WEBVTT - TechStuff Classic: DARPA Robotics Challenge Review 0:00:04.400 --> 0:00:12.480 Welcome to tex Stuff production from I Heart Radio. Hey there, 0:00:12.520 --> 0:00:15.720 and welcome to tex Stuff. I'm your host, John Than Strickland. 0:00:15.720 --> 0:00:18.160 I'm an executive producer with I Heart Radio and Hell 0:00:18.360 --> 0:00:22.840 the Tech Area. It is time for a tech Stuff 0:00:22.880 --> 0:00:28.440 classic episode. This episode originally published June two thousand fifteen. 0:00:28.520 --> 0:00:34.559 It is titled DARPA Robotics Challenge Review. It's about the 0:00:34.880 --> 0:00:39.720 DARPA Robotics Challenge where they were simulating UH an emergency 0:00:39.760 --> 0:00:44.040 situation and putting a robot through paces of doing certain 0:00:44.159 --> 0:00:47.919 basic tasks in an effort to see if the robots 0:00:47.960 --> 0:00:51.440 could do them, you know, more or less autonomously. Turns 0:00:51.479 --> 0:00:54.040 out that some things that you might think of as 0:00:54.040 --> 0:00:58.120 being incredibly easy to accomplish turned out to be pretty hard. 0:00:58.680 --> 0:01:03.360 Let's listen in So, to start off, we got to 0:01:03.360 --> 0:01:07.560 talk about DARPA itself. Remember that's the research and development 0:01:07.720 --> 0:01:11.600 arm of the Department of Defenses technically the US Defense 0:01:11.680 --> 0:01:17.200 Advanced Research Projects Agency UH and DARPA has been responsible 0:01:17.200 --> 0:01:21.680 for a lot of cutting edge research, largely in the 0:01:21.720 --> 0:01:25.720 mode of military use. But we have seen the benefits 0:01:25.800 --> 0:01:29.600 of that research hit us in other ways. Besides, you know, 0:01:29.840 --> 0:01:35.160 the military applications, So for example, our pa net, the 0:01:35.200 --> 0:01:39.560 predecessor to the Internet, was the platform upon which the 0:01:39.600 --> 0:01:43.400 technologies that allow the Internet to work were developed. And 0:01:43.480 --> 0:01:45.600 that was a DARPA project that was back when they 0:01:45.600 --> 0:01:50.800 were first just called ARPA rather than DARPA. But we 0:01:50.960 --> 0:01:53.600 wouldn't have the Internet, at least not the way we 0:01:53.680 --> 0:01:57.200 have it today if it were not for this agency. 0:01:57.240 --> 0:01:59.160 So the agency has done a lot of stuff that 0:01:59.240 --> 0:02:02.040 has benefited us in many ways. I'll talk about another 0:02:02.080 --> 0:02:04.560 one a little bit later on in the podcast, kind 0:02:04.600 --> 0:02:09.040 of draw parallels to what happened with the robotics challenge. 0:02:09.960 --> 0:02:13.800 So here we have this agency. They're all about research 0:02:13.800 --> 0:02:17.520 and development. They don't actually have lots of labs with 0:02:17.560 --> 0:02:21.360 scientists working in them. Know what this agency does is 0:02:21.400 --> 0:02:26.040 it sends out proposals to other groups and asks them 0:02:26.200 --> 0:02:30.959 to help work on various projects. Sometimes they hold challenges. 0:02:31.440 --> 0:02:34.320 In these challenges, it's kind of an open invitation for 0:02:34.440 --> 0:02:40.440 any group that can participate to uh to compete in 0:02:40.520 --> 0:02:43.440 some way, and the winner usually gets some sort of 0:02:43.480 --> 0:02:47.320 cash prize. And so often the cash prize will be 0:02:47.440 --> 0:02:51.040 less than the investment needed to actually participate in the 0:02:51.080 --> 0:02:54.320 competition to build out the thing that you need to 0:02:54.360 --> 0:02:57.639 make for us to to win, but it's still a 0:02:57.760 --> 0:03:03.119 very important UH and prestigious event to be part of 0:03:03.240 --> 0:03:06.359 if you are capable of competing. Now we're talking mostly 0:03:06.360 --> 0:03:10.960 about research departments engineering departments in universities, but there are 0:03:11.000 --> 0:03:15.800 also some private institutions, not just university type things, but 0:03:15.919 --> 0:03:19.640 actual research and development labs that will participate in DARPA 0:03:19.760 --> 0:03:25.440 projects as well. And it's collectively that these teams