0 00:00:00,000 --> 00:00:30,000 Dear viewer, these subtitles were generated by a machine via the service Trint and therefore are (very) buggy. If you are capable, please help us to create good quality subtitles: https://c3subtitles.de/talk/739 Thanks! 1 00:00:21,240 --> 00:00:23,099 Thank you very much. 2 00:00:23,100 --> 00:00:25,379 I just wanted to say first, thank 3 00:00:25,380 --> 00:00:27,629 you for coming to hear my talk and 4 00:00:27,630 --> 00:00:29,009 to let you know that there are a few 5 00:00:29,010 --> 00:00:30,359 first going on right now. 6 00:00:30,360 --> 00:00:32,639 So it's my first time being 7 00:00:32,640 --> 00:00:34,979 at C.C.C., so I'm excited 8 00:00:34,980 --> 00:00:35,980 to be here. 9 00:00:42,000 --> 00:00:44,309 It's also my first time giving 10 00:00:44,310 --> 00:00:46,409 a talk this late at night, so 11 00:00:46,410 --> 00:00:48,599 when I got my time slot, I asked myself 12 00:00:48,600 --> 00:00:50,699 the question, just how drunk will my 13 00:00:50,700 --> 00:00:52,949 audience be? So you can report 14 00:00:52,950 --> 00:00:55,079 back to me later about how 15 00:00:55,080 --> 00:00:56,129 drunk you are. 16 00:00:56,130 --> 00:00:58,239 And then the last thing that's a first 17 00:00:58,240 --> 00:01:00,419 that's a little more serious is 18 00:01:00,420 --> 00:01:02,159 this is my first time showing work in 19 00:01:02,160 --> 00:01:04,499 progress. So I'm a little bit nervous 20 00:01:04,500 --> 00:01:06,959 about that. So be gentle with me. 21 00:01:06,960 --> 00:01:09,659 So when we 22 00:01:09,660 --> 00:01:11,879 start talking about edible soft robotics 23 00:01:11,880 --> 00:01:13,979 and this exploration of candy 24 00:01:13,980 --> 00:01:16,199 as an engineered material, it 25 00:01:16,200 --> 00:01:18,539 was kicked off by just this thought I had 26 00:01:18,540 --> 00:01:21,149 while making cast silicone 27 00:01:21,150 --> 00:01:23,819 actuators that there's an intense 28 00:01:23,820 --> 00:01:26,249 similarity between the cast 29 00:01:26,250 --> 00:01:29,009 silicone and gummy candies. 30 00:01:29,010 --> 00:01:30,839 And it's like one of those idle thoughts 31 00:01:30,840 --> 00:01:32,219 you have when you're fabricating 32 00:01:32,220 --> 00:01:33,569 something. 33 00:01:33,570 --> 00:01:35,849 But I was really captured 34 00:01:35,850 --> 00:01:38,879 by this, like joking, not joking place. 35 00:01:38,880 --> 00:01:41,309 I wanted it to be something 36 00:01:41,310 --> 00:01:43,439 that is totally ridiculous that I 37 00:01:43,440 --> 00:01:45,209 could also take very seriously. 38 00:01:46,380 --> 00:01:48,479 And so when I was 39 00:01:48,480 --> 00:01:50,399 trying to identify, like, should I do 40 00:01:50,400 --> 00:01:52,229 this project or not, since I've been a 41 00:01:52,230 --> 00:01:54,479 government contractor, when you 42 00:01:54,480 --> 00:01:56,039 propose a project, you have to, like, 43 00:01:56,040 --> 00:01:57,689 turn in these sheets about what your 44 00:01:57,690 --> 00:01:58,889 project's going to be. 45 00:01:58,890 --> 00:02:02,039 So I went ahead and made one about 46 00:02:02,040 --> 00:02:03,449 making Candie's off robots. 47 00:02:03,450 --> 00:02:05,099 So this is my quad chart. 48 00:02:05,100 --> 00:02:06,989 If I were going to try to get the 49 00:02:06,990 --> 00:02:09,329 government to finance my project. 50 00:02:09,330 --> 00:02:11,909 But for me, it was really about 51 00:02:11,910 --> 00:02:15,209 this place in the potential benefit zone 52 00:02:15,210 --> 00:02:17,309 that says I see 53 00:02:17,310 --> 00:02:19,799 food as the highest form of interactivity 54 00:02:19,800 --> 00:02:22,049 and that that means it engages 55 00:02:22,050 --> 00:02:23,819 all five of your senses. 56 00:02:23,820 --> 00:02:25,889 And after you eat it, it becomes 57 00:02:25,890 --> 00:02:27,149 part of you. 58 00:02:27,150 --> 00:02:28,949 I couldn't think of anything more 59 00:02:28,950 --> 00:02:30,029 interactive than that. 60 00:02:30,030 --> 00:02:32,069 So I was that seemed like a good enough 61 00:02:32,070 --> 00:02:34,769 reason to go down this pathway. 62 00:02:34,770 --> 00:02:36,839 So kind of the next thing I do when I 63 00:02:36,840 --> 00:02:38,909 have a project is I ask myself, 64 00:02:38,910 --> 00:02:40,499 what's the history and the state of the 65 00:02:40,500 --> 00:02:42,269 art? I don't want to reinvent the wheel. 66 00:02:42,270 --> 00:02:43,919 I don't want to start from zero if I 67 00:02:43,920 --> 00:02:44,999 don't have to. 68 00:02:45,000 --> 00:02:47,009 So who is doing the work? 69 00:02:47,010 --> 00:02:48,209 What are the related fields? 70 00:02:48,210 --> 00:02:49,650 If you want to make a candy robot 71 00:02:51,060 --> 00:02:53,339 so soft, robotics 72 00:02:53,340 --> 00:02:56,039 was obviously a starting point. 73 00:02:56,040 --> 00:02:57,719 Luckily, it's something I've spent a lot 74 00:02:57,720 --> 00:02:59,399 of time thinking about, so I had a good 75 00:02:59,400 --> 00:03:00,959 foundation there. 76 00:03:00,960 --> 00:03:01,960 And then 77 00:03:03,270 --> 00:03:05,789 the next one is molecular 78 00:03:05,790 --> 00:03:06,869 gastronomy. 79 00:03:06,870 --> 00:03:09,329 So molecular gastronomy is this idea 80 00:03:09,330 --> 00:03:11,639 of using food as an art and 81 00:03:11,640 --> 00:03:13,769 an experience using the tools 82 00:03:13,770 --> 00:03:15,539 of science to get there. 83 00:03:15,540 --> 00:03:17,609 And the thing that was interesting to 84 00:03:17,610 --> 00:03:19,739 me about this field is 85 00:03:19,740 --> 00:03:21,749 that all cooking is technology. 86 00:03:21,750 --> 00:03:24,269 Just some technology is technology 87 00:03:24,270 --> 00:03:26,549 we're used to. So, for example, 88 00:03:26,550 --> 00:03:28,799 a microwave that was the cutting 89 00:03:28,800 --> 00:03:31,079 edge of technology not so long ago. 90 00:03:31,080 --> 00:03:33,029 And now it's just a typical tool. 91 00:03:33,030 --> 00:03:35,159 So as we look at this molecular 92 00:03:35,160 --> 00:03:37,259 gastronomy place, I'm interested to 93 00:03:37,260 --> 00:03:39,389 see which things that seem very 94 00:03:39,390 --> 00:03:41,729 far out there right now will become part 95 00:03:41,730 --> 00:03:42,809 of our regular cooking. 96 00:03:45,270 --> 00:03:47,459 If you look up edible robotics, the 97 00:03:47,460 --> 00:03:49,559 first thing that comes up is all health 98 00:03:49,560 --> 00:03:51,089 related projects. 