:55.520 --> 0:14:58.520 handle the stuff we handle without smashing it or dropping 0:14:58.560 --> 0:15:01.080 it or not knowing what to do with it. Yeah, 0:15:01.120 --> 0:15:05.320 otherwise you end up having to redesign all the tools, right, 0:15:05.360 --> 0:15:08.360 you'd have to redesign the tools for the robots, which 0:15:08.400 --> 0:15:11.240 seems like that would be a whole lot more work, 0:15:12.000 --> 0:15:15.000 depending on how tricky it is to really get a 0:15:15.040 --> 0:15:18.160 good set of robotic hands. Also, I want to point 0:15:18.160 --> 0:15:22.120 out I'm so sad that our listeners can't watch this 0:15:22.200 --> 0:15:25.960 podcast because we are just naturally gesturing with our hands 0:15:26.000 --> 0:15:30.600 a whole bunch a lot to begin with. That's true. Yeah, 0:15:30.640 --> 0:15:33.040 I do make a lot of crab claw emotions in general, 0:15:33.080 --> 0:15:36.520 and most podcasts, that's true, full of clamps. Yeah. But 0:15:36.640 --> 0:15:38.960 another thing, so one of them. One of the things 0:15:39.080 --> 0:15:42.760 is that we want generalist robots rather than specialist robots 0:15:42.760 --> 0:15:44.960 to be able to integrate into our lives in the future. 0:15:45.000 --> 0:15:47.520 But the other one is just the The human hand 0:15:47.680 --> 0:15:53.160 is an amazingly remarkably powerful machine. It's it's a it's 0:15:53.160 --> 0:15:55.600 a miracle of nature. It can be adapted to so 0:15:55.640 --> 0:15:58.720 many different jobs. And this isn't just because it's natural. 0:15:58.880 --> 0:16:02.320 I mean, I'm not saying any biological version of an 0:16:02.320 --> 0:16:04.360 indefector would be able to do this because think about 0:16:04.360 --> 0:16:07.520 a crabs claw or an ants four limbs. They can't 0:16:07.560 --> 0:16:10.120 do all the things that the primate hand can do. 0:16:10.800 --> 0:16:15.720 It's the primate hand, and specifically that's so amazingly powerful. 0:16:16.040 --> 0:16:22.080 It can push, pull, punch, pinch, slap, climb, throw, place 0:16:22.240 --> 0:16:25.840 things definitely, I mean it can do in thousands of things. 0:16:26.520 --> 0:16:28.720 And and it can do all these things with objects 0:16:28.760 --> 0:16:34.359 of vastly varying size, weight, and shape. The minimum roboticists 0:16:34.440 --> 0:16:37.360 say that you need to to approximate to human hands 0:16:37.440 --> 0:16:42.000 gripping and manipulation capacity is nine degrees of freedom. That's 0:16:42.200 --> 0:16:46.480 that's three fingers capable of moving in a three dimensional space. Yeah, 0:16:46.520 --> 0:16:48.600 and I feel like I've seen robotic hands that are 0:16:48.640 --> 0:16:51.480 basically three fingers. Yeah. Yeah, it looks a little bit 0:16:51.520 --> 0:16:55.200 like a like a nightcrawler hand. Yeah, um um. But 0:16:55.440 --> 0:16:57.520 but a fully functional hand is a lot more complicated 0:16:57.560 --> 0:16:59.920 than that. Yeah. The human the human hand has more 0:17:00.000 --> 0:17:02.560 than twenty independent degrees of freedom. I think I've seen 0:17:02.600 --> 0:17:06.080 somebody say twenty seven somewhere. That might be a disputable number, 0:17:06.119 --> 0:17:09.320 but definitely more than twenty. Uh. And it's controlled by 0:17:09.359 --> 0:17:12.680 more than thirty muscles, which in in robotics terms would 0:17:12.680 --> 0:17:17.000 be actuators. Yeah, and so, uh, you know, when you 0:17:17.040 --> 0:17:19.760 think about something that complex. Now, let's let's be fair. 0:17:20.359 --> 0:17:23.600 The human hand has had the benefit of millions of 0:17:23.800 --> 0:17:29.239 years of evolution to lead to that that that approach, right. 