were 0:03:26.200 --> 0:03:28.880 kind of they're the ones that are pushing the technology forward, 0:03:28.919 --> 0:03:32.040 and DARK is kind of the facilitator. They're offering up 0:03:32.080 --> 0:03:38.280 the possibility for people to actually UH participate, and not 0:03:38.280 --> 0:03:41.760 only the possibility, but the goals. They define the goals 0:03:41.840 --> 0:03:44.160 that have to be met, and as it turns out, 0:03:44.960 --> 0:03:48.360 defining those goals is a very important part of these 0:03:48.440 --> 0:03:53.520 challenges because you have to tell the engineers what it 0:03:53.680 --> 0:03:56.800 is they need to be able to do before they 0:03:56.840 --> 0:03:59.200 can build the stuff that does it. If if you 0:03:59.280 --> 0:04:03.160 have a very vague description of what needs to be done, 0:04:03.920 --> 0:04:08.800 the possibilities of achieving it are so varied that it 0:04:08.840 --> 0:04:13.280 can often paralyze a project nothing gets done. But if 0:04:13.280 --> 0:04:17.000 you make it very specific, then it narrows down the 0:04:17.080 --> 0:04:20.640 options and it gives more focus to the project. So 0:04:21.040 --> 0:04:26.359 the DARPA Robotics project itself was inspired by a real 0:04:26.480 --> 0:04:32.320 world disaster, specifically the Fukushima disaster in Japan. Now, this 0:04:32.520 --> 0:04:38.359 was the nuclear facility, the nuclear reactor that was flooded 0:04:38.839 --> 0:04:45.120 and then suffered some massive problems because of damage to 0:04:45.160 --> 0:04:50.800 the infrastructure. And we're talking about a nuclear facility, Radiation 0:04:51.000 --> 0:04:55.120 is a factor, so it's a very dangerous environment to 0:04:55.240 --> 0:05:01.240 send responders to. Dark As approach was to create a 0:05:01.279 --> 0:05:05.880 scenario in which a robot would need to go through 0:05:05.920 --> 0:05:10.120 an environment similar to that that was at Fukushima and 0:05:10.160 --> 0:05:14.680 be able to perform the tasks necessary to help avert 0:05:15.080 --> 0:05:20.440 true devastation and catastrophe without putting people at risk. So 0:05:20.480 --> 0:05:22.600 the idea is that you can use a robot which 0:05:22.640 --> 0:05:27.240 is not going to respond to radiation the way a 0:05:27.279 --> 0:05:31.560 living organism would and be able to actually carry out 0:05:31.839 --> 0:05:35.679 the stuff you need to prevent a really huge problem 0:05:35.760 --> 0:05:41.200 from getting worse. So uh, some people have referred to 0:05:41.320 --> 0:05:46.800 the Darker Robotics Challenge as the Robolympics because it was 0:05:46.839 --> 0:05:49.920 a series of tasks that a robot has to complete 0:05:49.960 --> 0:05:53.080 and it had to all be the same. Robot teams 0:05:53.120 --> 0:05:57.360 could not build individual robots designed to complete specific tasks. 0:05:57.680 --> 0:06:00.680 If they could, it would be way easier of a 0:06:00.760 --> 0:06:03.600 challenge because most of the robots that are in use 0:06:03.640 --> 0:06:09.200 today have very narrow parameters. There's just maybe a single task, 0:06:09.839 --> 0:06:12.359 or perhaps just a couple of small, you know, a 0:06:12.400 --> 0:06:14.719 small range of tasks that the robot has to do. 0:06:15.600 --> 0:06:17.880 It doesn't have to do anything outside of that. So 0:06:17.880 --> 0:06:20.719 when you're designing the robot, you just say, all right, 0:06:20.920 --> 0:06:24.200 what design elements will allow the robot to do what 0:06:24.240 --> 0:06:27.200 it has to do. Anything that's outside of that is unnecessary, 0:06:27.279 --> 0:06:30.359 we will not do it. Take the roomba as an example. 0:06:30.360 --> 0:06:33.280 It's a great example. It's a robot the vacuums that 0:06:33.360 --> 0:06:37.400 cleans floors. Anything that's not necessary for the roombot to 0:06:37.440 --> 0:06:41.520 do that is put aside. You'd want to make sure 0:06:41.560 --> 0:06:46.039 that all the design elements in your robot complement its purpose. 