99 00:03:51,090 --> 00:03:53,309 So this, for example, is 100 00:03:53,310 --> 00:03:55,439 a robot that is 101 00:03:55,440 --> 00:03:56,489 made out of meat. 102 00:03:56,490 --> 00:03:58,169 It's made out of sausage casings. 103 00:03:58,170 --> 00:04:00,509 And its purpose is to 104 00:04:00,510 --> 00:04:02,939 if you swallow a watch battery, 105 00:04:02,940 --> 00:04:04,199 it can do a lot of damage to your 106 00:04:04,200 --> 00:04:05,879 stomach. So this is supposed to go 107 00:04:05,880 --> 00:04:07,679 through your stomach and collect the 108 00:04:07,680 --> 00:04:09,269 watch battery and then flush it out of 109 00:04:09,270 --> 00:04:10,270 your system. 110 00:04:12,150 --> 00:04:14,309 Then the next one is kind of 111 00:04:14,310 --> 00:04:16,349 the other side of the same problem. 112 00:04:16,350 --> 00:04:17,999 So since you don't want to swallow a 113 00:04:18,000 --> 00:04:19,349 watch battery, but you might want to 114 00:04:19,350 --> 00:04:21,629 swallow a robot, they came 115 00:04:21,630 --> 00:04:23,729 up with a battery that's made 116 00:04:23,730 --> 00:04:25,799 from cuttlefish ink that you 117 00:04:25,800 --> 00:04:27,919 can swallow it and then if 118 00:04:27,920 --> 00:04:29,640 it gets digested over time. 119 00:04:31,720 --> 00:04:33,789 So then to move from the health 120 00:04:33,790 --> 00:04:35,409 space to the art space, which is the 121 00:04:35,410 --> 00:04:36,759 other people who are making edible 122 00:04:36,760 --> 00:04:38,949 robots, there's this project 123 00:04:38,950 --> 00:04:39,939 called Living Food. 124 00:04:39,940 --> 00:04:42,699 It's inspired by 125 00:04:42,700 --> 00:04:45,039 the synthetic biology 126 00:04:45,040 --> 00:04:47,199 movement. So the idea was that 127 00:04:47,200 --> 00:04:49,359 they wanted to make living food 128 00:04:49,360 --> 00:04:51,459 that moved and breathed and like 129 00:04:51,460 --> 00:04:52,719 that would be part of the dining 130 00:04:52,720 --> 00:04:54,219 experience. 131 00:04:54,220 --> 00:04:56,439 Sadly, this was only a theoretical 132 00:04:56,440 --> 00:04:58,419 project. So there weren't any, like 133 00:04:58,420 --> 00:05:00,429 technological takeaways I could apply to 134 00:05:00,430 --> 00:05:02,319 my work. But it was interesting to 135 00:05:02,320 --> 00:05:03,320 consider 136 00:05:04,840 --> 00:05:06,849 then. The other thing that happened is 137 00:05:06,850 --> 00:05:08,619 when I was talking to people about edible 138 00:05:08,620 --> 00:05:10,779 robots, people kept 139 00:05:10,780 --> 00:05:12,939 pointing me back at NYC Resistor, which 140 00:05:12,940 --> 00:05:14,979 is the hackerspace that I'm part of. 141 00:05:14,980 --> 00:05:16,959 So I was like, oh, you know, you fell in 142 00:05:16,960 --> 00:05:18,609 with the right group of hackers when your 143 00:05:18,610 --> 00:05:20,829 research leads you right back to your own 144 00:05:20,830 --> 00:05:21,830 door. 145 00:05:22,390 --> 00:05:24,579 So the top project is called 146 00:05:24,580 --> 00:05:26,979 Lady Godiva, and it's a chocolate 147 00:05:26,980 --> 00:05:30,519 that's a closed circuit made with silver 148 00:05:30,520 --> 00:05:32,079 edible silver foil. 149 00:05:32,080 --> 00:05:34,329 So when you eat it, the chocolate moans 150 00:05:34,330 --> 00:05:35,859 and you have kind of this experience. 151 00:05:35,860 --> 00:05:38,259 And then below 152 00:05:38,260 --> 00:05:41,199 is a Ajello piano. 153 00:05:41,200 --> 00:05:43,299 So the things about these projects is 154 00:05:43,300 --> 00:05:45,429 that they're dealing with sound. 155 00:05:45,430 --> 00:05:46,929 But I've always been much more interested 156 00:05:46,930 --> 00:05:48,879 in motion. So there wasn't a lot for me 157 00:05:48,880 --> 00:05:49,989 to take away from these. 158 00:05:51,400 --> 00:05:53,979 And then finally there was this robotic 159 00:05:53,980 --> 00:05:56,589 big project. It was a large collaboration 160 00:05:56,590 --> 00:05:59,079 through an art space and a university 161 00:05:59,080 --> 00:06:00,909 and a private company. 162 00:06:00,910 --> 00:06:03,789 So they were making edible robots 163 00:06:03,790 --> 00:06:05,979 that moved the Rice Krispie 164 00:06:05,980 --> 00:06:06,819 robot. 165 00:06:06,820 --> 00:06:09,549 Is Rice Krispies covering 166 00:06:09,550 --> 00:06:10,749 a cable system? 167 00:06:10,750 --> 00:06:13,029 So I really like the motion 168 00:06:13,030 --> 00:06:14,319 and I like the form, but I don't really 169 00:06:14,320 --> 00:06:15,819 like that you have to eat around the 170 00:06:15,820 --> 00:06:17,349 cables. 171 00:06:17,350 --> 00:06:19,419 I was a lot more interested 172 00:06:19,420 --> 00:06:20,949 in the cucumber robot because the 173 00:06:20,950 --> 00:06:23,049 cucumber is strong enough that 174 00:06:23,050 --> 00:06:25,329 it can use its own material to be 175 00:06:25,330 --> 00:06:26,589 the cables. 176 00:06:26,590 --> 00:06:28,719 So it's like this is this is where I 177 00:06:28,720 --> 00:06:31,329 want to be in the robotic space. 178 00:06:31,330 --> 00:06:33,459 But I didn't really want 179 00:06:33,460 --> 00:06:35,349 to eat a cucumber after I had been a 180 00:06:35,350 --> 00:06:36,939 robot. Like, I wasn't really drawn to 181 00:06:36,940 --> 00:06:38,829 that. So that's why I was more interested 182 00:06:38,830 --> 00:06:40,179 in candy. 183 00:06:40,180 --> 00:06:41,180 So that's this. 184 00:06:42,680 --> 00:06:44,120 That's the state of the art. 185 00:06:49,160 --> 00:06:51,229 So as we go down this road, then the next 186 00:06:51,230 --> 00:06:52,549 thing I like to do is consult with 187 00:06:52,550 --> 00:06:55,009 experts, which is a really interesting 188 00:06:55,010 --> 00:06:57,259 question. When you're working on an 189 00:06:57,260 --> 00:06:59,359 emerging problem, an emerging field. 190 00:06:59,360 --> 00:07:01,999 Who are the experts? 