1 00:00:04,200 --> 00:00:12,160 Speaker 1: Get in touch with technology with tex Stuff from dot Com. 2 00:00:12,200 --> 00:00:14,440 Speaker 1: Hey there, everyone, and welcome to tech Stuff. I am 3 00:00:14,760 --> 00:00:18,400 Speaker 1: Jonathan Strickland, and today but I want to talk about 4 00:00:18,800 --> 00:00:21,840 Speaker 1: the DARPA Robotics Challenge. You may remember I did an 5 00:00:21,840 --> 00:00:25,520 Speaker 1: episode where I interviewed someone from DARPA to talk about 6 00:00:25,680 --> 00:00:29,280 Speaker 1: the robots for US part of the DARPA Robotics Challenge, 7 00:00:29,880 --> 00:00:34,000 Speaker 1: and since then the challenge actually occurred. It happened. I'm 8 00:00:34,040 --> 00:00:37,640 Speaker 1: recording this on June eleven, two thousand fifteen, and the 9 00:00:37,640 --> 00:00:42,319 Speaker 1: previous weekend is when they held the DARPA Robotics Challenge Finals. UH. 10 00:00:42,400 --> 00:00:45,560 Speaker 1: This was the last step in a journey to create 11 00:00:45,600 --> 00:00:50,360 Speaker 1: a robot that could respond to a simulated emergency situation, 12 00:00:51,360 --> 00:00:57,120 Speaker 1: and it was a pretty interesting event. So to start off, 13 00:00:57,160 --> 00:01:01,080 Speaker 1: we gotta talk about DARPA itself. Remember that's the research 14 00:01:01,080 --> 00:01:04,800 Speaker 1: and development arm of the Department of Defense is technically 15 00:01:04,880 --> 00:01:10,280 Speaker 1: the US Defense Advanced Research Projects Agency UH, and DARPA 16 00:01:10,319 --> 00:01:14,080 Speaker 1: has been responsible for a lot of cutting edge research, 17 00:01:14,760 --> 00:01:18,640 Speaker 1: largely in the mode of military use. But we have 18 00:01:18,760 --> 00:01:22,880 Speaker 1: seen the benefits of that research hit us in other ways. 19 00:01:22,959 --> 00:01:28,160 Speaker 1: Besides you know the military applications. So for example, our 20 00:01:28,240 --> 00:01:32,880 Speaker 1: PA net, the predecessor to the Internet was the platform 21 00:01:33,000 --> 00:01:36,200 Speaker 1: upon which the technologies that allow the Internet to work 22 00:01:36,440 --> 00:01:39,320 Speaker 1: were developed, and that was a DARPA project that was 23 00:01:39,360 --> 00:01:43,200 Speaker 1: back when they were first just called ARPA rather than DARPA. 24 00:01:43,440 --> 00:01:47,440 Speaker 1: But we wouldn't have the Internet, at least not the 25 00:01:47,440 --> 00:01:50,000 Speaker 1: way we have it today, if it were not for 26 00:01:50,040 --> 00:01:52,880 Speaker 1: this agency. So the agency has done a lot of 27 00:01:52,920 --> 00:01:55,640 Speaker 1: stuff that has benefited us in many ways. I'll talk 28 00:01:55,680 --> 00:01:58,480 Speaker 1: about another one a little bit later on in the podcast. 29 00:01:58,600 --> 00:02:03,200 Speaker 1: Kind of drop parallels to what happened with the robotics challenge. 30 00:02:04,160 --> 00:02:07,960 Speaker 1: So here we have this agency. They're all about research 31 00:02:08,000 --> 00:02:11,720 Speaker 1: and development. They don't actually have lots of labs with 32 00:02:11,760 --> 00:02:15,520 Speaker 1: scientists working in them. Now, what this agency does is 33 00:02:15,560 --> 00:02:20,240 Speaker 1: it sends out proposals to other groups and asks them 34 00:02:20,360 --> 00:02:25,120 Speaker 1: to help work on various projects. Sometimes they hold challenges. 35 00:02:25,600 --> 00:02:28,520 Speaker 1: In these challenges, it's kind of an open invitation for 36 00:02:28,639 --> 00:02:34,639 Speaker 1: any group that can participate to uh to compete in 37 00:02:34,680 --> 00:02:37,640 Speaker 1: some way, and the winner usually gets some sort of 38 00:02:37,639 --> 00:02:41,520 Speaker 1: cash prize. And so often the cash prize will be 39 00:02:41,600 --> 00:02:45,200 Speaker 1: less than the investment needed to actually participate in the 40 00:02:45,280 --> 00:02:48,520 Speaker 1: competition to build out the thing that you need to 41 00:02:48,520 --> 00:02:52,639 Speaker 1: make for us to to win, but it's still very 42 00:02:52,680 --> 00:02:57,600 Speaker 1: important uh and prestigious event to be part of if 43 00:02:57,639 --> 00:03:00,520 Speaker 1: you are capable of competing. Now we're talking king mostly 44 00:03:00,560 --> 00:03:05,160 Speaker 1: about research departments engineering departments in universities, but there are 45 00:03:05,160 --> 00:03:10,000 Speaker 1: also some private institutions, not just university type things, but 46 00:03:10,120 --> 00:03:13,840 Speaker 1: actual research and development labs that will participate in DARPA 47 00:03:13,919 --> 00:03:19,440 Speaker 1: projects as well. And it's collectively that these teams were 48 00:03:20,360 --> 00:03:23,040 Speaker 1: kind of they're the ones that are pushing the technology forward, 49 00:03:23,080 --> 00:03:26,200 Speaker 1: and DARK is kind of the facilitator. They're offering up 50 00:03:26,240 --> 00:03:32,440 Speaker 1: the possibility for people to actually uh participate, and not 51 00:03:32,480 --> 00:03:35,960 Speaker 1: only the possibility, but the goals. They define the goals 52 00:03:36,000 --> 00:03:38,320 Speaker 1: that have to be met, and as it turns out, 53 00:03:39,120 --> 00:03:42,520 Speaker 1: defining those goals is a very important part of these 54 00:03:42,600 --> 00:03:47,680 Speaker 1: challenges because you have to tell the engineers what it 55 00:03:47,840 --> 00:03:51,000 Speaker 1: is they need to be able to do before they 56 00:03:51,000 --> 00:03:53,400 Speaker 1: can build the stuff that does it. If if you 57 00:03:53,440 --> 00:03:57,360 Speaker 1: have a very vague description of what needs to be done, 58 00:03:58,120 --> 00:04:03,000 Speaker 1: the possibilities of achieving it are so varied that it 59 00:04:03,040 --> 00:04:07,440 Speaker 1: can often paralyze a project nothing gets done. But if 60 00:04:07,440 --> 00:04:11,200 Speaker 1: you make it very specific, then it narrows down the 61 00:04:11,240 --> 00:04:14,840 Speaker 1: options and it gives more focus to the project. So 62 00:04:15,240 --> 00:04:20,520 Speaker 1: the Darker Robotics project itself was inspired by a real 63 00:04:20,680 --> 00:04:26,480 Speaker 1: world disaster, specifically the Fukushima disaster in Japan. Now, this 64 00:04:26,680 --> 00:04:32,560 Speaker 1: was the nuclear facility, the nuclear reactor that was flooded 65 00:04:33,040 --> 00:04:39,320 Speaker 1: and then suffered some massive problems because of damage to 66 00:04:39,360 --> 00:04:45,000 Speaker 1: the infrastructure. And we're talking about a nuclear facility, radiation 67 00:04:45,160 --> 00:04:49,320 Speaker 1: is a factor, so it's a very dangerous environment to 68 00:04:49,400 --> 00:04:56,000 Speaker 1: send responders to. Darpest approach was to create a scenario 69 00:04:56,680 --> 00:05:00,280 Speaker 1: in which a robot would need to go through an 70 00:05:00,360 --> 00:05:04,440 Speaker 1: environment similar to that that was at Fukushima and be 71 00:05:04,520 --> 00:05:09,640 Speaker 1: able to perform the tasks necessary to help avert true 72 00:05:09,760 --> 00:05:14,800 Speaker 1: devastation and catastrophe without putting people at