WEBVTT - The 2007 Urban Challenge 0:00:04.120 --> 0:00:07.160 Get in touch with technology with tech Stuff from how 0:00:07.200 --> 0:00:13.920 stuff works dot com. Hey there, and welcome to tech Stuff. 0:00:13.960 --> 0:00:17.239 I'm your host, Jonathan Strickland. I'm an executive producer with 0:00:17.320 --> 0:00:19.239 How Stuff Works and my Heart Radio and I love 0:00:19.360 --> 0:00:23.720 all things tech. And this is the third episode in 0:00:23.760 --> 0:00:27.160 a series about driver less cars. In the first two episodes, 0:00:27.200 --> 0:00:31.120 I covered the early concepts, which mostly involved building out 0:00:31.160 --> 0:00:35.120 an infrastructure that's outside of cars, like a smart super 0:00:35.240 --> 0:00:37.440 highway that would guide traffic to where it would need 0:00:37.479 --> 0:00:40.000 to go. And I also talked about the first two 0:00:40.080 --> 0:00:43.839 Grand Challenges that were sponsored by DARPA, that's the R 0:00:43.880 --> 0:00:47.520 and D division of the Department of Defense. Darpest goal 0:00:47.680 --> 0:00:52.600 was to create an incentive for engineers, researchers, computer scientists 0:00:52.600 --> 0:00:56.440 to to innovate in the space in that autonomous vehicle space, 0:00:56.960 --> 0:01:00.160 because the United States Congress had decided that a third 0:01:00.200 --> 0:01:04.280 of all military ground combat vehicles should be autonomous by 0:01:04.319 --> 0:01:08.760 the year and progress was just not going at the 0:01:08.800 --> 0:01:12.160 speed that was needed to make that happen. So dark 0:01:12.240 --> 0:01:16.559 As assessment was that the usual defense contractors just weren't 0:01:16.600 --> 0:01:19.760 making progress fast enough that that they needed to have 0:01:19.920 --> 0:01:24.240 some other way of getting new ideas, new blood into 0:01:24.280 --> 0:01:25.960 all of this, So the Grand Challenge was a way 0:01:25.959 --> 0:01:29.240 to invite numerous other engineers to solve these really difficult 0:01:29.280 --> 0:01:33.679 engineering problems. The two thousand four challenge saw no winner. 0:01:34.080 --> 0:01:36.840 No one was able to take home the prize money. 0:01:36.880 --> 0:01:39.959 The best performing team made it less than seven and 0:01:40.000 --> 0:01:43.600 a half miles through a one forty two mile course. 0:01:43.920 --> 0:01:46.759 But the two thousand five challenge turned out much better. 0:01:46.920 --> 0:01:51.360 Five teams completed the full course and one Stanford's racing 0:01:51.400 --> 0:01:55.520 team took the top prize that year, which was more 0:01:55.560 --> 0:01:58.640 than a million dollars. Now, there was no question that 0:01:58.720 --> 0:02:03.880 the challenges had encouraged invention and innovation, from creative computer 0:02:03.920 --> 0:02:07.360 algorithms that would guide an autonomous car through decisions on 0:02:07.440 --> 0:02:11.320 wind and how to act, to designing all new hardware 0:02:11.600 --> 0:02:14.720 that would be used to sense the environment. These challenges 0:02:14.840 --> 0:02:19.640 had pushed technological development considerably. But the driverless cars of 0:02:19.680 --> 0:02:23.840 two thousand five were meant to cross a desert course 0:02:24.160 --> 0:02:27.960 sort of an off road course. There was tricky terrain 0:02:28.040 --> 0:02:29.960 to cross, and the cars had to make some really 0:02:30.000 --> 0:02:32.600 complex maneuvers in a few cases, but there was no 0:02:32.720 --> 0:02:35.560 need to compensate for the types of stuff that your 0:02:35.600 --> 0:02:39.920 typical human driver encounters every day. Namely, there was no 0:02:40.000 --> 0:02:43.760 need to