191 00:07:02,000 --> 00:07:04,399 When I go down this path, I like to go 192 00:07:04,400 --> 00:07:06,499 as broadly as possible, like I 193 00:07:06,500 --> 00:07:08,749 might get information I might not 194 00:07:08,750 --> 00:07:09,649 use. 195 00:07:09,650 --> 00:07:11,779 But the further I go afield into who 196 00:07:11,780 --> 00:07:13,879 I think experts are, the more surprising 197 00:07:13,880 --> 00:07:16,609 information and the better synthesis 198 00:07:16,610 --> 00:07:18,649 of ideas. I will get across the problem 199 00:07:18,650 --> 00:07:21,139 space. So in this particular 200 00:07:21,140 --> 00:07:23,359 case, I consulted 201 00:07:23,360 --> 00:07:25,969 with Matthew Barguti, who is 202 00:07:25,970 --> 00:07:28,219 actually my collaborator 203 00:07:28,220 --> 00:07:30,319 at Super Replacer, and he is actually the 204 00:07:30,320 --> 00:07:32,569 person who convinced me to get into soft 205 00:07:32,570 --> 00:07:33,859 robotics at all. 206 00:07:33,860 --> 00:07:35,899 And he's also the person who 207 00:07:37,040 --> 00:07:38,929 told me that I had to make candy robots 208 00:07:38,930 --> 00:07:41,119 once I said it, because I, 209 00:07:41,120 --> 00:07:42,679 I said, well, it doesn't really fit our 210 00:07:42,680 --> 00:07:44,149 soft robotics agenda. 211 00:07:44,150 --> 00:07:46,249 And then he said, part of our soft 212 00:07:46,250 --> 00:07:49,399 robotics agenda is to be awesome. 213 00:07:49,400 --> 00:07:51,649 So you have him to thank or 214 00:07:51,650 --> 00:07:53,419 blame about this project. 215 00:07:53,420 --> 00:07:55,609 And you can watch his talk from 216 00:07:55,610 --> 00:07:58,099 last year, C.C.C., about 217 00:07:58,100 --> 00:07:59,419 soft robotics in general. 218 00:07:59,420 --> 00:08:00,829 If you want to learn more about the 219 00:08:00,830 --> 00:08:01,830 general field, 220 00:08:02,990 --> 00:08:05,299 the next person I consulted with is 221 00:08:05,300 --> 00:08:06,949 a friend of mine, Tim Rodriguez. 222 00:08:06,950 --> 00:08:09,169 He's most known for being 223 00:08:09,170 --> 00:08:11,329 a board game designer, but he also has 224 00:08:11,330 --> 00:08:13,189 a degree in food science. 225 00:08:13,190 --> 00:08:15,379 So one of the things that was 226 00:08:15,380 --> 00:08:17,089 really useful to talking to him about 227 00:08:17,090 --> 00:08:19,459 food science is that he took me down 228 00:08:19,460 --> 00:08:21,739 the path, way of looking at 229 00:08:21,740 --> 00:08:23,839 candy at the molecular level, 230 00:08:23,840 --> 00:08:26,299 because it's actually all of these 231 00:08:26,300 --> 00:08:28,489 very, very simple systems in 232 00:08:28,490 --> 00:08:30,739 terms of the chemistry. 233 00:08:30,740 --> 00:08:32,808 And he had some great insights 234 00:08:32,809 --> 00:08:36,079 to about the relationship between 235 00:08:36,080 --> 00:08:38,298 why Candy is sometimes not a very 236 00:08:38,299 --> 00:08:41,149 strong material where and why vegetables 237 00:08:41,150 --> 00:08:42,678 are strong. 238 00:08:42,679 --> 00:08:43,908 So that's something that will come up 239 00:08:43,909 --> 00:08:45,049 again a little later. 240 00:08:46,490 --> 00:08:49,639 And then my friend Liz Hara, she 241 00:08:49,640 --> 00:08:51,359 is a professional puppet builder. 242 00:08:51,360 --> 00:08:53,479 I have worked with her at the Jim 243 00:08:53,480 --> 00:08:55,549 Henson Company and other places, 244 00:08:55,550 --> 00:08:57,409 and she is also an extreme candy 245 00:08:57,410 --> 00:08:58,369 enthusiast. 246 00:08:58,370 --> 00:09:00,469 She has eaten every kind of candy 247 00:09:00,470 --> 00:09:02,629 that is mass produced in America, which 248 00:09:02,630 --> 00:09:05,269 is quite a feat. 249 00:09:05,270 --> 00:09:07,369 And then one of the other things that's 250 00:09:07,370 --> 00:09:09,469 been really wonderful about talking 251 00:09:09,470 --> 00:09:11,509 to her is because she made edible puppets 252 00:09:11,510 --> 00:09:13,639 once. So she gave me this fabulous 253 00:09:13,640 --> 00:09:15,949 insight. I can tell you stabbing licorice 254 00:09:15,950 --> 00:09:17,929 into a hot dog does not make a durable 255 00:09:17,930 --> 00:09:19,999 puppet. This is the voice 256 00:09:20,000 --> 00:09:21,000 of experience. 257 00:09:22,910 --> 00:09:24,739 And then the other thing that's been so 258 00:09:24,740 --> 00:09:26,149 great about working with her is that 259 00:09:26,150 --> 00:09:28,339 she's very enthusiastic about 260 00:09:28,340 --> 00:09:30,259 the candy actuator space. 261 00:09:30,260 --> 00:09:32,449 So she's been texting me more information 262 00:09:32,450 --> 00:09:33,529 even while I've been here. 263 00:09:35,390 --> 00:09:38,479 So now that I've talked to the experts, 264 00:09:38,480 --> 00:09:40,519 I needed to balance my project. 265 00:09:40,520 --> 00:09:42,889 I needed to decide what the evaluation 266 00:09:42,890 --> 00:09:44,959 criteria was, whether or not 267 00:09:44,960 --> 00:09:46,940 I was succeeding at making a soft robot. 268 00:09:48,020 --> 00:09:50,089 In this particular case, 269 00:09:50,090 --> 00:09:52,399 I decided you have to be able 270 00:09:52,400 --> 00:09:54,679 to eat the whole robot up to the computer 271 00:09:54,680 --> 00:09:56,989 control. As I said before, I'm not 272 00:09:56,990 --> 00:09:58,789 really a fan of the like eating around 273 00:09:58,790 --> 00:09:59,790 cables. 274 00:10:01,430 --> 00:10:04,189 Also, it needs to be easy to reproduce. 275 00:10:04,190 --> 00:10:06,529 The primary reason that I'm doing 276 00:10:06,530 --> 00:10:08,359 this project, other than to be awesome, 277 00:10:08,360 --> 00:10:10,489 is to make a how to to put into 278 00:10:10,490 --> 00:10:12,799 a book about soft robotics that I'm 279 00:10:12,800 --> 00:10:13,729 writing. 280 00:10:13,730 --> 00:10:16,369 And so what I decided 281 00:10:16,370 --> 00:10:18,499 to reproduce looked like was 282 00:10:18,500 --> 00:10:20,929 that it could 283 00:10:20,930 --> 00:10:23,659 use special specialty ingredients. 284 00:10:23,660 --> 00:10:25,639 You could have to go far afield to find 285 00:10:25,640 --> 00:10:27,229 the right things, but you shouldn't have 286 00:10:27,230 --> 00:10:29,419 to buy specialty equipment in order 287 00:10:29,420 --> 00:10:31,969 to do this project, because 288 00:10:31,970 --> 00:10:33,499 I don't want people filling up their 289 00:10:33,500 --> 00:10:34,849 lives with all of these things. 290 00:10:34,850 --> 00:10:37,009 They're only going to use once or 291 00:10:37,010 --> 00:10:38,929 twice to make a candy robot. 292 00:10:38,930 --> 00:10:41,119 The other was that it had to be not too 293 00:10:41,120 --> 00:10:42,979 fussy to make. 294 00:10:42,980 --> 00:10:45,049 When I talk about fuzzy to make, I don't 295 00:10:45,050 --> 00:10:46,279 know if any of you have ever made a 296 00:10:46,280 --> 00:10:48,529 meringue, but it's 297 00:10:48,530 --> 00:10:51,349 incredibly sensitive to the humidity 298 00:10:51,350 --> 00:10:53,149 and like you have to adjust everything 299 00:10:53,150 --> 00:10:54,949 very carefully. 