risk. So the 73 00:05:14,839 --> 00:05:16,920 Speaker 1: idea is that you can use a robot which is 74 00:05:16,960 --> 00:05:21,800 Speaker 1: not going to respond to radiation the way a living 75 00:05:21,920 --> 00:05:26,080 Speaker 1: organism would and be able to actually carry out the 76 00:05:26,120 --> 00:05:30,440 Speaker 1: stuff you need to prevent a really huge problem from 77 00:05:30,480 --> 00:05:35,599 Speaker 1: getting worse. So uh, some people have referred to the 78 00:05:35,680 --> 00:05:41,200 Speaker 1: Darker Robotics Challenge as the Robolympics because it was a 79 00:05:41,200 --> 00:05:44,200 Speaker 1: series of tasks that a robot has to complete and 80 00:05:44,240 --> 00:05:47,479 Speaker 1: it had to all be the same. Robot teams could 81 00:05:47,520 --> 00:05:52,040 Speaker 1: not build individual robots designed to complete specific tasks. If 82 00:05:52,080 --> 00:05:55,400 Speaker 1: they could, it would be way easier of a challenge 83 00:05:55,720 --> 00:05:58,159 Speaker 1: because most of the robots that are in use today 84 00:05:58,720 --> 00:06:03,360 Speaker 1: have very narrow parameters. There's just maybe a single task, 85 00:06:04,000 --> 00:06:06,560 Speaker 1: or perhaps just a couple of small, you know, a 86 00:06:06,600 --> 00:06:08,880 Speaker 1: small range of tasks that the robot has to do. 87 00:06:09,760 --> 00:06:12,039 Speaker 1: It doesn't have to do anything outside of that. So 88 00:06:12,080 --> 00:06:14,960 Speaker 1: when you're designing the robot, you just say, all right, 89 00:06:15,120 --> 00:06:18,360 Speaker 1: what design elements will allow the robot to do what 90 00:06:18,440 --> 00:06:21,440 Speaker 1: it has to do. Anything that's outside of that is unnecessary, 91 00:06:21,480 --> 00:06:24,520 Speaker 1: we will not do it. Take the roomba as an example. 92 00:06:24,560 --> 00:06:27,479 Speaker 1: It's a great example. It's a robot the vacuums that 93 00:06:27,560 --> 00:06:31,599 Speaker 1: cleans floors. Anything that's not necessary for the rumbo to 94 00:06:31,600 --> 00:06:35,680 Speaker 1: do that is put aside. You'd want to make sure 95 00:06:35,720 --> 00:06:40,200 Speaker 1: that all the design elements in your robot complement its purpose. 96 00:06:41,360 --> 00:06:45,320 Speaker 1: But if you have to complete a lot of different tasks, 97 00:06:45,400 --> 00:06:49,120 Speaker 1: and those tasks are not always similar to one another, 98 00:06:50,040 --> 00:06:57,120 Speaker 1: you complicate the whole design process by incredible factors. It's 99 00:06:57,160 --> 00:07:02,160 Speaker 1: hard to even express how much more difficult this is. Now, 100 00:07:02,240 --> 00:07:05,440 Speaker 1: if you've watched any of the videos that came out 101 00:07:05,480 --> 00:07:08,960 Speaker 1: of the Darker Robotics Challenge, you probably saw at least 102 00:07:08,960 --> 00:07:12,880 Speaker 1: one or two that were montages of robots falling over. 103 00:07:13,640 --> 00:07:15,640 Speaker 1: And it's a little funny to watch. You see these 104 00:07:15,840 --> 00:07:21,880 Speaker 1: big robots just topple over sometimes apparently for no reason. 105 00:07:22,080 --> 00:07:24,480 Speaker 1: They just they just they're standing and then they collapse. 106 00:07:25,240 --> 00:07:27,720 Speaker 1: And there's something comic about that, you know that you 107 00:07:27,800 --> 00:07:32,200 Speaker 1: can't deny that it's funny, but it also demonstrates that 108 00:07:32,280 --> 00:07:34,880 Speaker 1: these robots are trying to do something that while we 109 00:07:35,000 --> 00:07:38,640 Speaker 1: as human beings might find easy, it's a real challenge 110 00:07:38,920 --> 00:07:44,960 Speaker 1: from a robotics standpoint. So this Darker Robotics Challenge had 111 00:07:45,040 --> 00:07:49,880 Speaker 1: three phases. The first one was a virtual challenge, which 112 00:07:50,560 --> 00:07:54,440 Speaker 1: in which teams were to design software that would allow 113 00:07:54,560 --> 00:07:58,640 Speaker 1: a virtual robot to complete um certain tasks within a 114 00:07:58,720 --> 00:08:03,000 Speaker 1: virtual environment. So there were no real robotics here, but 115 00:08:03,080 --> 00:08:05,840 Speaker 1: it was a test to see if if the teams 116 00:08:05,840 --> 00:08:09,320 Speaker 1: could actually build the software necessary to make the robot 117 00:08:09,400 --> 00:08:13,560 Speaker 1: do the tasks that needed to be done. The next 118 00:08:14,200 --> 00:08:17,400 Speaker 1: had a trial where it was a physical trial where 119 00:08:17,440 --> 00:08:22,120 Speaker 1: where teams actually had to uh place a real robot 120 00:08:22,240 --> 00:08:25,600 Speaker 1: through too, you know, go to the next level. And 121 00:08:25,640 --> 00:08:27,480 Speaker 1: then you had the finals, which were the ones that 122 00:08:27,880 --> 00:08:31,440 Speaker 1: happened most recently. UH and I'll go through each of 123 00:08:31,480 --> 00:08:33,800 Speaker 1: the tasks that they had to do in just a second. 124 00:08:34,720 --> 00:08:37,000 Speaker 1: If we look back at the virtual challenge, there were 125 00:08:37,040 --> 00:08:40,199 Speaker 1: twenty six teams that were part of it to build 126 00:08:40,200 --> 00:08:44,079 Speaker 1: that software. The robot they were controlling was a virtual 127 00:08:44,240 --> 00:08:50,000 Speaker 1: Atlas robot, which is a humanoid robot and was a 128 00:08:50,080 --> 00:08:52,400 Speaker 1: popular choice for a lot of the teams. When they 129 00:08:52,440 --> 00:08:56,720 Speaker 1: moved on to the more physical trials, they still would 130 00:08:56,840 --> 00:08:58,880 Speaker 1: use the Atlas robots. The the challenge, by the way, 131 00:08:58,880 --> 00:09:01,600 Speaker 1: did not require of the teams to build their own 132 00:09:01,679 --> 00:09:05,600 Speaker 1: robots from scratch. They could use pre existing robots, but 133 00:09:05,600 --> 00:09:08,160 Speaker 1: they had to program them so that they could complete 134 00:09:08,160 --> 00:09:11,600 Speaker 1: the necessary tasks. So not every robot out there would 135 00:09:11,600 --> 00:09:13,600 Speaker 1: have been a good fit for this. Let's go back 136 00:09:13,640 --> 00:09:16,000 Speaker 1: to the roomba. The roombo would have been a terrible 137 00:09:16,080 --> 00:09:19,280 Speaker 1: choice to enter into the Darbo Robotics Challenge. Would have 138 00:09:19,320 --> 00:09:23,880 Speaker 1: been funny, but it wouldn't have worked. So uh. Here 139 00:09:23,920 --> 00:09:27,960 Speaker 1: are some of the virtual tasks that the virtual robots 140 00:09:28,160 --> 00:09:31,880 Speaker 1: had to complete using this software. They had to walk 141 00:09:31,920 --> 00:09:37,280 Speaker 1: through an area of uneven ground that had debris included there, 142 00:09:37,280 --> 00:09:40,480 Speaker 1: so simulated debris had to have a software that would 143 00:09:40,480 --> 00:09:43,880 Speaker 1: allow the robot to maintain its balance even while tracking 144 00:09:43,880 --> 00:09:48,160 Speaker 1: a path to a specific destination. Through this this uneven terrain, 145 00:09:48,840 --> 00:09:51,480 Speaker 1: they had to attach a hose to a spigot and 146 00:09:51,640 --> 00:09:56,560 Speaker 1: turn a valve. Uh. They also were given some artificial 147 00:09:56,640 --> 00:10:01,400 Speaker 1: limits because and an emergency WHO situation, you cannot always 148 00:10:01,480 --> 00:10:07,560 Speaker 1: count on your communication lines being perfectly sound, such as 149 00:10:07,559 --> 00:10:11,840 Speaker 1: that the Fukushima disaster. You need to be sure that 150 00:10:11,920 --> 00:10:14,360 Speaker 1: your robot can contend with the fact that there might 151 00:10:14,400 --> 00:10:17,160 Speaker 1: be lag between when you can issue a command and 152 00:10:17,160 --> 00:10:20,720 Speaker 1: when the robot is ready for the next command. UH. 153 00:10:21,000 --> 00:10:23,280 Speaker 1: It meant that some of the teams started to really 154 00:10:23,320 --> 00:10:28,120 Speaker 1: look at ways to create a semi autonomous robot. So 155 00:10:28,200 --> 00:10:31,440 Speaker 1: some of the teams built robots that were capable of 156 00:10:32,480 --> 00:10:39,040 Speaker 1: taking on some tasks autonomously. Uh. Some teams avoided that entirely. 