0:17:29.480 --> 0:17:32.040 Robotic hands, on the other hand, have had a few decades. Right, 0:17:32.119 --> 0:17:37.480 So we are trying to replicate what took nature millions 0:17:37.480 --> 0:17:40.520 of years to create. And uh, and so we shouldn't 0:17:40.720 --> 0:17:47.240 be so not skeptical, but so dismissive of how quote 0:17:47.280 --> 0:17:52.160 unquote primitive robotic hands are right now, because honestly, they're 0:17:52.160 --> 0:17:56.760 getting to be pretty sophisticated. And also, I did want 0:17:56.760 --> 0:17:59.080 to point out that, you know, just because the human 0:17:59.119 --> 0:18:03.120 hand works so for us, doesn't necessarily mean that it's 0:18:03.200 --> 0:18:05.560 the only design that we can come up with. It 0:18:05.640 --> 0:18:08.639 can do this kind of amazing stuff. There's some really 0:18:08.680 --> 0:18:12.600 cool generalist end defectors that are being designed that look 0:18:12.680 --> 0:18:15.440 nothing like a human hand. I wanted to mention one 0:18:15.840 --> 0:18:18.720 sort of as an example of how advanced they can 0:18:18.800 --> 0:18:23.080 be yet yet still failed to match our hands design ultimately, 0:18:23.520 --> 0:18:26.000 and I think we've we've mentioned this type on on 0:18:26.040 --> 0:18:30.800 the show before. Um, it's based on jamming transition technology 0:18:30.840 --> 0:18:33.600 and okay, and in physics, jamming is what happens when 0:18:33.960 --> 0:18:37.679 granular materials that would usually flow, like a sand or 0:18:37.680 --> 0:18:41.720 sugar or coffee grounds or whatever, they become rigid when 0:18:41.760 --> 0:18:44.800 they're like really packed in. Yeah. Like like if if, 0:18:44.840 --> 0:18:47.959 if you've ever handled a package of vacuum sealed coffee grounds, 0:18:48.400 --> 0:18:51.200 it's a it's a brick. It's solid, completely solid, but 0:18:51.200 --> 0:18:53.159 but if you open it up, you can pour the 0:18:53.200 --> 0:18:58.560 grounds into your coffee maker. Um sand flows through our glasses, 0:18:59.320 --> 0:19:03.480 but in a hack sec it becomes a solid, bouncy unit. Yeah, so, 0:19:03.560 --> 0:19:05.040 and I like that. We I think this is the 0:19:05.080 --> 0:19:07.800 second mention of hackey sex we've had in this episode 0:19:07.840 --> 0:19:10.800 so far. Good times. I think we actually mentioned hacky 0:19:10.800 --> 0:19:14.800 sex before we start. It's okay because it's always in 0:19:14.800 --> 0:19:18.800 our hearts and sometimes on our knees and feet. Jonathan, 0:19:18.840 --> 0:19:20.639 were you a high school hacky sack? I am not 0:19:20.680 --> 0:19:27.080 answering that question, Joe, please go on, Lauren, no problem, Jonathan, Okay, So, So, 0:19:27.240 --> 0:19:31.399 jamming transition, which is the jamming transition technology, is the 0:19:31.440 --> 0:19:33.520 word I used before, the term I used before. A 0:19:33.600 --> 0:19:38.440 jamming transition occurs when grains of stuff become dense enough 0:19:38.480 --> 0:19:41.720 to act as a solid. And this, as it turns out, 0:19:41.840 --> 0:19:45.560 is a really useful property because researchers figured out that 0:19:45.720 --> 0:19:48.520 if you fill a soft rubber ball with a granular 0:19:48.560 --> 0:19:51.199 material and then attached that rubber ball to the end 0:19:51.200 --> 0:19:54.680 of a vacuum tube, you can quickly and easily move 0:19:54.800 --> 0:19:58.600 air into and out of the ball, thus changing the 0:19:58.680 --> 0:20:02.560 density and the behavior year of the material inside um. 0:20:02.720 --> 0:20:04.720 And you can use such a device to to grip 0:20:04.760 --> 0:20:07.439 and handle objects of all kinds of different shapes and 0:20:07.480 --> 0:20:11.440 