0:06:47.200 --> 0:06:50.640 But if you have to complete a lot of different tasks, 0:06:51.240 --> 0:06:54.960 and those tasks are not always similar to one another, 0:06:55.880 --> 0:07:02.960 you complicate the whole design process by incredible factors. It's 0:07:03.000 --> 0:07:08.000 hard to even express how much more difficult this is. Now, 0:07:08.040 --> 0:07:11.280 if you've watched any of the videos that came out 0:07:11.320 --> 0:07:14.800 of the Darker Robotics Challenge. You probably saw at least 0:07:14.800 --> 0:07:18.720 one or two that were montages of robots falling over 0:07:19.480 --> 0:07:21.440 and it's a little funny to watch. You see these 0:07:21.640 --> 0:07:27.680 big robots just topple over sometimes apparently for no reason. 0:07:27.880 --> 0:07:30.280 They just they just they're standing and then they collapse. 0:07:31.080 --> 0:07:33.520 And there's something comic about that, you know that you 0:07:33.640 --> 0:07:38.000 can't deny that it's funny, but it also demonstrates that 0:07:38.080 --> 0:07:40.680 these robots are trying to do something that while we 0:07:40.800 --> 0:07:44.480 as human beings might find easy, it's a real challenge 0:07:44.720 --> 0:07:50.800 from a robotics standpoint. So this Darker Robotics Challenge had 0:07:50.840 --> 0:07:55.720 three phases. The first one was a virtual challenge, which 0:07:56.360 --> 0:08:00.200 in which teams were to design software that would allow 0:08:00.400 --> 0:08:04.440 a virtual robot to complete um certain tasks within a 0:08:04.520 --> 0:08:08.800 virtual environment. So there were no real robotics here, but 0:08:08.880 --> 0:08:11.679 it was a test to see if if the teams 0:08:11.680 --> 0:08:15.160 could actually build the software necessary to make the robot 0:08:15.240 --> 0:08:19.480 do the tasks that needed to be done. The next 0:08:20.000 --> 0:08:23.200 had a trial where it was a physical trial where 0:08:23.560 --> 0:08:28.320 teams actually had to uh place a real robot through 0:08:28.600 --> 0:08:31.600 to you know, go to the next level. And then 0:08:31.640 --> 0:08:33.400 you had the finals, which were the ones that have 0:08:33.720 --> 0:08:37.240 happened most recently. Uh, and I'll go through each of 0:08:37.320 --> 0:08:39.560 the tasks that they had to do in just a second. 0:08:40.559 --> 0:08:42.840 If we look back at the Virtual challenge, there were 0:08:42.840 --> 0:08:46.000 twenty six teams that were part of it to build 0:08:46.040 --> 0:08:49.880 that software. The robot they were controlling was a virtual 0:08:50.040 --> 0:08:55.840 Atlas robot, which is a humanoid robot and was a 0:08:55.880 --> 0:08:58.240 popular choice for a lot of the teams. When they 0:08:58.240 --> 0:09:02.560 moved on to the more physical trials, they still would 0:09:02.640 --> 0:09:04.679 use the Atlas robots. The the challenge, by the way, 0:09:04.720 --> 0:09:07.880 did not require the teams to build their own robots 0:09:07.920 --> 0:09:11.520 from scratch. They could use pre existing robots, but they 0:09:11.600 --> 0:09:14.280 had to program them so that they could complete the 0:09:14.320 --> 0:09:17.520 necessary tasks. So not every robot out there would have 0:09:17.559 --> 0:09:19.520 been a good fit for this. Let's go back to 0:09:19.559 --> 0:09:22.160 the Roomba. The roombo would have been a terrible choice 0:09:22.200 --> 0:09:26.000 to enter into the Darbo Robotics Challenge. We'll be back 0:09:26.040 --> 0:09:28.679 with more of this classic episode of tech Stuff after 0:09:28.720 --> 0:09:39.520 this quick break. So, uh, here are some of the 0:09:39.640 --> 0:09:44.959 virtual tasks that the virtual robots had to complete using 0:09:45.000 --> 0:09:48.320 this software. They had to walk through an area of 0:09:48.720 --> 0:09:53.800 uneven ground that had debris included there, so simulated debris 0:09:53.920 --> 0:09:56.319 had to have a software that would allow the robot 0:09:56.360 --> 0:10:00.120 to maintain its balance even while tracking a path to 0:10:00.160 --> 0:10:04.319 a specific destination. Through this this uneven terrain, they had 0:10:04.360 --> 0:10:08.640 to attach a hose to a spigot and turn a valve. Uh. 0:10:09.040 --> 0:10:14.400 They also were given some artificial limits because and an 