157 00:10:39,280 --> 00:10:42,560 Speaker 1: They focused on using the robots as a direct extension 158 00:10:42,840 --> 00:10:47,280 Speaker 1: of the controls that the humans were behind. So, in 159 00:10:47,280 --> 00:10:51,120 Speaker 1: other words, you might think of those robots as enormous 160 00:10:51,360 --> 00:10:56,920 Speaker 1: technological puppets. The puppets would respond to direct human commands 161 00:10:57,640 --> 00:11:00,160 Speaker 1: and wouldn't be able to do anything on their own. 162 00:11:00,200 --> 00:11:03,559 Speaker 1: They didn't have any autonomy. There were other ones that, 163 00:11:03,679 --> 00:11:06,440 Speaker 1: you know, you you could have semi autonomy. Only show 164 00:11:06,480 --> 00:11:08,240 Speaker 1: it a set of stairs and you send it the 165 00:11:08,280 --> 00:11:12,240 Speaker 1: command to climb those stairs, and it could do the rest. 166 00:11:12,280 --> 00:11:14,360 Speaker 1: It could calculate how high it needed to put its 167 00:11:14,400 --> 00:11:18,240 Speaker 1: feet and how to shift its weight, that sort of stuff. UH, 168 00:11:18,240 --> 00:11:20,600 Speaker 1: and there were a lot of different strategies employed and 169 00:11:20,880 --> 00:11:24,599 Speaker 1: to varying degrees of success. It wasn't like the autonomous 170 00:11:24,720 --> 00:11:28,720 Speaker 1: robots automatically were better than the human controlled or vice versa. 171 00:11:28,760 --> 00:11:32,600 Speaker 1: There was actually it all depended on the implementation. So 172 00:11:32,800 --> 00:11:40,440 Speaker 1: that virtual trial ended up having one team winning pretty 173 00:11:40,480 --> 00:11:45,360 Speaker 1: pretty decisively. UM. It was the Institute for Human and 174 00:11:45,440 --> 00:11:51,120 Speaker 1: Machine Cognition, which is UH, an organization in Pensacola, Florida, 175 00:11:51,360 --> 00:11:54,800 Speaker 1: and eight other teams qualified for the trials that we're 176 00:11:54,840 --> 00:11:59,920 Speaker 1: going to be held in December. Now, by the time 177 00:12:00,120 --> 00:12:04,520 Speaker 1: those trials happened, some of the teams had merged UH 178 00:12:04,520 --> 00:12:08,800 Speaker 1: and some other teams were coming into it. Six of 179 00:12:08,840 --> 00:12:11,600 Speaker 1: the teams would go on to compete at the trials 180 00:12:11,600 --> 00:12:16,040 Speaker 1: that were held at the Homestead Miami Speedway, and they 181 00:12:16,080 --> 00:12:19,000 Speaker 1: had eight tasks they had to complete that were similar 182 00:12:19,040 --> 00:12:23,280 Speaker 1: to the ones in the Darker Robotics Final and those 183 00:12:23,320 --> 00:12:28,679 Speaker 1: tasks included manning a vehicle, um walking through uneven terrain 184 00:12:28,920 --> 00:12:32,960 Speaker 1: or otherwise moving through uneven terrain. Robots did not have 185 00:12:33,080 --> 00:12:36,640 Speaker 1: to be bipedal. They could move around on however many 186 00:12:36,679 --> 00:12:38,760 Speaker 1: limbs they had. They just had to be also able 187 00:12:38,800 --> 00:12:41,760 Speaker 1: to do these other tasks. Uh. They had to be 188 00:12:41,840 --> 00:12:43,800 Speaker 1: able to climb a ladder. They had to be able 189 00:12:43,840 --> 00:12:46,400 Speaker 1: to remove debris from a door and then open the door. 190 00:12:46,640 --> 00:12:49,520 Speaker 1: They had to be able to break through a wall. Uh, 191 00:12:49,559 --> 00:12:51,719 Speaker 1: they had to be able to handle valves and to 192 00:12:51,880 --> 00:12:55,960 Speaker 1: attach a hose to a spigot. The winning team of 193 00:12:56,120 --> 00:13:00,679 Speaker 1: that set of trials was a group called Shaft spelled 194 00:13:00,960 --> 00:13:05,199 Speaker 1: s C H A f T. This was a group 195 00:13:05,200 --> 00:13:09,160 Speaker 1: out of Japan. Now, when this group was competing, they 196 00:13:09,200 --> 00:13:12,880 Speaker 1: were this little independent group. But then they got acquired 197 00:13:12,960 --> 00:13:16,199 Speaker 1: by a little company that we've talked about several times 198 00:13:16,200 --> 00:13:19,520 Speaker 1: on this podcast. You know him, you may or may 199 00:13:19,520 --> 00:13:23,240 Speaker 1: not love him Google. So just before the trials, Google 200 00:13:23,240 --> 00:13:28,679 Speaker 1: had acquired uh, this this this group, this organization. Now 201 00:13:28,760 --> 00:13:32,320 Speaker 1: Shaft did really well, uh, which you know makes sense. 202 00:13:32,320 --> 00:13:35,040 Speaker 1: I mean, you've you've heard the theme song. But out 203 00:13:35,120 --> 00:13:38,240 Speaker 1: of thirty two points that it could possibly score, it 204 00:13:38,520 --> 00:13:43,320 Speaker 1: earned twenty seven, So pretty good score for for a 205 00:13:43,440 --> 00:13:46,920 Speaker 1: robot and also that was seven points more than the 206 00:13:46,960 --> 00:13:51,040 Speaker 1: next closest competitor, which was from I H M C, 207 00:13:51,520 --> 00:13:55,800 Speaker 1: the same group that won the virtual trial earlier. Third 208 00:13:55,880 --> 00:14:01,520 Speaker 1: place was Tartan Rescue from Carnegie Melon And that will 209 00:14:01,559 --> 00:14:03,200 Speaker 1: be important in a second to when we get into 210 00:14:03,200 --> 00:14:08,120 Speaker 1: the finals. Now, Shaft did not go on to compete 211 00:14:08,120 --> 00:14:11,719 Speaker 1: in the DARPA Robotics Finals because Google announced that they 212 00:14:11,720 --> 00:14:17,679 Speaker 1: were refocusing the team, dedicating it to actual commercial Google projects, 213 00:14:18,240 --> 00:14:23,320 Speaker 1: So uh, they were no longer being dedicated to this challenge. 