300 00:10:54,950 --> 00:10:57,109 So I wanted the robot to 301 00:10:57,110 --> 00:10:58,819 be something that it wouldn't be 302 00:10:58,820 --> 00:11:00,350 temperature or weather dependent. 303 00:11:01,670 --> 00:11:03,619 I wanted it to elicit an emotional 304 00:11:03,620 --> 00:11:05,449 response that was strong. 305 00:11:05,450 --> 00:11:07,129 And for me, when you deal with candy, you 306 00:11:07,130 --> 00:11:08,359 either get two directions. 307 00:11:08,360 --> 00:11:10,369 You get like this joyful space, like this 308 00:11:10,370 --> 00:11:12,439 fun and happy space, or you get something 309 00:11:12,440 --> 00:11:15,259 that's like really gross like snot 310 00:11:15,260 --> 00:11:16,260 or like 311 00:11:17,360 --> 00:11:19,879 those jelly beans that tastes like vomit. 312 00:11:19,880 --> 00:11:22,189 So those were the two most 313 00:11:22,190 --> 00:11:24,439 exciting places for me in the edible 314 00:11:24,440 --> 00:11:25,759 robotics place. 315 00:11:25,760 --> 00:11:28,309 And then I wanted there to be a baseline 316 00:11:28,310 --> 00:11:30,469 fit flavor standard, so 317 00:11:30,470 --> 00:11:32,929 I didn't want it just to be technically 318 00:11:32,930 --> 00:11:34,999 edible. So a lot of people I talked to 319 00:11:35,000 --> 00:11:36,559 about this were like, well, you can eat 320 00:11:36,560 --> 00:11:38,779 fabric or you can eat paper. 321 00:11:38,780 --> 00:11:40,849 And I was like, I think that's really not 322 00:11:40,850 --> 00:11:42,529 in the spirit of the project. 323 00:11:42,530 --> 00:11:44,929 Like, I think that there has to be 324 00:11:44,930 --> 00:11:47,179 a baseline deliciousness 325 00:11:47,180 --> 00:11:48,079 that makes this. 326 00:11:48,080 --> 00:11:50,689 An appealing thing, and then 327 00:11:50,690 --> 00:11:52,519 it needs to be durable enough to make the 328 00:11:52,520 --> 00:11:54,619 day before, I don't like the 329 00:11:54,620 --> 00:11:56,929 pressure of like having to perform, 330 00:11:56,930 --> 00:11:58,849 making something to the last minute right 331 00:11:58,850 --> 00:12:00,440 before I serve it to someone, 332 00:12:01,460 --> 00:12:03,169 especially for something as complicated 333 00:12:03,170 --> 00:12:04,729 as a robot. 334 00:12:04,730 --> 00:12:06,619 So I wanted you to be able to make it the 335 00:12:06,620 --> 00:12:08,749 day before and then serve it. 336 00:12:08,750 --> 00:12:10,819 And then also, 337 00:12:10,820 --> 00:12:12,319 there was no need for long term 338 00:12:12,320 --> 00:12:14,509 stability, though, because once you get 339 00:12:14,510 --> 00:12:17,089 into the place where you're making a food 340 00:12:17,090 --> 00:12:19,219 actuator, long term 341 00:12:19,220 --> 00:12:21,319 storage just sounds like dusty, 342 00:12:21,320 --> 00:12:24,049 dirty handled by a lot of people, 343 00:12:24,050 --> 00:12:26,209 which then went away from kind of 344 00:12:26,210 --> 00:12:28,579 the the the delightful 345 00:12:28,580 --> 00:12:30,379 edible component I was looking for. 346 00:12:30,380 --> 00:12:32,509 So those are the specs that I'm 347 00:12:32,510 --> 00:12:34,039 going to measure success against. 348 00:12:36,500 --> 00:12:38,839 So now, 349 00:12:38,840 --> 00:12:41,179 since I got into technology 350 00:12:41,180 --> 00:12:43,549 kind of through being a fabricator 351 00:12:43,550 --> 00:12:45,799 and through my experience working 352 00:12:45,800 --> 00:12:47,989 with materials and like 353 00:12:47,990 --> 00:12:49,969 direct relationship with the material 354 00:12:49,970 --> 00:12:52,069 world, it makes the most sense 355 00:12:52,070 --> 00:12:54,139 to me to spend time with the 356 00:12:54,140 --> 00:12:56,129 materials and to touch the materials and 357 00:12:56,130 --> 00:12:57,919 to experience what they do. 358 00:12:57,920 --> 00:13:00,109 And aside from it being my comfort 359 00:13:00,110 --> 00:13:02,269 zone. The other thing I wanted to 360 00:13:02,270 --> 00:13:04,459 bring to the process 361 00:13:04,460 --> 00:13:06,799 was that I have seen many, 362 00:13:06,800 --> 00:13:08,389 many people build projects out of 363 00:13:08,390 --> 00:13:10,099 inappropriate materials. 364 00:13:10,100 --> 00:13:12,109 And you can see that they're spending a 365 00:13:12,110 --> 00:13:14,329 lot of time over and over again trying 366 00:13:14,330 --> 00:13:15,769 to make something do. 367 00:13:18,190 --> 00:13:20,409 An object that it cannot achieve, 368 00:13:20,410 --> 00:13:22,629 so because I don't know anything about 369 00:13:22,630 --> 00:13:24,699 Candy as a engineering 370 00:13:24,700 --> 00:13:26,859 material, before I started, I 371 00:13:26,860 --> 00:13:28,899 needed to spend time with the materials 372 00:13:28,900 --> 00:13:30,579 to know what's appropriate. 373 00:13:30,580 --> 00:13:31,580 Use cases were 374 00:13:33,730 --> 00:13:34,730 so. 375 00:13:35,460 --> 00:13:37,649 First thing I did was start acquiring 376 00:13:37,650 --> 00:13:39,989 some materials, so Kallu 377 00:13:39,990 --> 00:13:42,089 stands the place to buy hydro 378 00:13:42,090 --> 00:13:44,159 colloids in New York City. 379 00:13:45,660 --> 00:13:47,759 I'm going to say hydro quite a 380 00:13:47,760 --> 00:13:49,019 lot. 381 00:13:49,020 --> 00:13:51,149 That's what makes a gummy candy, a gummy 382 00:13:51,150 --> 00:13:52,150 candy. 383 00:13:52,740 --> 00:13:54,809 So just so we have a definition to start 384 00:13:54,810 --> 00:13:56,879 with, a hydro keloid is 385 00:13:56,880 --> 00:13:59,369 a stable suspension of hydrophilic 386 00:13:59,370 --> 00:14:00,659 molecules. 387 00:14:00,660 --> 00:14:03,179 So most of them 388 00:14:03,180 --> 00:14:05,159 are sugars, but some of them are 389 00:14:05,160 --> 00:14:07,169 proteins. So, for example, gelatins are 390 00:14:07,170 --> 00:14:09,539 proteins and there are hydrochloride. 391 00:14:09,540 --> 00:14:11,729 So moving on, I acquired 392 00:14:11,730 --> 00:14:12,730 a bunch of heidrich. 393 00:14:14,280 --> 00:14:16,709 Now, what I didn't realize when I started 394 00:14:16,710 --> 00:14:17,759 down this pathway 395 00:14:19,500 --> 00:14:21,809 gelling of hydro colloids 396 00:14:21,810 --> 00:14:24,179 is a foundational problem of molecular 397 00:14:24,180 --> 00:14:26,159 gastronomy. I thought that I was going 398 00:14:26,160 --> 00:14:28,229 into a space where everything was going 399 00:14:28,230 --> 00:14:29,699 to be mostly solved. 