0:10:14.440 --> 0:10:18.720 emergency situation, you cannot always count on your communication lines 0:10:19.120 --> 0:10:24.960 being perfectly sound, such as that the Fukushima disaster. You 0:10:25.200 --> 0:10:28.600 need to be sure that your robot can contend with 0:10:28.640 --> 0:10:31.040 the fact that there might be lag between when you 0:10:31.080 --> 0:10:34.079 can issue a command and when the robot is ready 0:10:34.120 --> 0:10:37.240 for the next command. UH. That meant that some of 0:10:37.280 --> 0:10:40.360 the teams started to really look at ways to create 0:10:40.520 --> 0:10:44.800 a semi autonomous robot. So some of the teams built 0:10:44.880 --> 0:10:51.240 robots that were capable of taking on some tasks autonomously. UH. 0:10:51.320 --> 0:10:55.800 Some teams avoided that entirely. They focused on using the 0:10:55.920 --> 0:11:00.640 robots as a direct extension of the control roles that 0:11:00.720 --> 0:11:03.240 the humans were behind. So, in other words, you might 0:11:03.280 --> 0:11:09.320 think of those robots as enormous technological puppets. The puppets 0:11:09.320 --> 0:11:14.040 would respond to direct human commands and wouldn't be able 0:11:14.080 --> 0:11:17.040 to do anything on their own. They didn't have any autonomy. 0:11:17.720 --> 0:11:20.280 There are other ones that you know, you could have 0:11:20.360 --> 0:11:22.559 semi autonomy in the show it a set of stairs 0:11:22.640 --> 0:11:26.080 and you send it the command to climb those stairs, 0:11:26.160 --> 0:11:28.440 and it could do the rest. It could calculate how 0:11:28.520 --> 0:11:30.480 high it needed to put its feet and how to 0:11:30.480 --> 0:11:33.640 shift its weight, that sort of stuff. UH, And there 0:11:33.679 --> 0:11:36.600 were a lot of different strategies employed and to varying 0:11:36.600 --> 0:11:41.439 degrees of success. It wasn't like the autonomous robots automatically 0:11:41.440 --> 0:11:44.040 were better than the human controlled or vice versa. There 0:11:44.120 --> 0:11:48.120 was actually it all depended on the implementation. So that 0:11:48.200 --> 0:11:57.960 virtual trial ended up having one team winning pretty pretty decisively. UM. 0:11:58.000 --> 0:12:02.360 It was the Institute for Human and Machine Cognition, which 0:12:02.440 --> 0:12:07.320 is UH, an organization in Pensacola, Florida, and eight other 0:12:07.360 --> 0:12:10.360 teams qualified for the trials that we're going to be 0:12:10.400 --> 0:12:16.320 held in December. Now, by the time those trials happened, 0:12:17.080 --> 0:12:21.920 some of the teams had merged UH and some other 0:12:22.000 --> 0:12:24.600 teams were coming into it. Six of the teams would 0:12:24.600 --> 0:12:27.280 go on to compete at the trials that were held 0:12:27.320 --> 0:12:32.160 at the Homestead Miami Speedway, and they had eight tasks 0:12:32.200 --> 0:12:34.720 they had to complete that were similar to the ones 0:12:34.760 --> 0:12:40.400 in the Darker robotics final, and those tasks included manning 0:12:40.400 --> 0:12:45.600 a vehicle, um walking through uneven terrain, or otherwise moving 0:12:45.640 --> 0:12:49.280 through uneven terrain. Robots did not have to be bipedal. 0:12:49.840 --> 0:12:52.480 They could move around on however many limbs they had. 0:12:52.520 --> 0:12:54.480 They just had to be also able to do these 0:12:54.480 --> 0:12:57.480 other tasks. Uh. They had to be able to climb 0:12:57.480 --> 0:12:59.720 a ladder. They had to be able to remove debris 0:12:59.760 --> 0:13:02.120 from a door and then opened the door. They had 0:13:02.200 --> 0:13:04.840 to be able to break through a wall. Uh, they 0:13:04.880 --> 0:13:07.439 had to be able to handle valves and to attach 0:13:07.480 --> 0:13:11.480 a hose to a spigott. The winning team of that 0:13:12.080 --> 0:13:16.320 set of trials was a group called Shaft spelled s 0:13:16.360 --> 0:13:20.520 C H A f T. This was a group out 0:13:20.520 --> 0:13:24.560 of Japan. Now, when this group was competing, they were 0:13:24.760 --> 0:13:28.560 this little independent group. But then they got acquired by 0:13:28.600 --> 0:13:31.440 a little