214 00:14:23,360 --> 00:14:26,200 Speaker 1: They were being dedicated to a real world product of 215 00:14:26,240 --> 00:14:34,440 Speaker 1: some sort. Google's also had other involvement with some DARPA stuff, 216 00:14:34,480 --> 00:14:37,680 Speaker 1: at least after the fact. Google does not tend to 217 00:14:37,720 --> 00:14:42,360 Speaker 1: get involved in these challenges directly, um probably because of 218 00:14:42,400 --> 00:14:47,440 Speaker 1: the military association of DARPA and Google is very careful 219 00:14:47,480 --> 00:14:51,680 Speaker 1: to avoid those kind of associations. Now, the finals took 220 00:14:51,720 --> 00:14:56,440 Speaker 1: place over June five and June six and Pomona, California, 221 00:14:56,680 --> 00:15:00,960 Speaker 1: and uh, it was at the Fair Plex, and there 222 00:15:01,040 --> 00:15:04,840 Speaker 1: was quite a crowd watching these events, and the reports 223 00:15:04,840 --> 00:15:07,120 Speaker 1: I read were really interesting, and also the videos I 224 00:15:07,160 --> 00:15:11,280 Speaker 1: watched were interesting because the crowd was extremely enthusiastic cheering 225 00:15:11,360 --> 00:15:15,840 Speaker 1: on the robots, groaning every time a robot failed a 226 00:15:15,960 --> 00:15:20,240 Speaker 1: task or fell over, gasping when that sort of stuff happened, 227 00:15:20,640 --> 00:15:24,280 Speaker 1: because everyone wanted to see these robots succeed. No one 228 00:15:24,360 --> 00:15:28,520 Speaker 1: wanted any group to fail. And uh. Also, the other 229 00:15:28,600 --> 00:15:31,840 Speaker 1: interesting thing was that it was really hard to tell 230 00:15:32,040 --> 00:15:34,920 Speaker 1: if you were in the audience when a robot was 231 00:15:35,040 --> 00:15:39,960 Speaker 1: operating autonomously as opposed to being controlled by humans. Um. 232 00:15:40,040 --> 00:15:44,320 Speaker 1: That says two things, One that the line between these 233 00:15:44,360 --> 00:15:48,040 Speaker 1: two is getting further and further blurred, and too that 234 00:15:48,280 --> 00:15:51,600 Speaker 1: sometimes the robot would behave in a weird way, and 235 00:15:51,640 --> 00:15:54,520 Speaker 1: you couldn't be sure if it was because the autonomous 236 00:15:54,560 --> 00:16:00,360 Speaker 1: programming was lacking or because the control mechanisms were limited. So, 237 00:16:00,400 --> 00:16:03,840 Speaker 1: in other words, it's not that necessarily that the art 238 00:16:03,880 --> 00:16:06,280 Speaker 1: has gotten so advanced that we can't tell the difference. 239 00:16:06,320 --> 00:16:08,720 Speaker 1: It may be that the art has not advanced enough 240 00:16:09,200 --> 00:16:13,600 Speaker 1: that we can't spot what is the cause of incompetence. 241 00:16:15,040 --> 00:16:18,640 Speaker 1: Uh And and I use the word incompetence mainly as 242 00:16:18,680 --> 00:16:22,960 Speaker 1: a joke. It's just a joke because truthfully it displays 243 00:16:23,000 --> 00:16:27,280 Speaker 1: how difficult robotics as a discipline actually is, how how 244 00:16:27,400 --> 00:16:30,960 Speaker 1: challenging it is to design a robot is capable of 245 00:16:31,000 --> 00:16:33,440 Speaker 1: doing the same sort of things that humans can do. 246 00:16:34,360 --> 00:16:36,800 Speaker 1: It shows that the things that we take for granted 247 00:16:37,440 --> 00:16:42,520 Speaker 1: as being pretty simple are fiendish lee difficult when you 248 00:16:42,560 --> 00:16:45,800 Speaker 1: get to a design and programming level when you're actually 249 00:16:45,800 --> 00:16:47,920 Speaker 1: building a robot that's going to be capable of doing 250 00:16:47,960 --> 00:16:51,800 Speaker 1: the same thing. So here are the eight tasks that 251 00:16:51,840 --> 00:16:55,400 Speaker 1: the robots had to complete, and one of these tasks 252 00:16:55,760 --> 00:16:58,680 Speaker 1: was called Surprise. I'll get to it, but the reason 253 00:16:58,720 --> 00:17:01,960 Speaker 1: it's called surprise is that DARPA told all the teams, 254 00:17:02,680 --> 00:17:07,200 Speaker 1: your robot will encounter a task similar to these other 255 00:17:07,280 --> 00:17:10,520 Speaker 1: ones that you already know about, but we're not going 256 00:17:10,560 --> 00:17:13,320 Speaker 1: to tell you what it is. And it meant that 257 00:17:13,359 --> 00:17:16,440 Speaker 1: the teams knew that there was going to be something 258 00:17:17,320 --> 00:17:21,680 Speaker 1: on that list of tasks that was not not defined, 259 00:17:22,359 --> 00:17:24,040 Speaker 1: and that the robot would have to be able to 260 00:17:24,080 --> 00:17:26,600 Speaker 1: contend with it in order to get a point for 261 00:17:26,720 --> 00:17:30,919 Speaker 1: that particular task. And so the tasks were um there 262 00:17:30,920 --> 00:17:33,800 Speaker 1: were eight, Like I said, eight of them. You were 263 00:17:33,880 --> 00:17:37,240 Speaker 1: awarded a point if your robot could successfully complete that task. 264 00:17:38,640 --> 00:17:42,680 Speaker 1: So the score, the final scores once everything was done, 265 00:17:43,000 --> 00:17:46,439 Speaker 1: were determined by how many tasks were completed. Successfully and 266 00:17:46,560 --> 00:17:51,280 Speaker 1: what was the time of the robots performance. Now, when 267 00:17:51,520 --> 00:17:55,320 Speaker 1: they were first designing the robots, teams were told they 268 00:17:55,320 --> 00:17:58,480 Speaker 1: would have half an hour per task, So you have 269 00:17:58,560 --> 00:18:01,280 Speaker 1: eight tasks half an hour per task, that's four hours total. 270 00:18:01,800 --> 00:18:04,200 Speaker 1: But the actual competition, they were told they would have 271 00:18:04,240 --> 00:18:09,240 Speaker 1: one hour to complete all eight tasks. Um. And while 272 00:18:09,320 --> 00:18:11,560 Speaker 1: you might think that how is that fair as a 273 00:18:11,560 --> 00:18:14,639 Speaker 1: as a bait and switch, you also have to understand 274 00:18:14,760 --> 00:18:20,000 Speaker 1: this was all about simulating an emergency response. So under 275 00:18:20,080 --> 00:18:23,840 Speaker 1: those emergency conditions, you can't expect to ask for more time. 276 00:18:24,080 --> 00:18:27,679 Speaker 1: That's not realistic. Uh. And it added an extra layer 277 00:18:27,720 --> 00:18:32,280 Speaker 1: of pressure on the teams. So those eight tasks, the 278 00:18:32,320 --> 00:18:35,240 Speaker 1: first one was to drive, so the robots had to 279 00:18:35,359 --> 00:18:39,919 Speaker 1: drive a Polaris Ranger XP nine vehicle. These if you 280 00:18:39,920 --> 00:18:42,000 Speaker 1: haven't seen them, they look like sometimes you might call 281 00:18:42,119 --> 00:18:44,520 Speaker 1: mcgator or you might think of it as a golf 282 00:18:44,600 --> 00:18:49,359 Speaker 1: cart on steroids. These are um vehicles that are similar 283 00:18:49,400 --> 00:18:53,040 Speaker 1: to golf carts. They're they're largely open, they've got heavier 284 00:18:53,040 --> 00:18:56,200 Speaker 1: wheels than golf carts, do more warm horsepower than a 285 00:18:56,280 --> 00:19:00,520 Speaker 1: golf cart would. But it's definitely in that rain between 286 00:19:00,520 --> 00:19:04,439 Speaker 1: golf cart and real car teams had five minutes to 287 00:19:04,520 --> 00:19:09,600 Speaker 1: alter the vehicle without using tools to make sure that 288 00:19:09,640 --> 00:19:12,760 Speaker 1: their robots could actually operate it. So what the robots 289 00:19:12,800 --> 00:19:15,760 Speaker 1: had to do was drive from the starting point to 290 00:19:16,480 --> 00:19:21,080 Speaker 1: a destination, and it was only considered a success if 291 00:19:21,080 --> 00:19:24,720 Speaker 1: the vehicle had completely moved past a finishing mark. The 292 00:19:24,800 --> 