400 00:14:29,700 --> 00:14:31,979 But what I found instead is in 401 00:14:31,980 --> 00:14:34,229 this book, this is 22 402 00:14:34,230 --> 00:14:37,289 different options for gelling agents. 403 00:14:37,290 --> 00:14:39,539 So there's a lot 404 00:14:39,540 --> 00:14:42,149 of ways to achieve this outcome. 405 00:14:42,150 --> 00:14:43,679 And I'm actually going to read directly a 406 00:14:43,680 --> 00:14:45,179 quote from the book because it made me 407 00:14:45,180 --> 00:14:46,649 laugh. 408 00:14:46,650 --> 00:14:48,749 Unfortunately, relatively 409 00:14:48,750 --> 00:14:50,999 little theory exists that can help cookes 410 00:14:51,000 --> 00:14:53,279 know which gelling agents or coagulants 411 00:14:53,280 --> 00:14:55,739 will have synergistic effects. 412 00:14:55,740 --> 00:14:57,179 We can recommend several good 413 00:14:57,180 --> 00:14:59,249 combinations based on trial and error 414 00:14:59,250 --> 00:15:01,349 in our kitchen and others, 415 00:15:01,350 --> 00:15:03,839 but more information would be helpful. 416 00:15:03,840 --> 00:15:05,699 There are many possibilities out there 417 00:15:05,700 --> 00:15:07,889 waiting, awaiting discovery. 418 00:15:07,890 --> 00:15:09,719 Now, the reason why this quote is funny 419 00:15:09,720 --> 00:15:12,149 to me is because I work in soft robotics 420 00:15:12,150 --> 00:15:14,129 and essentially if you swap out Kitchin 421 00:15:14,130 --> 00:15:16,649 for lab and like one other noun, 422 00:15:16,650 --> 00:15:18,479 it's exactly what you would read in an 423 00:15:18,480 --> 00:15:19,480 academic paper. 424 00:15:22,590 --> 00:15:24,839 So I started on 425 00:15:24,840 --> 00:15:27,029 my hydrograph flight path by doing 426 00:15:27,030 --> 00:15:29,249 pectin, now, pectin was a good 427 00:15:29,250 --> 00:15:31,319 case because it actually illustrated 428 00:15:31,320 --> 00:15:33,209 what doesn't work. 429 00:15:33,210 --> 00:15:35,519 It's a Heidrich Cleide that 430 00:15:35,520 --> 00:15:37,799 is made of monomers that are very dense 431 00:15:37,800 --> 00:15:39,929 and very small, and they only have 432 00:15:39,930 --> 00:15:41,759 a single point where they joined to the 433 00:15:41,760 --> 00:15:42,659 other ones. 434 00:15:42,660 --> 00:15:44,909 And because there is no motion 435 00:15:44,910 --> 00:15:47,009 within the monomer 436 00:15:47,010 --> 00:15:48,629 and then there's the only the single 437 00:15:48,630 --> 00:15:49,559 point of connection. 438 00:15:49,560 --> 00:15:51,149 This makes a gel that's incredibly 439 00:15:51,150 --> 00:15:53,369 brittle and breaks apart under 440 00:15:53,370 --> 00:15:55,079 any kind of snapping force. 441 00:15:55,080 --> 00:15:57,299 So it's totally inappropriate for 442 00:15:57,300 --> 00:15:59,429 soft robots, but it is very 443 00:15:59,430 --> 00:16:01,350 good for learning about hydrate quoits. 444 00:16:02,700 --> 00:16:04,859 Number two, iteration or number two 445 00:16:04,860 --> 00:16:07,079 material exploration was 446 00:16:07,080 --> 00:16:09,449 I burned the gelatin. 447 00:16:09,450 --> 00:16:10,619 This was horrible. 448 00:16:10,620 --> 00:16:12,659 It smelled awful. 449 00:16:12,660 --> 00:16:14,819 And it was also really interesting 450 00:16:14,820 --> 00:16:16,649 because it reminded me that I was working 451 00:16:16,650 --> 00:16:18,269 with a protein. I think that we don't 452 00:16:18,270 --> 00:16:20,639 really think about how the materials 453 00:16:20,640 --> 00:16:22,799 within candy are things that 454 00:16:22,800 --> 00:16:25,199 are kind of surprising if you if you 455 00:16:25,200 --> 00:16:26,490 pick it apart a little more. 456 00:16:28,510 --> 00:16:30,939 So I made 457 00:16:30,940 --> 00:16:33,159 my third version, my third batch 458 00:16:33,160 --> 00:16:35,439 was too much gelatin, 459 00:16:35,440 --> 00:16:37,329 and this was yet another reminder that I 460 00:16:37,330 --> 00:16:38,769 was working with an animal product 461 00:16:38,770 --> 00:16:40,839 because the outcome looked and smelled 462 00:16:40,840 --> 00:16:43,059 like and taste like skin, 463 00:16:43,060 --> 00:16:45,129 which is not a desirable outcome. 464 00:16:45,130 --> 00:16:46,130 And Candy. 465 00:16:48,310 --> 00:16:50,829 So now we get to some slightly 466 00:16:50,830 --> 00:16:52,599 more successful gels. 467 00:16:52,600 --> 00:16:54,729 So this is a Jello brand 468 00:16:54,730 --> 00:16:55,899 gelatin. 469 00:16:55,900 --> 00:16:57,789 It was a good compromise between flavor 470 00:16:57,790 --> 00:16:59,259 and ease of handling because it already 471 00:16:59,260 --> 00:17:00,759 had sugar. It already had 472 00:17:01,870 --> 00:17:04,029 some some measurements done for 473 00:17:04,030 --> 00:17:06,098 you. But this is an example 474 00:17:06,099 --> 00:17:07,838 of an elastic hydrochloride. 475 00:17:07,839 --> 00:17:10,838 So what happens is 476 00:17:10,839 --> 00:17:13,059 collagen is a 477 00:17:13,060 --> 00:17:14,348 tri helix form. 478 00:17:14,349 --> 00:17:16,929 So it's inherently very, very stretchy. 479 00:17:16,930 --> 00:17:19,029 As you can imagine, the space between 480 00:17:19,030 --> 00:17:21,068 the three three components of the tri 481 00:17:21,069 --> 00:17:24,129 helix that stretch out 482 00:17:24,130 --> 00:17:25,899 well, when you make gelatin, it breaks 483 00:17:25,900 --> 00:17:27,338 into these single strands. 484 00:17:27,339 --> 00:17:29,469 So they're all like a free 485 00:17:29,470 --> 00:17:31,959 for the floating spiral. 486 00:17:31,960 --> 00:17:34,389 As you treat the 487 00:17:34,390 --> 00:17:37,539 as you treat the material, the gelatin, 488 00:17:37,540 --> 00:17:39,729 as you heat it up to a certain point, you 489 00:17:39,730 --> 00:17:41,439 actually start making these bonds again 490 00:17:41,440 --> 00:17:42,849 and you start getting tri helix 491 00:17:42,850 --> 00:17:45,159 sometimes, but not all of the time. 492 00:17:45,160 --> 00:17:46,869 So you end up with this combination where 493 00:17:46,870 --> 00:17:49,089 the tri helical parts are very 494 00:17:49,090 --> 00:17:51,369 stretchy. Still, all of the free 495 00:17:51,370 --> 00:17:53,839 floating ends that didn't make bonds 496 00:17:53,840 --> 00:17:56,319 end up in this tangled mess 497 00:17:56,320 --> 00:17:58,539 that a chemist told me is a lot 498 00:17:58,540 --> 00:18:00,489 like spaghetti. 499 00:18:00,490 --> 00:18:03,309 So one spaghetti is really hot and moist. 