company that we've talked about several times on 0:13:31.480 --> 0:13:34.760 this podcast. You know him. You may or may not 0:13:34.840 --> 0:13:38.520 love them Google. So just before the trials, Google had 0:13:38.520 --> 0:13:44.240 acquired uh, this this this group, this organization. Now Shaft 0:13:44.280 --> 0:13:47.719 did really well, uh, which you know makes sense. I mean, 0:13:47.760 --> 0:13:50.840 you've you've heard the theme song. But out of thirty 0:13:50.840 --> 0:13:55.160 two points that it could possibly score. It earned twenty seven, 0:13:56.160 --> 0:13:59.680 so pretty good score for for a robot and also 0:14:00.240 --> 0:14:05.120 was seven points more than the next closest competitor, which 0:14:05.200 --> 0:14:07.920 was from I H M C, the same group that 0:14:07.960 --> 0:14:13.120 won the virtual trial earlier. Third place was Tartan Rescue 0:14:13.480 --> 0:14:17.280 from Carnegie Mellon And that will be important in a 0:14:17.320 --> 0:14:21.080 second to when we get into the finals. Now, Shaft 0:14:21.240 --> 0:14:24.480 did not go on to compete in the DARPA Robotics 0:14:24.520 --> 0:14:28.880 Finals because Google announced that they were refocusing the team, 0:14:28.960 --> 0:14:34.960 dedicating it to actual commercial Google projects. So uh, they 0:14:35.040 --> 0:14:38.800 were no longer being dedicated to this challenge. They were 0:14:38.800 --> 0:14:41.920 being dedicated to a real world product of some sort. 0:14:43.120 --> 0:14:49.760 Google's also had other involvement with some DARPA stuff, at 0:14:49.800 --> 0:14:53.040 least after the fact. Google does not tend to get 0:14:53.080 --> 0:14:57.640 involved in these challenges directly. Um, probably because of the 0:14:57.760 --> 0:15:02.720 military association of DARPA and Google is very careful to 0:15:02.880 --> 0:15:07.240 avoid those kind of associations. Now, the finals took place 0:15:07.720 --> 0:15:12.760 over June five and June six and Pomona, California, and uh, 0:15:12.800 --> 0:15:16.640 it was at the Fair Plex and there was quite 0:15:16.640 --> 0:15:20.280 a crowd watching these events, and the reports I read 0:15:20.320 --> 0:15:22.760 were really interesting, and also the videos I watched were 0:15:22.800 --> 0:15:27.440 interesting because the crowd was extremely enthusiastic cheering on the robots, 0:15:28.080 --> 0:15:32.280 groaning every time a robot failed a task or fell over, 0:15:33.440 --> 0:15:36.800 gasping when that sort of stuff happened, because everyone wanted 0:15:36.840 --> 0:15:40.520 to see these robots succeed. No one wanted any group 0:15:40.560 --> 0:15:44.840 to fail. And uh. Also, the other interesting thing was 0:15:44.920 --> 0:15:47.560 that it was really hard to tell if you were 0:15:47.560 --> 0:15:52.080 in the audience when a robot was operating autonomously as 0:15:52.080 --> 0:15:56.640 opposed to being controlled by humans. Um. That says two things, 0:15:56.880 --> 0:16:00.320 One that the line between these two is getting further 0:16:00.360 --> 0:16:05.520 and further blurred. And to that sometimes the robot would 0:16:05.560 --> 0:16:07.480 behave in a weird way, and you couldn't be sure 0:16:07.520 --> 0:16:11.520 if it was because the autonomous programming was lacking or 0:16:11.640 --> 0:16:16.160 because the control mechanisms were limited. So, in other words, 0:16:16.880 --> 0:16:19.640 it's not that necessarily that the art has gotten so 0:16:19.680 --> 0:16:21.800 advanced that we can't tell the difference. It may be 0:16:21.960 --> 0:16:25.000 that the art has not advanced enough that we can't 0:16:26.120 --> 0:16:31.480 spot what is the cause of incompetence. And and I 0:16:31.600 --> 0:16:34.720 use the word incompetence mainly as a joke. It's just 0:16:34.840 --> 0:16:40.440 a joke because truthfully, it displays how difficult robotics as 0:16:40.440 --> 0:16:43.480 a discipline actually is how how challenging it is to 0:16:43.560 --> 0:16:46.840 design a robot that is capable of doing the same 0:16:46.880 --> 0:16:50.440 sort of things that humans can do. It shows that 0:16:50.520 --> 0:16:54.160 the things that we take