sizes because they mold to it. And and then like 0:20:11.680 --> 0:20:13.800 if you've got a lego brick, the ball can bend 0:20:13.840 --> 0:20:16.359 around the top of the brick or the top of 0:20:16.400 --> 0:20:19.800 a bottle cap and solidify around it in order to 0:20:19.800 --> 0:20:22.760 pick it up. Then it can soften again to put 0:20:22.800 --> 0:20:27.159 it down. And this is pretty amazing. Yeah, yeah, you 0:20:27.240 --> 0:20:29.119 might have you might have seen one that the bright 0:20:29.280 --> 0:20:32.680 green versible um. It looks like a like a weird 0:20:32.720 --> 0:20:35.679 little tennis ball on the end of a tube. And 0:20:35.680 --> 0:20:37.880 and about a year or two ago, I think there 0:20:37.880 --> 0:20:40.480 were a lot of videos going around interesting. But yeah, yeah, 0:20:40.480 --> 0:20:44.280 it's a it's great. It's really fascinating and beautiful to 0:20:44.320 --> 0:20:47.760 watch inaction. Also, it can't play a piano at all. 0:20:48.560 --> 0:20:51.280 It would suck at that. What's amazing to me is 0:20:51.320 --> 0:20:54.359 that this actually reminds me of I saw a comedy show. 0:20:54.520 --> 0:20:57.480 It was improvised comedy show, and one of the characters 0:20:57.480 --> 0:21:00.600 in the show was a robot that had become horrified 0:21:00.600 --> 0:21:03.040 that had had killed his best friend because did not 0:21:03.119 --> 0:21:07.679 realize that humans can't reboot, and so had replaced replaced 0:21:07.680 --> 0:21:10.119 these hands. He would say that over and over in 0:21:10.160 --> 0:21:13.639 the episode. He replaced his hands with oven mits filled 0:21:13.680 --> 0:21:16.000 with cookie dough, is what he described it, so that 0:21:16.040 --> 0:21:17.760 way he could no longer hurt anything. It would just 0:21:17.800 --> 0:21:21.680 flop against people and uh and it was. It was 0:21:21.800 --> 0:21:25.080 hilarious and heartbreaking at the same time. So when you're 0:21:25.119 --> 0:21:29.720 describing this like, I can't believe that that incredibly weird 0:21:29.880 --> 0:21:34.679 improvised comedy thing I saw fifteen years ago is kind 0:21:34.680 --> 0:21:37.160 of sort of the same thing you're talking about. It's 0:21:37.200 --> 0:21:39.159 just it's just it's a cookie dough. It would be 0:21:39.200 --> 0:21:44.320 some sort of granular type stuff. But that's amazing and weird. 0:21:45.600 --> 0:21:49.200 Are the are the future thinkers? I guess, I guess 0:21:49.240 --> 0:21:52.000 the Jules Verne of today. I have to I have 0:21:52.040 --> 0:21:53.800 to let him know. I know the guy who did it, 0:21:53.880 --> 0:21:56.280 so I have to tell him, Hey, you were prescient 0:21:56.960 --> 0:22:01.440 in your goofy makeup ups. Some people who have been 0:22:01.440 --> 0:22:04.720 working with robot hands even earlier than this improv team 0:22:04.760 --> 0:22:09.040 of yours. Um was it? Was it with Dad's Garage, 0:22:09.200 --> 0:22:13.639 Christ's Garage. You guys, if you're ever in Atlanta, if 0:22:13.680 --> 0:22:16.359 you're from Atlanta, you should totally check out the work 0:22:16.400 --> 0:22:19.280 that Dad's Garage does. It's a terrific theater company anyway. 0:22:19.359 --> 0:22:22.280 Um okay, So, so even before Dad's Garage, people were 0:22:22.280 --> 0:22:25.840 working on robot hands. And if we were going to 0:22:25.960 --> 0:22:28.800 go into the full history of this, it would the 0:22:28.880 --> 0:22:31.320 history of the development in the technology. That would be 0:22:31.320 --> 0:22:35.040 a whole episode unto itself, probably a couple And maybe 0:22:35.119 --> 0:22:37.000 we should do that on tech stuff sometimes that might 0:22:37.000 --> 0:22:39.040