00:19:29,119 Speaker 1: vehicle also had to go through essentially a driving course 293 00:19:29,400 --> 00:19:33,200 Speaker 1: with obstacles and cones set up, and it was determined 294 00:19:33,200 --> 00:19:36,560 Speaker 1: that if the robot were to collide with one of 295 00:19:36,560 --> 00:19:38,600 Speaker 1: those obstacles or one of those cones and cause it 296 00:19:38,640 --> 00:19:42,119 Speaker 1: to move as a result of that collision, the robot 297 00:19:42,160 --> 00:19:47,639 Speaker 1: would not receive a point for that task uh the robot. Also, 298 00:19:47,960 --> 00:19:51,200 Speaker 1: the team could choose for the robot too, instead of driving, 299 00:19:51,280 --> 00:19:54,359 Speaker 1: to walk to the destination, but in that case they 300 00:19:54,400 --> 00:19:56,879 Speaker 1: would not be awarded a point for that task. They 301 00:19:56,880 --> 00:20:01,920 Speaker 1: would forfeit the point um. But ideally the robot would 302 00:20:01,960 --> 00:20:04,359 Speaker 1: be able to operate the accelerator and the steering wheel 303 00:20:05,200 --> 00:20:08,800 Speaker 1: and maybe even the break and the shift. The cars 304 00:20:08,840 --> 00:20:12,240 Speaker 1: were already or the vehicles were already started, the engines 305 00:20:12,240 --> 00:20:16,760 Speaker 1: were already running, and they were already in high gear. 306 00:20:17,160 --> 00:20:20,000 Speaker 1: Because it was considered to be the smoothest way for 307 00:20:20,040 --> 00:20:24,479 Speaker 1: the robots to to maneuver the vehicle. But if teams 308 00:20:24,480 --> 00:20:26,919 Speaker 1: wanted to, they could design a robot that would be 309 00:20:26,920 --> 00:20:30,520 Speaker 1: capable of shifting gears. It was not a requirement, it 310 00:20:30,600 --> 00:20:32,400 Speaker 1: was just something they could do if they wanted to. 311 00:20:33,560 --> 00:20:36,040 Speaker 1: At any rate, the robot needed to be able to 312 00:20:36,119 --> 00:20:40,280 Speaker 1: drive safely from the starting point to the destination and 313 00:20:40,320 --> 00:20:44,119 Speaker 1: then halt the vehicle, either by letting off of the 314 00:20:44,160 --> 00:20:46,760 Speaker 1: accelerator so it coasted to a stop or actually applying 315 00:20:46,760 --> 00:20:50,720 Speaker 1: the brake. The next task. That was just task number one. 316 00:20:51,280 --> 00:20:54,080 Speaker 1: The next task was called egress, which is really just 317 00:20:54,280 --> 00:20:57,800 Speaker 1: getting out of the car. It sounds incredibly simple, and 318 00:20:57,840 --> 00:21:00,359 Speaker 1: for humans, for most of us, it really is pretty simple, 319 00:21:00,960 --> 00:21:04,680 Speaker 1: you know it. We we intuitively know how to maintain 320 00:21:04,720 --> 00:21:07,439 Speaker 1: our balance and to shift our weight so that we 321 00:21:07,560 --> 00:21:10,800 Speaker 1: go from a seated position to standing. But that's not 322 00:21:10,880 --> 00:21:13,119 Speaker 1: the case with the robots. You have to design the 323 00:21:13,200 --> 00:21:15,399 Speaker 1: robot to do that. You have to program the robot 324 00:21:15,440 --> 00:21:17,320 Speaker 1: to do that. You have to the robot be able 325 00:21:17,359 --> 00:21:21,760 Speaker 1: to anticipate what a shift and its weight will do, 326 00:21:21,880 --> 00:21:25,359 Speaker 1: what how it's momentum will carry it forward. As it 327 00:21:25,359 --> 00:21:28,240 Speaker 1: turns out, this was one of the trickier tasks that 328 00:21:28,840 --> 00:21:31,600 Speaker 1: teams had to complete to get out of a vehicle. 329 00:21:31,720 --> 00:21:36,600 Speaker 1: It was not an easy thing to do um but 330 00:21:36,800 --> 00:21:38,200 Speaker 1: they were told that they could get out of the 331 00:21:38,280 --> 00:21:40,800 Speaker 1: vehicle in either direction. They didn't have to exit out 332 00:21:40,800 --> 00:21:42,879 Speaker 1: all the left side versus the right side. That the 333 00:21:42,960 --> 00:21:46,840 Speaker 1: robot could get out of either side, and again the 334 00:21:46,880 --> 00:21:50,200 Speaker 1: task was considered complete if the robot could get out 335 00:21:50,200 --> 00:21:52,720 Speaker 1: of the vehicle and maintain its balance and move on 336 00:21:52,800 --> 00:21:55,920 Speaker 1: to the next challenge, and if it could, it would 337 00:21:55,920 --> 00:22:00,760 Speaker 1: receive a point. The next one was door. Now, egress 338 00:22:00,840 --> 00:22:04,800 Speaker 1: was tough. Door was surprisingly tough. You would think that 339 00:22:04,920 --> 00:22:07,919 Speaker 1: opening a door would be a pretty simple task. This 340 00:22:08,000 --> 00:22:10,960 Speaker 1: was a push door operated by a lever style handle. 341 00:22:11,040 --> 00:22:13,560 Speaker 1: You had to push up or pull you know, pull 342 00:22:13,640 --> 00:22:16,640 Speaker 1: up or push down rather on the handle and then 343 00:22:16,680 --> 00:22:20,320 Speaker 1: push the door to open it, then step through the door. 344 00:22:20,720 --> 00:22:22,600 Speaker 1: The door did not have a threshold, so there was 345 00:22:22,640 --> 00:22:25,600 Speaker 1: nothing that you had to step over. However, the door 346 00:22:25,800 --> 00:22:29,480 Speaker 1: was a standard thirty six inch wide doorway and with 347 00:22:29,600 --> 00:22:31,840 Speaker 1: the door jam that's closer to thirty three and a 348 00:22:31,840 --> 00:22:34,960 Speaker 1: half inches. And some of these robots were too wide 349 00:22:35,240 --> 00:22:38,520 Speaker 1: to walk through the door, you know, heading with their 350 00:22:38,560 --> 00:22:41,679 Speaker 1: torso facing forward. They actually had to turn sideways and 351 00:22:41,720 --> 00:22:45,280 Speaker 1: then shuffle through the doorway, and there were a lot 352 00:22:45,320 --> 00:22:48,600 Speaker 1: of robots that fell down at this stage of the 353 00:22:49,080 --> 00:22:54,800 Speaker 1: the challenge. They would lose their balance either when pushing 354 00:22:54,840 --> 00:22:57,640 Speaker 1: the door or when trying to maneuver through the doorway. 355 00:22:58,800 --> 00:23:02,399 Speaker 1: One robot fell down at this stage and was able 356 00:23:02,440 --> 00:23:06,480 Speaker 1: to pick itself back up very very slowly and deliberately, 357 00:23:07,640 --> 00:23:12,720 Speaker 1: which is remarkable because it's the only robot that managed 358 00:23:12,840 --> 00:23:15,439 Speaker 1: to pick itself back up after falling over, and that 359 00:23:15,520 --> 00:23:19,080 Speaker 1: was the Carnegie Melon robot, the same one we mentioned 360 00:23:19,080 --> 00:23:22,080 Speaker 1: in the Trials part in just a few moments ago. 361 00:23:23,600 --> 00:23:25,200 Speaker 1: But anyway, that was it. You just had to open 362 00:23:25,200 --> 00:23:27,000 Speaker 1: the door and walk through the doorway, and that was 363 00:23:27,040 --> 00:23:30,200 Speaker 1: the end of that task, and yet it was deceptively difficult. 