worry about traffic laws or traffic for that matter, 0:02:44.200 --> 0:02:45.760 and it was important for the cars to be able 0:02:45.760 --> 0:02:48.359 to detect obstacles in order to work their way around them, 0:02:48.360 --> 0:02:51.400 but there was less worry about more specific challenges like 0:02:51.720 --> 0:02:55.919 navigating through a city that has lots of potential moving 0:02:56.120 --> 0:03:01.320 obstacles pedestrians, bicyclists, other cars. So for the next challenge, 0:03:01.360 --> 0:03:04.360 which would take place in two thousand seven, DARPA wanted 0:03:04.360 --> 0:03:06.519 to put teams to the test and have them design 0:03:06.560 --> 0:03:10.320 a car that can navigate through a simulated urban environment. 0:03:10.880 --> 0:03:13.400 Just as the agency had increased the prize money from 0:03:13.440 --> 0:03:16.720 two thousand four to two thousand and five, the the 0:03:16.760 --> 0:03:19.359 prize kitty was a million dollars for two thousand four, 0:03:19.360 --> 0:03:21.800 it was two million for two thousand five. The urban 0:03:21.919 --> 0:03:24.200 version of the challenge up to it again to three 0:03:24.240 --> 0:03:28.240 and a half million dollars in prize money total. DARPA 0:03:28.360 --> 0:03:31.880 offered two different routes to enter the Urban Challenge. You 0:03:31.919 --> 0:03:37.400 could apply using one of two different approaches. They called 0:03:37.400 --> 0:03:41.720 it Track A and Track B. A team submitting to 0:03:41.760 --> 0:03:46.040 participate in Track A would be eligible to receive funding 0:03:46.240 --> 0:03:50.080 from DARPA of up to a million dollars, but to qualify, 0:03:50.120 --> 0:03:52.440 the team would have to submit a detailed proposal for 0:03:52.520 --> 0:03:56.400 review and then, if selected, they would have to demonstrate 0:03:56.560 --> 0:04:00.840 the progress of the experiment of the prototype at four 0:04:00.960 --> 0:04:04.920 different milestone events in order to still be eligible for 0:04:05.040 --> 0:04:08.520 that program. If the team could not meet the standards 0:04:08.560 --> 0:04:11.839 that DARPA had set, then they would be removed from 0:04:11.880 --> 0:04:16.240 Track A and all financial support would stop. Sixty five 0:04:16.360 --> 0:04:21.799 teams submitted for Track A consideration, DARPA would select eleven 0:04:21.839 --> 0:04:26.600 of them to participate. Track B teams were self supported, 0:04:26.800 --> 0:04:30.200 meaning they were responsible for securing the funding necessary for 0:04:30.440 --> 0:04:34.920 technology development, execution, that sort of stuff. Those teams could 0:04:34.920 --> 0:04:38.560 seek out funding from sponsors and from other sources. So 0:04:38.680 --> 0:04:42.960 Track B teams also had to participate in certain qualifying 0:04:43.040 --> 0:04:46.520 events in order to continue the competition. There were site visits, 0:04:46.520 --> 0:04:49.160 and then there was a big qualifier that a lot 0:04:49.200 --> 0:04:51.200 of teams had to participate, and I'll talk about that 0:04:51.400 --> 0:04:54.400 a little bit later. Now, those qualifiers were meant to 0:04:54.400 --> 0:04:57.760 demonstrate that the respective teams vehicles could have a decent 0:04:57.880 --> 0:05:02.000 chance of completing a competition's task on the actual day 0:05:02.000 --> 0:05:05.680 of the final event. So It's really all about whittling 0:05:05.800 --> 0:05:10.440 down the competitors to the ones most likely to achieve success. 