for granted as being pretty 0:16:54.200 --> 0:16:59.040 simple are fiendishly difficult when you get to a design 0:16:59.120 --> 0:17:02.000 and programming level when you're actually building a robot that's 0:17:02.040 --> 0:17:04.840 going to be capable of doing the same thing. So 0:17:04.920 --> 0:17:09.320 here are the eight tasks that the robots had to complete, 0:17:09.400 --> 0:17:12.560 and one of these tasks was called Surprise. I'll get 0:17:12.600 --> 0:17:15.600 to it. But the reason it's called surprises that DARPA 0:17:16.000 --> 0:17:20.080 told all the teams, your robot will encounter a task 0:17:20.400 --> 0:17:24.600 similar to these other ones that you already know about, 0:17:25.160 --> 0:17:26.639 but we're not going to tell you what it is. 0:17:27.720 --> 0:17:30.720 And it meant that the teams knew that there was 0:17:30.760 --> 0:17:34.120 going to be something on that list of tasks that 0:17:34.560 --> 0:17:38.600 was not not defined, and that the robot would have 0:17:38.640 --> 0:17:40.399 to be able to contend with it in order to 0:17:40.440 --> 0:17:44.520 get a point for that particular task. And so the 0:17:44.640 --> 0:17:47.040 tasks were um there were eight, Like I said, eight 0:17:47.040 --> 0:17:50.600 of them. You were awarded a point if your robot 0:17:50.600 --> 0:17:55.680 could successfully complete that task. So the score the final 0:17:55.760 --> 0:17:59.640 scores once everything was done were determined by how many 0:17:59.680 --> 0:18:03.879 tasks were completed successfully and what was the time of 0:18:03.960 --> 0:18:09.119 the robots performance. Now, when they were first designing the robots, 0:18:09.440 --> 0:18:12.240 teams were told they would have half an hour per task. 0:18:12.640 --> 0:18:15.160 So you have eight tasks, half an hour per test, 0:18:15.240 --> 0:18:18.639 that's four hours total. But the actual competition they were 0:18:18.640 --> 0:18:23.359 told they would have one hour to complete all eight tests. Um. 0:18:23.440 --> 0:18:26.280 And why you might think that, how is that fair? 0:18:26.320 --> 0:18:28.639 As a as a bait and switch, you also have 0:18:28.720 --> 0:18:33.080 to understand this was all about simulating an emergency response. 0:18:34.200 --> 0:18:38.320 So under those emergency conditions, you can't expect to ask 0:18:38.359 --> 0:18:41.879 for more time. That's not realistic. Uh. And it added 0:18:41.920 --> 0:18:46.720 an extra layer of pressure on the teams. So those 0:18:46.760 --> 0:18:49.760 eight tasks, the first one was to drive, so the 0:18:49.840 --> 0:18:53.800 robots had to drive a Polaris Ranger XP nine vehicle. 0:18:54.560 --> 0:18:56.760 These if you haven't seen them, they look like sometimes 0:18:56.760 --> 0:18:59.080 you might call mcgator or you might think of it 0:18:59.119 --> 0:19:03.600 as a golf car on steroids. These are um vehicles 0:19:03.680 --> 0:19:06.600 that are similar to golf carts. They're they're largely open, 0:19:06.640 --> 0:19:10.320 They've got heavier wheels than golf carts, do more warm 0:19:10.400 --> 0:19:14.520 horsepower than a golf cart would, but it's definitely in 0:19:14.640 --> 0:19:18.760 that range between golf cart and real car. Teams had 0:19:18.800 --> 0:19:23.960 five minutes to alter the vehicle without using tools to 0:19:24.080 --> 0:19:27.280 make sure that their robots could actually operate it. So 0:19:27.320 --> 0:19:29.760 what the robots had to do was drive from the 0:19:29.840 --> 0:19:35.320 starting point to a destination, and it was only considered 0:19:35.480 --> 0:19:38.359 a success if the vehicle had completely moved past a 0:19:38.560 --> 0:19:43.320 finishing mark. The vehicle also had to go through essentially 0:19:43.320 --> 0:19:47.560 a driving course with obstacles and cones set up, and 0:19:47.600 --> 0:19:51.240 it was determined that if the robot were to collide 0:19:51.320 --> 0:19:53.200 with one of those obstacles or one of those cones 0:19:53.240 --> 0:19:56.200 and cause it to move as a result of that collision, 0:19:56.720 --> 0:20:00.359 the robot would not receive a point for that ask 0:20:01.880 --> 0:20:05.000 the robot. Also, the team could choose for the robot too, 0:20:05.119 --> 0:20:08.959 instead of driving, to walk to the destination, but in 0:20:08.960 --> 0:20:11.439 that case they would not be awarded a point for 0:20:11.480 --> 0:20:16.400 that task. They would forfeit the point um. But ideally 0:20:16.440 --> 0:20:18.600 the robot would be able to operate the accelerator and 0:20:18.680 --> 0:20:22.439 the steering wheel and maybe even the break and the shift. 