364 00:23:30,680 --> 00:23:35,560 Speaker 1: The next one was Valve, which was not about the 365 00:23:35,760 --> 00:23:39,359 Speaker 1: game company, but rather about turning a control as if 366 00:23:39,400 --> 00:23:42,720 Speaker 1: you were turning a valve to operate a fire hose 367 00:23:42,960 --> 00:23:46,159 Speaker 1: or to shut down water to a specific part of 368 00:23:46,160 --> 00:23:49,040 Speaker 1: the system, which makes sense in a nuclear facility you 369 00:23:49,119 --> 00:23:52,520 Speaker 1: might have to shut off or open a valve. In 370 00:23:52,560 --> 00:23:57,359 Speaker 1: that case, the valve had to be turned counterclockwise in 371 00:23:57,480 --> 00:24:00,600 Speaker 1: three sixty degrees in order for the task to complete, 372 00:24:01,080 --> 00:24:03,679 Speaker 1: and the team was only told that the valve would 373 00:24:03,920 --> 00:24:07,240 Speaker 1: have a size between four and sixteen inches in diameter, 374 00:24:08,000 --> 00:24:10,520 Speaker 1: which meant that you had to create a robot that 375 00:24:10,560 --> 00:24:13,959 Speaker 1: would be capable of gripping anything within that range and 376 00:24:13,960 --> 00:24:19,080 Speaker 1: then turning it in that counterclockwise direction three degrees again 377 00:24:19,160 --> 00:24:22,840 Speaker 1: trigger than it sounds the next step. The next task 378 00:24:23,000 --> 00:24:25,160 Speaker 1: was called wall, in which a robot had to pick 379 00:24:25,240 --> 00:24:28,080 Speaker 1: up a cordless drill. There were two of them available, 380 00:24:28,520 --> 00:24:31,520 Speaker 1: They were not automatically on. The robot had to operate 381 00:24:31,560 --> 00:24:35,040 Speaker 1: the cordless drill and turn it on by squeezing the trigger, 382 00:24:35,920 --> 00:24:39,000 Speaker 1: and use the cordless drill to cut through some drywall 383 00:24:39,480 --> 00:24:43,720 Speaker 1: in a shape that was drawn on the drywall. The 384 00:24:43,800 --> 00:24:47,080 Speaker 1: drywall was a half inch thick, and the robot had 385 00:24:47,119 --> 00:24:49,840 Speaker 1: to hold the drill, operate the drill, and move in 386 00:24:49,840 --> 00:24:53,000 Speaker 1: this shape and then remove the rubble. The idea being 387 00:24:53,040 --> 00:24:56,320 Speaker 1: that sometimes the robot might have to cut through a 388 00:24:56,480 --> 00:25:00,239 Speaker 1: surface in order to get access to certain controls uh 389 00:25:00,440 --> 00:25:04,119 Speaker 1: that might otherwise be blocked by a collapse of a 390 00:25:04,200 --> 00:25:08,399 Speaker 1: room or something along those lines. Also pretty tricky, you know, 391 00:25:08,560 --> 00:25:11,680 Speaker 1: using tools that were designed to be held in human hands. 392 00:25:12,280 --> 00:25:14,720 Speaker 1: You know, humans, we we have the ability to detect 393 00:25:14,960 --> 00:25:19,360 Speaker 1: how much pressure we're using and to um to change 394 00:25:19,440 --> 00:25:22,919 Speaker 1: that based upon what's happening. Robots a little more tricky. 395 00:25:22,960 --> 00:25:26,399 Speaker 1: I mean, you can have sensors that alert the the 396 00:25:26,400 --> 00:25:30,800 Speaker 1: operator what's happening to what's happening, but it's not intuitive. Again, 397 00:25:30,840 --> 00:25:35,760 Speaker 1: you have to program it. After wall was the surprise. Now, 398 00:25:35,840 --> 00:25:39,080 Speaker 1: the surprise was something that DARPA could change out from 399 00:25:39,160 --> 00:25:41,600 Speaker 1: day to day. There were two days of this series 400 00:25:41,640 --> 00:25:45,200 Speaker 1: of challenges, and I know that on one day, I'm 401 00:25:45,320 --> 00:25:47,640 Speaker 1: not sure if this was the same for the other day, 402 00:25:47,680 --> 00:25:50,600 Speaker 1: but on one day it involved picking up a plug 403 00:25:51,240 --> 00:25:55,879 Speaker 1: and plugging it into a socket. Ah. I think they 404 00:25:56,000 --> 00:25:58,200 Speaker 1: probably changed it for the second day, but I never 405 00:25:58,240 --> 00:26:00,960 Speaker 1: could find out what the surprise was on that case. 406 00:26:01,400 --> 00:26:03,960 Speaker 1: But at any rate, it also ended up being pretty 407 00:26:04,000 --> 00:26:07,960 Speaker 1: tricky that I saw one robot that attempted several times 408 00:26:08,000 --> 00:26:10,600 Speaker 1: to plug the plug into a socket and it was 409 00:26:12,200 --> 00:26:15,240 Speaker 1: sad and funny at the same time. But yeah, again, 410 00:26:15,320 --> 00:26:17,080 Speaker 1: it was one of those things where it was similar 411 00:26:17,119 --> 00:26:20,399 Speaker 1: to tasks the robots had had to do previously, but 412 00:26:20,640 --> 00:26:23,199 Speaker 1: wasn't something they were expressly told they would have to 413 00:26:23,240 --> 00:26:26,320 Speaker 1: do during the actual challenge, so that's what made it 414 00:26:26,400 --> 00:26:31,120 Speaker 1: really hard. The next task was rubble, in which the 415 00:26:31,240 --> 00:26:34,399 Speaker 1: robot had to walk through a debris field or on 416 00:26:34,520 --> 00:26:36,720 Speaker 1: top of the debris, either through it or on top 417 00:26:36,760 --> 00:26:40,960 Speaker 1: of it. Again very challenging for robots to maintain their balance. 418 00:26:41,280 --> 00:26:44,960 Speaker 1: There were lots of you know, Boston Dynamics robots that 419 00:26:45,000 --> 00:26:48,320 Speaker 1: were pretty good at doing this, but still pretty tricky. 420 00:26:48,320 --> 00:26:49,840 Speaker 1: You saw a lot of robots fall over at this 421 00:26:49,880 --> 00:26:53,280 Speaker 1: point too. And then there were stairs that was the 422 00:26:53,320 --> 00:26:56,440 Speaker 1: final task was to climb us of stairs that had 423 00:26:56,480 --> 00:26:59,040 Speaker 1: a rail on the left side but no rail on 424 00:26:59,080 --> 00:27:03,400 Speaker 1: the right side, and once the robot was completely on 425 00:27:03,440 --> 00:27:05,960 Speaker 1: the top of the stairs, it was considered to have 426 00:27:06,040 --> 00:27:08,800 Speaker 1: completed the task and the course and it's time would 427 00:27:08,800 --> 00:27:15,280 Speaker 1: be logged. So again, those those tasks, for the most part, 428 00:27:15,400 --> 00:27:18,720 Speaker 1: pretty simple for your average human to do. But imagine 429 00:27:18,720 --> 00:27:20,600 Speaker 1: that it's your job to build a robot that could 430 00:27:20,600 --> 00:27:23,840 Speaker 1: do all of those things. It has to be able 431 00:27:23,880 --> 00:27:26,440 Speaker 1: to have some form of perception, has to be able 432 00:27:26,480 --> 00:27:31,320 Speaker 1: to see either for the human operators to uh pick 433 00:27:31,480 --> 00:27:36,320 Speaker 1: up various tools or navigate through certain areas, or for 434 00:27:36,359 --> 00:27:39,960 Speaker 1: it to to see and operate autonomously, and has to 435 00:27:39,960 --> 00:27:44,200 Speaker 1: be able to perceive its environment understand depth. In fact, 436 00:27:44,200 --> 00:27:46,840 Speaker 1: a lot of teams had trouble with the the wall 437 00:27:46,880 --> 00:27:50,560 Speaker 1: task because it was very hard to perceive how far 438 00:27:50,600 --> 00:27:53,480 Speaker 1: away the drill was from a hand to pick it 439 00:27:53,560 --> 00:27:57,760 Speaker 1: up and then use it again. Things that are pretty 440 00:27:57,800 --> 00:28:04,400 Speaker 1: easy for most of us very hard from a robotic standpoint. Um. 441 00:28:04,440 --> 00:28:05,960 Speaker 1: And like I said, the door was one of