500 00:18:03,310 --> 00:18:05,559 It slides out and you can pull 501 00:18:05,560 --> 00:18:07,629 some strands away, but as soon 502 00:18:07,630 --> 00:18:09,579 as it gets cold, that tangled mess is 503 00:18:09,580 --> 00:18:11,679 stuck together and becomes a very, 504 00:18:11,680 --> 00:18:13,329 very solid thing. 505 00:18:13,330 --> 00:18:15,639 And so that relationship between 506 00:18:15,640 --> 00:18:18,099 the tangled mess and the tri helix 507 00:18:18,100 --> 00:18:19,419 is what gives gummies. 508 00:18:19,420 --> 00:18:21,370 They're like very, very. 509 00:18:23,860 --> 00:18:25,569 Signature texture. 510 00:18:28,120 --> 00:18:30,639 So this is an experiment 511 00:18:30,640 --> 00:18:33,039 that I made out of Gelin Gellin 512 00:18:33,040 --> 00:18:35,739 is a different kind of hydrolyzed it's a 513 00:18:35,740 --> 00:18:37,839 plant based and actually is a 514 00:18:37,840 --> 00:18:40,089 bacteria that grows on the bottom of lily 515 00:18:40,090 --> 00:18:42,309 pads in nature, which is pretty 516 00:18:42,310 --> 00:18:43,239 interesting. 517 00:18:43,240 --> 00:18:45,039 Now, the great thing about Gellin is that 518 00:18:45,040 --> 00:18:47,109 it makes this like extremely 519 00:18:47,110 --> 00:18:49,299 stretchy material, 520 00:18:49,300 --> 00:18:50,829 but at the same time it's very tender and 521 00:18:50,830 --> 00:18:51,849 soft. 522 00:18:51,850 --> 00:18:53,169 And when they were talking about like 523 00:18:53,170 --> 00:18:55,329 synergies between Heidrick Colditz, 524 00:18:55,330 --> 00:18:57,369 it's often combined with carrageenan to 525 00:18:57,370 --> 00:18:59,079 give it a stiffness. 526 00:19:00,370 --> 00:19:02,319 But the thing that's interesting about 527 00:19:02,320 --> 00:19:04,479 mixing hydrochloride is 528 00:19:04,480 --> 00:19:06,669 that you don't get an average between 529 00:19:06,670 --> 00:19:08,709 the two. When you start mixing them, you 530 00:19:08,710 --> 00:19:10,839 get entirely new behaviors that 531 00:19:10,840 --> 00:19:13,010 aren't reflected by the individual. 532 00:19:14,920 --> 00:19:17,799 So that is a future experiment. 533 00:19:17,800 --> 00:19:20,019 So I went back to the Jello brand gelatin 534 00:19:20,020 --> 00:19:22,509 recipe and then I was starting 535 00:19:22,510 --> 00:19:24,789 to be like, OK, this is this is working 536 00:19:24,790 --> 00:19:26,529 a little bit. So I'm going to go ahead 537 00:19:26,530 --> 00:19:28,900 and cast it in a soft, robotic mold. 538 00:19:30,160 --> 00:19:32,349 So this is dusted with 539 00:19:32,350 --> 00:19:33,849 cornstarch. 540 00:19:33,850 --> 00:19:36,129 Traditional gummies are actually made 541 00:19:36,130 --> 00:19:37,479 in cornstarch molds. 542 00:19:37,480 --> 00:19:39,789 They press metal shapes 543 00:19:39,790 --> 00:19:42,339 into corn starch and then pour the 544 00:19:42,340 --> 00:19:44,859 pour the gel directly into it. 545 00:19:44,860 --> 00:19:46,959 This is not food safe, so please 546 00:19:46,960 --> 00:19:47,959 don't do this in the future. 547 00:19:47,960 --> 00:19:50,289 So when I do future robots, I'm probably 548 00:19:50,290 --> 00:19:52,569 going to make silicone mold. 549 00:19:52,570 --> 00:19:54,639 But this mold was just one that I 550 00:19:54,640 --> 00:19:55,640 3D printed. 551 00:19:57,970 --> 00:20:00,309 So then my seventh 552 00:20:00,310 --> 00:20:02,679 experiment, making gelatin based 553 00:20:02,680 --> 00:20:05,199 making hydrochloride by stealth, was 554 00:20:05,200 --> 00:20:08,049 I used what they called a Haribo recipe. 555 00:20:08,050 --> 00:20:09,339 I don't know if that's how you pronounce 556 00:20:09,340 --> 00:20:11,559 it here, but the thing 557 00:20:11,560 --> 00:20:13,119 that was really interesting about this 558 00:20:13,120 --> 00:20:15,429 one was that they sub out 559 00:20:15,430 --> 00:20:17,859 regular white sugar for other sugars. 560 00:20:17,860 --> 00:20:20,019 You use glycol and submittal 561 00:20:20,020 --> 00:20:22,419 because it changes the moisture 562 00:20:22,420 --> 00:20:24,549 amounts in the mixture and you can 563 00:20:24,550 --> 00:20:26,499 eliminate some of the water when you 564 00:20:26,500 --> 00:20:27,940 switch the sugar source. 565 00:20:29,380 --> 00:20:31,569 So unsurprisingly, this one 566 00:20:31,570 --> 00:20:34,389 was the most delicious one that I made. 567 00:20:34,390 --> 00:20:36,699 And I think part of that was 568 00:20:36,700 --> 00:20:39,519 that it has added acid. 569 00:20:39,520 --> 00:20:42,069 So you can use citric acid or you can use 570 00:20:42,070 --> 00:20:43,689 I can't remember the name of the other 571 00:20:43,690 --> 00:20:45,969 acid, but a lot of different 572 00:20:45,970 --> 00:20:47,889 food acids can be used, but it really 573 00:20:47,890 --> 00:20:49,809 improves the flavor profile of your 574 00:20:49,810 --> 00:20:50,810 candy. 575 00:20:51,710 --> 00:20:53,889 Oops. So 576 00:20:53,890 --> 00:20:55,959 in the future, I have some plans for 577 00:20:55,960 --> 00:20:58,419 testing the assessment 578 00:20:58,420 --> 00:20:59,859 of these gels. 579 00:20:59,860 --> 00:21:02,109 And I didn't even get to it because 580 00:21:02,110 --> 00:21:04,269 the gel problem was so much deeper than 581 00:21:04,270 --> 00:21:05,529 I anticipated. 582 00:21:05,530 --> 00:21:07,839 So I'm planning to test 583 00:21:07,840 --> 00:21:09,639 the diameter, the elasticity, the 584 00:21:09,640 --> 00:21:11,739 flexibility and the durability of 585 00:21:11,740 --> 00:21:14,319 all my future iterations 586 00:21:14,320 --> 00:21:17,679 of the candy making. 587 00:21:17,680 --> 00:21:19,539 So in the store about material side, I 588 00:21:19,540 --> 00:21:20,769 went to economy candy. 589 00:21:20,770 --> 00:21:22,629 I bought a lot of different candy. 590 00:21:22,630 --> 00:21:24,639 I spent way too much money on candy. 591 00:21:24,640 --> 00:21:27,339 Also, I learned, don't eat your mistakes 592 00:21:27,340 --> 00:21:28,299 if you start working. 593 00:21:28,300 --> 00:21:29,410 And candy robotics. 594 00:21:38,320 --> 00:21:40,419 I ate three iterations and 595 00:21:40,420 --> 00:21:41,709 then I didn't feel so well. 596 00:21:43,840 --> 00:21:45,939 So I'm not a purist when it comes to soft 597 00:21:45,940 --> 00:21:48,129 robotics. I also love hybrid robots. 