0:20:23.160 --> 0:20:26.920 The cars were already or the vehicles were already started, 0:20:26.920 --> 0:20:31.119 The engines were already running, and they were already in 0:20:31.240 --> 0:20:34.359 high gear. Because it was considered to be the smoothest 0:20:34.680 --> 0:20:38.639 way for the robots to to maneuver the vehicle. But 0:20:39.000 --> 0:20:41.720 if teams wanted to, they could design a robot that 0:20:41.760 --> 0:20:45.600 would be capable of shifting gears. It was not a requirement, 0:20:45.600 --> 0:20:47.560 it was just something they could do if they wanted to. 0:20:48.720 --> 0:20:51.199 At any rate, the robot needed to be able to 0:20:51.280 --> 0:20:55.439 drive safely from the starting point to the destination and 0:20:55.480 --> 0:20:59.280 then halt the vehicle either by letting off of the 0:20:59.280 --> 0:21:01.919 accelerators so it coasted to a stop, are actually applying 0:21:01.920 --> 0:21:05.880 the brake. The next task. That was just task number one. 0:21:06.440 --> 0:21:09.240 The next task was called egress, which is really just 0:21:09.440 --> 0:21:12.960 getting out of the car. It sounds incredibly simple, and 0:21:12.960 --> 0:21:15.520 for humans, for most of us, it really is pretty simple. 0:21:16.119 --> 0:21:20.000 You know. We we intuitively know how to maintain our 0:21:20.000 --> 0:21:22.800 balance and to shift our weight so that we go 0:21:22.880 --> 0:21:26.080 from a seated position to standing. But that's not the 0:21:26.119 --> 0:21:28.679 case with the robots. You have to design the robot 0:21:28.760 --> 0:21:30.679 to do that. You have to program the robot to 0:21:30.680 --> 0:21:34.160 do that. You know, the robot be able to anticipate 0:21:34.600 --> 0:21:37.560 what a shift in its weight will do, what how 0:21:37.600 --> 0:21:40.920 it's momentum will carry it forward. As it turns out, 0:21:40.920 --> 0:21:44.520 This was one of the trickier tasks that teams had 0:21:44.520 --> 0:21:47.159 to complete to get out of a vehicle. It was 0:21:47.200 --> 0:21:52.320 not an easy thing to do um but they were 0:21:52.400 --> 0:21:54.000 told that they could get out of the vehicle in 0:21:54.040 --> 0:21:56.400 either direction. They didn't have to exit outll the left 0:21:56.440 --> 0:21:58.760 side versus the right side. That the robot could get 0:21:58.800 --> 0:22:02.880 out of either side, and again the task was considered 0:22:02.920 --> 0:22:05.919 complete if the robot could get out of the vehicle 0:22:05.960 --> 0:22:09.520 and maintain its balance and move on to the next challenge, 0:22:09.880 --> 0:22:12.919 and if they could, it would receive a point. The 0:22:13.000 --> 0:22:17.199 next one was door. Now egress was tough. Door was 0:22:17.240 --> 0:22:21.320 surprisingly tough. You would think that opening a door would 0:22:21.359 --> 0:22:24.040 be a pretty simple task. This was a push door 0:22:24.119 --> 0:22:27.080 operated by a lever style handle. You had to push 0:22:27.160 --> 0:22:29.680 up or pull you know, pull up or push down 0:22:29.840 --> 0:22:32.919 rather on the handle and then push the door to 0:22:33.000 --> 0:22:36.440 open it, then step through the door. The door did 0:22:36.440 --> 0:22:38.399 not have a threshold, so there was nothing that you 0:22:38.440 --> 0:22:42.080 had to step over. However, the door was a standard 0:22:42.119 --> 0:22:45.520 thirty six inch wide doorway and with the door jam 0:22:45.600 --> 0:22:48.320 that's closer to thirty three and a half inches, and 0:22:48.359 --> 0:22:51.080 some of these robots were too wide to walk through 0:22:51.119 --> 0:22:54.920 the door, you know, heading with their torso facing forward. 