the 442 00:28:05,960 --> 00:28:09,000 Speaker 1: hardest ones. Uh, a lot of robots had a lot 443 00:28:09,040 --> 00:28:11,600 Speaker 1: of issues just walking through a doorway, which is both 444 00:28:11,600 --> 00:28:15,240 Speaker 1: funny and and really does bring to light that we've 445 00:28:15,240 --> 00:28:19,520 Speaker 1: got a long way to go with robotics. Um. Well, 446 00:28:19,560 --> 00:28:22,560 Speaker 1: if if you, as someone in the crowd had said, 447 00:28:22,720 --> 00:28:26,119 Speaker 1: if the robotic uprising happens, just close the door and 448 00:28:26,200 --> 00:28:30,360 Speaker 1: you'll be fine. Uh, Because it's tricky stuff. It really 449 00:28:30,359 --> 00:28:32,919 Speaker 1: shows the automation is. We take a lot of it 450 00:28:32,920 --> 00:28:36,320 Speaker 1: for granted because the examples we see tend to be 451 00:28:36,359 --> 00:28:40,800 Speaker 1: pretty elegant because they're designed for a specific purpose. But 452 00:28:41,000 --> 00:28:43,480 Speaker 1: as we start looking at a more general purpose robot, 453 00:28:43,640 --> 00:28:48,560 Speaker 1: we begin to understand exactly how complicated we human beings are. 454 00:28:49,160 --> 00:28:51,440 Speaker 1: So to design a machine that can operate within our 455 00:28:51,520 --> 00:28:58,000 Speaker 1: human world seamlessly is an incredible challenge. So who won well, 456 00:28:58,040 --> 00:29:00,640 Speaker 1: the first place went to a South Korean team k 457 00:29:01,000 --> 00:29:05,320 Speaker 1: A I ST. It had its robot d d r 458 00:29:05,440 --> 00:29:10,200 Speaker 1: C Hubo, which completed all eight tasks in forty four 459 00:29:10,240 --> 00:29:14,959 Speaker 1: minutes and twenty eight seconds. Those eight tests forty four 460 00:29:15,000 --> 00:29:17,680 Speaker 1: minutes and twenty eight seconds, remember that's driving from one 461 00:29:17,720 --> 00:29:21,400 Speaker 1: point to another a very short distance, actually getting out 462 00:29:21,440 --> 00:29:25,400 Speaker 1: of that vehicle, walking through a doorway, cutting a hole 463 00:29:25,480 --> 00:29:29,600 Speaker 1: through a wall, turning a valve, plugging in a plug, 464 00:29:30,280 --> 00:29:33,280 Speaker 1: um walking over some debris, and climbing some stairs, something 465 00:29:33,320 --> 00:29:38,600 Speaker 1: that would take you know, a healthy person. If it 466 00:29:38,640 --> 00:29:40,640 Speaker 1: took the more than ten minutes, you would wonder what 467 00:29:40,760 --> 00:29:45,920 Speaker 1: was going on. So another example that the winning team 468 00:29:46,080 --> 00:29:48,240 Speaker 1: did it in forty four minutes and twenty eight seconds. 469 00:29:48,960 --> 00:29:51,560 Speaker 1: There's also a great preparation video for the d r 470 00:29:51,640 --> 00:29:54,480 Speaker 1: C Hubo that was put out by the South Korean 471 00:29:54,520 --> 00:29:57,840 Speaker 1: team that I loved because it was like a training 472 00:29:57,880 --> 00:30:01,360 Speaker 1: montage from Rocky and and they It was clear the 473 00:30:01,400 --> 00:30:03,959 Speaker 1: team had a sense of humor about this because it 474 00:30:04,000 --> 00:30:07,240 Speaker 1: was showing the robot not just completing tasks that would 475 00:30:07,240 --> 00:30:10,120 Speaker 1: be similar to the ones that were in the actual challenge, 476 00:30:10,120 --> 00:30:13,760 Speaker 1: including things like climbing a ladder or um or lifting 477 00:30:13,960 --> 00:30:18,200 Speaker 1: a tool, but also doing your typical you know, training 478 00:30:18,240 --> 00:30:21,760 Speaker 1: montage stuff like doing push ups or taking a fighting 479 00:30:21,840 --> 00:30:26,520 Speaker 1: stance like you're a kung fu master, And that really 480 00:30:26,560 --> 00:30:29,240 Speaker 1: tickled me. I thought it was pretty entertaining but also 481 00:30:29,320 --> 00:30:31,560 Speaker 1: very impressive about what they were capable of doing with 482 00:30:31,560 --> 00:30:34,880 Speaker 1: that robot. The second place team was I h M C. 483 00:30:35,200 --> 00:30:37,840 Speaker 1: That pensaclea Florida group we talked about a minute ago. 484 00:30:38,000 --> 00:30:43,720 Speaker 1: They were using the Boston Dynamics running Man Atlas butt Um. 485 00:30:43,760 --> 00:30:48,280 Speaker 1: It took it took about fifty minutes in twenty six 486 00:30:48,280 --> 00:30:50,959 Speaker 1: seconds to complete all eight tasks. After he did so, 487 00:30:51,040 --> 00:30:53,960 Speaker 1: after it climbed the stairs, it lifted its arms in 488 00:30:54,080 --> 00:30:59,640 Speaker 1: victory and then fell over, So I guess that's Hubris. 489 00:31:00,160 --> 00:31:04,280 Speaker 1: The third place team was Carnegie Melon. Their team Tartan Rescue, 490 00:31:04,440 --> 00:31:07,240 Speaker 1: the one that had come in third place in the 491 00:31:07,280 --> 00:31:11,160 Speaker 1: trials as well. Their robot was called Chimp, which had 492 00:31:11,320 --> 00:31:14,200 Speaker 1: very long arms. Is a red robot with arms that 493 00:31:14,240 --> 00:31:17,120 Speaker 1: looked kind of freakishly long if you compare it to 494 00:31:17,200 --> 00:31:21,880 Speaker 1: a human. And it also had treads like a like 495 00:31:21,920 --> 00:31:24,200 Speaker 1: a treadmill type thing on both of its legs and 496 00:31:24,240 --> 00:31:26,560 Speaker 1: on both of its arms. And that was the robot 497 00:31:26,600 --> 00:31:28,920 Speaker 1: that fell over in the door challenge and was able 498 00:31:28,960 --> 00:31:33,160 Speaker 1: to right itself by itself, the only one that could 499 00:31:33,160 --> 00:31:34,880 Speaker 1: get up by itself and didn't need a team of 500 00:31:34,960 --> 00:31:38,040 Speaker 1: humans to actually lift it up. Now, those were the 501 00:31:38,080 --> 00:31:41,480 Speaker 1: only three teams that actually completed all of the eight tasks, 502 00:31:41,560 --> 00:31:44,560 Speaker 1: and they won the prizes. First place was two million dollars, 503 00:31:44,600 --> 00:31:47,200 Speaker 1: second place was one million, and third place was five thousand. 504 00:31:48,040 --> 00:31:50,720 Speaker 1: The other teams broke down like this. Four of them 505 00:31:50,920 --> 00:31:54,920 Speaker 1: were able to complete seven of the tasks, only one 506 00:31:55,400 --> 00:31:59,320 Speaker 1: was capable of completing just six tasks. You had two 507 00:31:59,320 --> 00:32:03,040 Speaker 1: teams that com leaded five, two teams that completed four, 508 00:32:03,760 --> 00:32:08,480 Speaker 1: four teams that completed three, two that completed to one 509 00:32:08,520 --> 00:32:12,040 Speaker 1: that completed one, and four teams weren't able to complete 510 00:32:12,080 --> 00:32:16,120 Speaker 1: any of the tasks successfully. Um. Now, the whole point 511 00:32:16,120 --> 00:32:19,720 Speaker 1: of this was to really push that state of the 512 00:32:19,840 --> 00:32:23,600 Speaker 1: art forward, to have engineers think, what would you need 513 00:32:23,680 --> 00:32:27,080 Speaker 1: to do to build a robot capable of actually responding 514 00:32:27,080 --> 00:32:31,760 Speaker 1: to real world situations. Who are the challenges that are 515 00:32:31,800 --> 00:32:33,920 Speaker 1: in the way of