598 00:21:48,130 --> 00:21:50,349 So I did some tests 599 00:21:50,350 --> 00:21:52,509 where I drilled into chalky candies 600 00:21:52,510 --> 00:21:54,129 to see whether or not they could be 601 00:21:54,130 --> 00:21:56,349 spacer components 602 00:21:56,350 --> 00:21:58,599 which successful they were totally 603 00:21:58,600 --> 00:22:00,069 clean and had durable holes. 604 00:22:00,070 --> 00:22:01,779 And I could drill in a variety of sizes 605 00:22:03,610 --> 00:22:05,859 inspired by the octopus, which 606 00:22:05,860 --> 00:22:08,019 is a soft 607 00:22:08,020 --> 00:22:11,289 robot that runs a liquid across. 608 00:22:11,290 --> 00:22:12,699 I think it's a platinum plate and that 609 00:22:12,700 --> 00:22:13,959 that creates gas. 610 00:22:13,960 --> 00:22:15,699 I was like, oh, maybe I can make an 611 00:22:15,700 --> 00:22:18,399 explosive gas actuator 612 00:22:18,400 --> 00:22:21,279 by using pop rocks, but you can tell that 613 00:22:21,280 --> 00:22:23,559 it does not have any explosive action 614 00:22:23,560 --> 00:22:25,569 at all. Nothing happened. 615 00:22:25,570 --> 00:22:26,949 Just a very slow fizzle. 616 00:22:26,950 --> 00:22:29,439 So if I want to make a chemical 617 00:22:29,440 --> 00:22:30,999 reaction robot, I'll probably go to 618 00:22:31,000 --> 00:22:33,709 Mentos and Diet Coke because. 619 00:22:33,710 --> 00:22:35,689 The Internet has really well documented 620 00:22:35,690 --> 00:22:36,690 that that works well. 621 00:22:38,990 --> 00:22:41,179 Another thing I tried was buying 622 00:22:41,180 --> 00:22:43,039 pre purchased candy and then heating 623 00:22:43,040 --> 00:22:44,359 them. So one of the things that's 624 00:22:44,360 --> 00:22:46,039 interesting about this set of 625 00:22:46,040 --> 00:22:48,349 hydrochloride is that I'm exploring is 626 00:22:48,350 --> 00:22:50,509 that they are not a single state 627 00:22:50,510 --> 00:22:52,789 like one time reactions. 628 00:22:52,790 --> 00:22:54,259 They actually can revert. 629 00:22:54,260 --> 00:22:56,209 So you can keep them up and cool them 630 00:22:56,210 --> 00:22:58,489 down and it will cost the same chemical 631 00:22:58,490 --> 00:23:00,299 reaction set over and over again. 632 00:23:00,300 --> 00:23:02,389 So actually, there's lots of do overs if 633 00:23:02,390 --> 00:23:03,519 you're working and gummies. 634 00:23:04,550 --> 00:23:06,589 And so this these ones were microwaved. 635 00:23:06,590 --> 00:23:07,669 And I thought that would be an 636 00:23:07,670 --> 00:23:09,619 interesting option if people didn't want 637 00:23:09,620 --> 00:23:11,329 to make candy robots from scratch 638 00:23:12,830 --> 00:23:13,879 and then fruit leather. 639 00:23:13,880 --> 00:23:16,189 I was inspired by the fact that there's 640 00:23:16,190 --> 00:23:18,349 a lot of soft robots that are made out 641 00:23:18,350 --> 00:23:20,869 of airtight fabric that doesn't flex 642 00:23:20,870 --> 00:23:22,549 but just holds air. 643 00:23:22,550 --> 00:23:24,919 So this is me using a compressor 644 00:23:24,920 --> 00:23:27,529 to pump air into 645 00:23:27,530 --> 00:23:28,759 some fruit leather. 646 00:23:28,760 --> 00:23:31,189 And it also made a pretty good seal 647 00:23:31,190 --> 00:23:33,139 just with its own stickiness, which was a 648 00:23:33,140 --> 00:23:34,140 surprise. 649 00:23:36,730 --> 00:23:38,799 So I said something before 650 00:23:38,800 --> 00:23:41,019 about vegetable matter, a lot 651 00:23:41,020 --> 00:23:42,669 of the things I used for pulling force 652 00:23:42,670 --> 00:23:44,889 that were made of candy didn't 653 00:23:44,890 --> 00:23:46,569 hold up to the amount of force I needed 654 00:23:46,570 --> 00:23:48,339 to cause the situation. 655 00:23:48,340 --> 00:23:50,679 So I tried dried mangoes, 656 00:23:50,680 --> 00:23:52,059 which are excellent. 657 00:23:52,060 --> 00:23:54,069 You can't they they fit the taste 658 00:23:54,070 --> 00:23:56,649 profile. They go well with candy, 659 00:23:56,650 --> 00:23:58,839 but you can tie it in knots, you can find 660 00:23:58,840 --> 00:23:59,469 things off. 661 00:23:59,470 --> 00:24:02,199 You can use them as cable control's 662 00:24:02,200 --> 00:24:04,209 excellent material for the space. 663 00:24:04,210 --> 00:24:07,119 And then also there's these licorice 664 00:24:07,120 --> 00:24:09,189 vines that are hollow so 665 00:24:09,190 --> 00:24:10,540 you can pull things through them. 666 00:24:12,830 --> 00:24:15,019 They're also airtight, so you 667 00:24:15,020 --> 00:24:17,209 can use them as tubing, so 668 00:24:17,210 --> 00:24:18,710 if you want to make a pneumatic system. 669 00:24:20,720 --> 00:24:22,849 So I'm going to move 670 00:24:22,850 --> 00:24:25,279 on now to my next plan 671 00:24:25,280 --> 00:24:27,139 in this process was that I was going to 672 00:24:27,140 --> 00:24:29,150 do an iteration of a wide field. 673 00:24:30,170 --> 00:24:32,359 So usually what I do is 674 00:24:32,360 --> 00:24:34,969 I take all of my promising materials 675 00:24:34,970 --> 00:24:36,679 and I'll do a single iteration of each of 676 00:24:36,680 --> 00:24:39,589 the problem, promising materials. 677 00:24:39,590 --> 00:24:42,350 But the thing is that in this case, 678 00:24:43,460 --> 00:24:45,679 I ran out of time, like the 679 00:24:45,680 --> 00:24:47,629 Dell problem was so much bigger than I 680 00:24:47,630 --> 00:24:49,729 ever thought it would be that I 681 00:24:49,730 --> 00:24:51,169 only got through one iteration. 682 00:24:51,170 --> 00:24:52,969 So I wanted to let you know that right 683 00:24:52,970 --> 00:24:55,399 now. So in addition to iterating 684 00:24:55,400 --> 00:24:57,679 on all the promising materials, I wanted 685 00:24:57,680 --> 00:24:59,449 to do one iteration of each control 686 00:24:59,450 --> 00:25:01,369 method that is popular in soft robotics. 687 00:25:01,370 --> 00:25:03,559 So that's cable controlled pneumatic 688 00:25:03,560 --> 00:25:06,379 hydraulic and chemical reaction. 689 00:25:06,380 --> 00:25:07,999 The only one I got through is cable 690 00:25:08,000 --> 00:25:10,249 controlled, but this is my first 691 00:25:10,250 --> 00:25:12,679 iteration of a cable controlled 692 00:25:12,680 --> 00:25:13,760 robot made up candy. 693 00:25:23,210 --> 00:25:25,279 So it kind 694 00:25:25,280 --> 00:25:26,280 of works, 695 00:25:28,070 --> 00:25:30,229 and then when I do my 696 00:25:30,230 --> 00:25:31,969 first iterations, I have to go back and 697 00:25:31,970 --> 00:25:33,349 assess what worked and didn't work. 698 00:25:33,350 --> 00:25:34,909 And one of the things that I found about 699 00:25:34,910 --> 00:25:37,189 a cable controlled candy robot 700 00:25:37,190 --> 00:25:39,349 was I wanted a lubricant, 701 00:25:39,350 --> 00:25:41,209 so I tried oil. 702 00:25:41,210 --> 00:25:42,319 Oil was terrible. 