0:22:55.160 --> 0:22:58.200 They actually had to turn sideways and then shuffle through 0:22:58.240 --> 0:23:01.119 the doorway. And there were a lot of robots that 0:23:01.200 --> 0:23:07.600 fell down at this stage of the the challenge. They 0:23:08.040 --> 0:23:10.639 would lose their balance either when pushing the door or 0:23:10.720 --> 0:23:15.560 when trying to maneuver through the doorway. One robot fell 0:23:15.600 --> 0:23:18.320 down at this stage and was able to pick itself 0:23:18.359 --> 0:23:24.280 back up very very slowly and deliberately, which is remarkable 0:23:24.320 --> 0:23:28.800 because it's the only robot that managed to pick itself 0:23:28.840 --> 0:23:31.360 back up after falling over, and that was the Carnegie 0:23:31.440 --> 0:23:35.000 Melon robot, the same one we mentioned in the Trials 0:23:35.160 --> 0:23:39.320 part in just a few moments ago. But anyway, that 0:23:39.359 --> 0:23:40.720 was it. You just had to open the door and 0:23:40.760 --> 0:23:42.520 walk through the doorway, and that was the end of 0:23:42.560 --> 0:23:46.320 that task, and yet it was deceptively difficult. The next 0:23:46.320 --> 0:23:51.600 one was Valve, which was not about the game company, 0:23:51.640 --> 0:23:54.800 but rather about turning a control as if you were 0:23:54.840 --> 0:23:58.359 turning a valve to operate a fire hose or to 0:23:58.960 --> 0:24:01.760 shut down water to a specific part of the system, 0:24:01.800 --> 0:24:04.720 which makes sense. In a nuclear facility, you might have 0:24:04.880 --> 0:24:08.200 to shut off or open a valve. In that case, 0:24:09.720 --> 0:24:13.480 the valve had to be turned counterclockwise in three sixty 0:24:13.560 --> 0:24:16.320 degrees in order for the task to be complete, and 0:24:16.359 --> 0:24:19.159 the team was only told that the valve would have 0:24:19.320 --> 0:24:23.399 a size between four and sixteen inches in diameter, which 0:24:23.400 --> 0:24:25.920 meant that you had to create a robot that would 0:24:25.920 --> 0:24:29.399 be capable of gripping anything within that range and then 0:24:29.600 --> 0:24:34.240 turning it in that counterclockwise direction three sixty degrees. Again 0:24:34.280 --> 0:24:37.560 triggier than it sounds. We've got more to say in 0:24:37.600 --> 0:24:40.920 this classic episode of tech Stuff. After these quick messages 0:24:48.760 --> 0:24:51.600 the next step. The next task was called wall, in 0:24:51.600 --> 0:24:53.720 which a robot had to pick up a cordless drill. 0:24:53.920 --> 0:24:57.760 There were two of them available. They were not automatically on. 0:24:57.840 --> 0:25:00.040 The robot had to operate the cordless drill and and 0:25:00.160 --> 0:25:04.880 it on by squeezing the trigger, and use the cordless 0:25:04.960 --> 0:25:08.320 drill to cut through some drywall in a shape that 0:25:08.440 --> 0:25:12.120 was drawn on the drywall. The drywall was a half 0:25:12.119 --> 0:25:15.600 inch thick, and the robot had to hold the drill, 0:25:15.960 --> 0:25:18.440 operate the drill and move in this shape and then 0:25:18.480 --> 0:25:21.639 remove the rubble. The idea being that sometimes the robot 0:25:21.760 --> 0:25:25.200 might have to cut through a surface in order to 0:25:25.200 --> 0:25:28.840 get access to certain controls uh that might otherwise be 0:25:28.880 --> 0:25:33.119 blocked by collapse of a room or something along those lines. 0:25:34.720 --> 0:25:37.600 Also pretty tricky, you know, using tools that were designed 0:25:37.640 --> 0:25:40.640 to be held in human hands. You know, humans, we 0:25:40.640 --> 0:25:43.400 we have the ability to detect how much pressure we're 0:25:43.480 --> 0:25:48.760 using and to um to change that based upon what's happening. 0:25:49.119 --> 0:25:51.000 Robots a little more tricky. I mean, you can have 0:25:51.040 --> 0:25:56.159 sensors that alert the operator what's happening to what's happening, 0:25:56.160 --> 0:25:59.159 but it's not intuitive. Again, you have to program it. 0:26:00.520 --> 0:26:04.359 After wall was the surprise. Now, the surprise was something 0:26:04.400 --> 0:26:07.280