that. How can we advance the 516 00:32:33,960 --> 00:32:38,000 Speaker 1: technology both in the hardware and in the software to 517 00:32:38,480 --> 00:32:43,000 Speaker 1: overcome these challenges. And the goal was not to build 518 00:32:43,000 --> 00:32:45,440 Speaker 1: a super robot that's going to save the world. This 519 00:32:45,520 --> 00:32:48,320 Speaker 1: is not you know, the Avengers age of Ultron, That's 520 00:32:48,320 --> 00:32:52,480 Speaker 1: not what's going on here. It's all about incremental improvements 521 00:32:52,520 --> 00:32:56,719 Speaker 1: in the art and discipline of robotics so that the 522 00:32:56,760 --> 00:33:00,480 Speaker 1: next generation of robots can benefit from the research and 523 00:33:00,520 --> 00:33:03,800 Speaker 1: development done in this first generation, and a good way 524 00:33:03,800 --> 00:33:06,440 Speaker 1: of seeing how this plays out in the real world, 525 00:33:06,560 --> 00:33:08,800 Speaker 1: because you know, you might think, well, that was a 526 00:33:08,800 --> 00:33:13,360 Speaker 1: clever demonstration, but what what can we expect here on 527 00:33:13,400 --> 00:33:17,320 Speaker 1: our day to day lives. Take a look at how 528 00:33:17,400 --> 00:33:20,920 Speaker 1: autonomous cars are coming along, because of course, DARPA held 529 00:33:20,960 --> 00:33:25,720 Speaker 1: the Grand Challenge back in the mid to late two thousand's, 530 00:33:25,760 --> 00:33:28,800 Speaker 1: and that was the challenge in which different groups tried 531 00:33:28,840 --> 00:33:32,640 Speaker 1: to build self driving cars that could complete a course, 532 00:33:33,240 --> 00:33:36,400 Speaker 1: whether it was on the desert or simulated urban environment. 533 00:33:37,280 --> 00:33:41,440 Speaker 1: And we are now seeing people who worked on the 534 00:33:41,520 --> 00:33:45,440 Speaker 1: various teams that that competed in that DARPA Grand Challenge 535 00:33:46,160 --> 00:33:50,360 Speaker 1: building what will be the first generation of driverless cars 536 00:33:50,400 --> 00:33:54,040 Speaker 1: that will eventually make it to the consumer market. Google 537 00:33:54,280 --> 00:33:58,640 Speaker 1: has people on its team that competed in the those 538 00:33:58,680 --> 00:34:03,800 Speaker 1: Grand Challenges, so we're seeing that expertise that was developed 539 00:34:03,840 --> 00:34:07,680 Speaker 1: as as engineers were given a problem and told find 540 00:34:07,720 --> 00:34:09,960 Speaker 1: a way to solve this. We're seeing that expertise they 541 00:34:09,960 --> 00:34:12,799 Speaker 1: developed come into play in the real world now, and 542 00:34:12,840 --> 00:34:15,240 Speaker 1: it's sure it's gonna be a few more years before 543 00:34:15,239 --> 00:34:20,920 Speaker 1: we get autonomous cars as a reality that an actual 544 00:34:21,080 --> 00:34:23,719 Speaker 1: human being can go out and purchase, as opposed to 545 00:34:23,760 --> 00:34:27,720 Speaker 1: a representative of a company or you know, just uh, 546 00:34:27,719 --> 00:34:30,839 Speaker 1: you know, someone who is is there to demonstrate the 547 00:34:30,920 --> 00:34:33,279 Speaker 1: viability of the technology. It's going to be a few 548 00:34:33,280 --> 00:34:35,560 Speaker 1: more years before this becomes something that you or I 549 00:34:35,640 --> 00:34:39,880 Speaker 1: could go to the dealership and actually purchase. But the 550 00:34:39,920 --> 00:34:42,680 Speaker 1: reality is there on the horizon. Same here with the 551 00:34:42,800 --> 00:34:47,960 Speaker 1: Darba Robotics challenge. It will likely take a decade or 552 00:34:48,040 --> 00:34:52,080 Speaker 1: more to develop a humanoid robot or a robot capable 553 00:34:52,640 --> 00:34:57,640 Speaker 1: of operating within a world designed by humanoids for humanoids 554 00:34:58,520 --> 00:35:02,480 Speaker 1: and do so seem lessly. It will take a lot 555 00:35:02,480 --> 00:35:06,000 Speaker 1: of time and a lot more development, but the foundation 556 00:35:06,080 --> 00:35:10,680 Speaker 1: has been laid, so I'm particularly excited to see what 557 00:35:10,800 --> 00:35:14,719 Speaker 1: the future holds for us based upon these results. And 558 00:35:14,760 --> 00:35:18,280 Speaker 1: while those montages of robots falling over are really funny, 559 00:35:18,719 --> 00:35:23,160 Speaker 1: I am actually optimistic about what robotics will be able 560 00:35:23,400 --> 00:35:27,120 Speaker 1: to accomplish in the future. I'm just also realistic that 561 00:35:27,160 --> 00:35:30,400 Speaker 1: it's going to take time. It's not We're not on 562 00:35:30,440 --> 00:35:35,120 Speaker 1: the verge of a terminator like future. The state of 563 00:35:35,160 --> 00:35:39,160 Speaker 1: the art and artificial intelligence has gotten gone very far 564 00:35:39,239 --> 00:35:42,279 Speaker 1: in certain realms of computer science, but when it comes 565 00:35:42,280 --> 00:35:44,600 Speaker 1: to robotics, there's a there are a lot of problems 566 00:35:44,600 --> 00:35:47,600 Speaker 1: that are still very difficult. You know that vision is 567 00:35:47,600 --> 00:35:52,440 Speaker 1: a big one. Uh, and again, just being able to 568 00:35:52,480 --> 00:35:55,200 Speaker 1: maintain balance again one of those things that we take 569 00:35:55,200 --> 00:35:59,680 Speaker 1: for granted really a challenging problem. I hope you guys 570 00:35:59,760 --> 00:36:03,240 Speaker 1: enjoy aoid this episode, this look at the darker robotics 571 00:36:03,320 --> 00:36:06,839 Speaker 1: challenge outcome. I recommend that you go check out some 572 00:36:06,960 --> 00:36:10,480 Speaker 1: videos that were shot while this was going on, because 573 00:36:10,719 --> 00:36:14,640 Speaker 1: they are really interesting, they're entertaining, and uh they they 574 00:36:14,760 --> 00:36:16,880 Speaker 1: I think they're really exciting. You know, it's something that 575 00:36:16,960 --> 00:36:19,439 Speaker 1: really should get you excited about robotics in general. If 576 00:36:19,440 --> 00:36:22,000 Speaker 1: that's a field you're interested in, definitely check it out. 577 00:36:23,280 --> 00:36:26,080 Speaker 1: If you have any suggestions for future topics of tech stuff, 578 00:36:26,920 --> 00:36:31,839 Speaker 1: whether it's actually a subject like a general type of technology, 579 00:36:32,120 --> 00:36:34,440 Speaker 1: or maybe it's a company or a person that I 580 00:36:34,440 --> 00:36:37,680 Speaker 1: should focus on, or maybe there's a guest you would 581 00:36:37,680 --> 00:36:40,680 Speaker 1: like me to interview or to have on as a 582 00:36:40,719 --> 00:36:44,400 Speaker 1: guest co host. Let me know. Send me a message. 583 00:36:44,680 --> 00:36:47,280 Speaker 1: The email address is tech stuff at how stuff works 584 00:36:47,280 --> 00:36:50,440 Speaker 1: dot com, or drop me a line on Facebook, Twitter 585 00:36:50,640 --> 00:36:53,319 Speaker 1: or Tumbler. The handle it all three of those is 586 00:36:53,400 --> 00:36:56,880 Speaker 1: tech stuff hs W and I'll talk to you again 587 00:36:57,600 --> 00:37:04,080 Speaker 1: really soon. For moral notes in thousands of other topics, 588 00:37:04,400 --> 00:37:15,080 Speaker 1: is the housetop works dot com.