703 00:25:42,320 --> 00:25:43,849 It was a solvent. It worked a lot like 704 00:25:43,850 --> 00:25:45,829 water. Everything started breaking down 705 00:25:45,830 --> 00:25:46,969 and getting stickier. 706 00:25:46,970 --> 00:25:48,739 And then I was like, oops, of course, 707 00:25:48,740 --> 00:25:50,239 sugar dissolves in moisture. 708 00:25:50,240 --> 00:25:51,139 Should have thought of that. 709 00:25:51,140 --> 00:25:52,729 But sometimes when you switch from one 710 00:25:52,730 --> 00:25:54,379 domain to the next, the things you know 711 00:25:54,380 --> 00:25:55,880 don't transfer with it. 712 00:25:57,550 --> 00:25:59,419 I tried cornstarch because I was like, 713 00:25:59,420 --> 00:26:01,369 dry lubricant seems the way to go. 714 00:26:01,370 --> 00:26:03,289 It got a little gummy as it rolled. 715 00:26:03,290 --> 00:26:05,359 So it was good short term lubricant, 716 00:26:05,360 --> 00:26:07,549 but it you couldn't 717 00:26:07,550 --> 00:26:10,219 have high cycles with confectioner's 718 00:26:10,220 --> 00:26:11,749 sugar. It was the most promising. 719 00:26:11,750 --> 00:26:14,329 So confectioners sugar is a mixture of 720 00:26:14,330 --> 00:26:16,609 very, very fine sugar with cornstarch. 721 00:26:17,750 --> 00:26:19,519 The second iteration would be a narrowed 722 00:26:19,520 --> 00:26:21,409 field based on my first set of 723 00:26:21,410 --> 00:26:23,410 iterations, TBD. 724 00:26:24,770 --> 00:26:27,019 And the same thing is true to getting 725 00:26:27,020 --> 00:26:29,029 the final soft robot. 726 00:26:29,030 --> 00:26:31,159 So repeat iterations till your specs are 727 00:26:31,160 --> 00:26:33,319 met. No excess and 728 00:26:33,320 --> 00:26:35,000 iterate yes, you're done 729 00:26:36,680 --> 00:26:37,680 TBD. 730 00:26:38,150 --> 00:26:40,549 So the thing, though, about 731 00:26:40,550 --> 00:26:42,589 this whole conversation is 732 00:26:43,820 --> 00:26:46,279 I really wanted 733 00:26:46,280 --> 00:26:48,079 to take the risk of showing work in 734 00:26:48,080 --> 00:26:50,509 progress and showing unfinished work. 735 00:26:50,510 --> 00:26:52,309 And the reason why I wanted to show 736 00:26:52,310 --> 00:26:53,809 unfinished work is because it 737 00:26:53,810 --> 00:26:55,909 incentivized me to show 738 00:26:55,910 --> 00:26:57,529 you my failures. 739 00:26:57,530 --> 00:26:59,179 I think that innovators spend a lot of 740 00:26:59,180 --> 00:27:00,799 time talking about how we're willing to 741 00:27:00,800 --> 00:27:02,929 fail. But what happens is you either get 742 00:27:02,930 --> 00:27:04,339 all the way to the end or you give up on 743 00:27:04,340 --> 00:27:06,469 the project and there's not any air time 744 00:27:06,470 --> 00:27:08,569 for whether or not you succeed, what your 745 00:27:08,570 --> 00:27:10,430 failures were and. 746 00:27:19,770 --> 00:27:22,049 And so and the other 747 00:27:22,050 --> 00:27:24,299 thing I was interested in, 748 00:27:24,300 --> 00:27:26,999 why I wanted to give this talk is because 749 00:27:27,000 --> 00:27:29,249 I'm talking about candy robots today, 750 00:27:29,250 --> 00:27:31,529 but my deeper interest is 751 00:27:31,530 --> 00:27:33,599 in systems and processes 752 00:27:33,600 --> 00:27:36,719 and particularly how a robust process 753 00:27:36,720 --> 00:27:38,819 can give you a framework in 754 00:27:38,820 --> 00:27:40,529 which you can succeed on something that's 755 00:27:40,530 --> 00:27:41,819 never been done before. 756 00:27:41,820 --> 00:27:44,099 And also how being in a good 757 00:27:44,100 --> 00:27:46,409 process means when you get stuck, there's 758 00:27:46,410 --> 00:27:47,669 a way to correct for that. 759 00:27:47,670 --> 00:27:49,799 There's a way to get unstuck for me in 760 00:27:49,800 --> 00:27:51,659 this process. When I get stuck, I can 761 00:27:51,660 --> 00:27:53,609 always go back to an expert or find a new 762 00:27:53,610 --> 00:27:55,979 expert. And it means that I never 763 00:27:55,980 --> 00:27:58,049 hit a bottleneck where I stop innovating. 764 00:27:59,940 --> 00:28:02,519 So hopefully people will 765 00:28:02,520 --> 00:28:04,349 think about processes for working on 766 00:28:04,350 --> 00:28:06,119 emerging technologies that are from their 767 00:28:06,120 --> 00:28:08,249 toolkit. I'd love to hear about that or 768 00:28:08,250 --> 00:28:09,869 use this one to work on your own 769 00:28:09,870 --> 00:28:10,870 problems. 770 00:28:11,820 --> 00:28:13,319 Some final thoughts. 771 00:28:13,320 --> 00:28:14,879 I think design thinking and other 772 00:28:14,880 --> 00:28:17,189 processes yield effective pathways 773 00:28:17,190 --> 00:28:19,199 through the unknown. 774 00:28:19,200 --> 00:28:21,329 I think people in general should work on 775 00:28:21,330 --> 00:28:23,339 emerging fields wherever you find them. 776 00:28:23,340 --> 00:28:24,839 It's where you find the most exciting 777 00:28:24,840 --> 00:28:26,399 problems and the most interesting 778 00:28:26,400 --> 00:28:27,839 opportunities. 779 00:28:27,840 --> 00:28:29,909 And then in general, I want people to 780 00:28:29,910 --> 00:28:31,649 reach out. If you want to share or 781 00:28:31,650 --> 00:28:34,319 collaborate about edible robotics, 782 00:28:34,320 --> 00:28:36,599 I could use more help in chemistry 783 00:28:36,600 --> 00:28:38,849 control systems for SOF actuators 784 00:28:38,850 --> 00:28:40,529 is a is a juicy problem. 785 00:28:40,530 --> 00:28:42,569 Or you tell me maybe you're an expert in 786 00:28:42,570 --> 00:28:44,189 something related to this that I haven't 787 00:28:44,190 --> 00:28:45,190 even thought of yet. 788 00:28:46,140 --> 00:28:47,969 And also, please reach out if you want to 789 00:28:47,970 --> 00:28:49,709 talk process. 790 00:28:49,710 --> 00:28:51,779 So this is me and where 791 00:28:51,780 --> 00:28:53,039 you can find me on the Internet. 792 00:28:53,040 --> 00:28:54,539 And if you want to talk to me about the 793 00:28:54,540 --> 00:28:57,449 other things she introduced me before, 794 00:28:57,450 --> 00:28:59,849 I like to talk about soft robotics, 795 00:28:59,850 --> 00:29:02,009 spacesuits, costumes and puppets. 796 00:29:02,010 --> 00:29:03,989 So thank you. 797 00:29:03,990 --> 00:29:04,990 Thank you.