0:05:11.360 --> 0:05:14.240 Track B also had several other additional rules they had 0:05:14.279 --> 0:05:18.440 to follow. This is from DARPA's actual official rule book 0:05:18.720 --> 0:05:21.720 for the two thousand seven Urban Challenge. So, for example, 0:05:21.880 --> 0:05:25.560 the team leader of a Track B team would be 0:05:26.040 --> 0:05:30.039 ineligible to participate in any other Urban Challenge team. They 0:05:30.080 --> 0:05:33.279 could only work on their own team that they were leading. 0:05:34.160 --> 0:05:37.680 If you were not team leader, you could switch teams. 0:05:37.880 --> 0:05:40.159 You could work on one team and then switch to 0:05:40.200 --> 0:05:42.520 another one. Let's say that you're working on one team 0:05:42.560 --> 0:05:48.280 and it is unable to meet the qualifications that are 0:05:48.320 --> 0:05:50.960 necessary at a certain milestone, then you would be allowed 0:05:51.000 --> 0:05:54.559 to switch to another team if another team wanted to. Also, 0:05:54.600 --> 0:05:55.960 if you were a team leader, you had to be 0:05:55.960 --> 0:05:57.919 at least twenty one years old, and you had to 0:05:57.920 --> 0:06:01.760 be a US citizen. Non citizens could participate on teams, 0:06:01.800 --> 0:06:04.719 but they could not be team leaders. This is another 0:06:04.760 --> 0:06:12.880 reminder that ultimately this competition was keyed into national defense. 0:06:14.040 --> 0:06:17.640 The spirit of the competition was very jovial from what 0:06:17.720 --> 0:06:21.719 I can understand, it was very cooperative. But when you 0:06:21.800 --> 0:06:25.080 take a step back and you look at the overall picture. 0:06:25.160 --> 0:06:29.679 You remember, Oh, this is so that DARPA can develop 0:06:29.720 --> 0:06:33.840 the technologies or can can encourage the development of the 0:06:33.839 --> 0:06:39.360 technologies that will be needed to power the future ground 0:06:39.440 --> 0:06:44.240 combat vehicles. Unlike the previous Grand Challenges, DARPA even allowed 0:06:44.279 --> 0:06:47.920 teams to secure funding from government sources the previous ones. 0:06:47.960 --> 0:06:50.160 If you listen to my last episode, you know they 0:06:50.160 --> 0:06:53.680 said you couldn't get money from federal sources, but all 0:06:53.720 --> 0:06:57.400 such funding had to be explicitly approved and authorized by 0:06:57.440 --> 0:07:01.480 whatever the respective government department was for the use of 0:07:01.520 --> 0:07:05.839 this Grand Challenge. Government sponsored teams were, however, prevented from 0:07:05.920 --> 0:07:10.320 using any technology or information that was under any kind 0:07:10.320 --> 0:07:14.400 of classified status. So you couldn't use any top secret 0:07:14.480 --> 0:07:19.040 stuff because that was against the rules. I mentioned briefly 0:07:19.440 --> 0:07:22.280 what the overall goal of the Urban Challenge was, this 0:07:22.360 --> 0:07:26.560 idea of maneuvering through a simulated urban environment. But I 0:07:26.600 --> 0:07:29.160 think it's more helpful if we actually go through all 0:07:29.240 --> 0:07:32.200 the elements that DARPA spelled out in the official rules. 0:07:32.240 --> 0:07:35.880 This is what was sent to any interested party that 0:07:36.000 --> 0:07:39.000 was applying to be part of the Urban Challenge. This 0:07:39.080 --> 0:07:42.920 illustrates exactly how challenging the whole thing ended up being 0:07:43.040 --> 0:07:45.680 and the specific objectives of the Grand Challenge were and 0:07:45.760 --> 0:07:49.600 these are all quotes from the rule book. Complete a 0:07:49.600 --> 0:07:53.840 mission defined by an ordered series of checkpoints in a 0:07:54.000 --> 0:07:57.760 complex route network. The vehicle will have five minutes to 0:07:57.800 --> 0:08:02.880 process a mission description, but for attempting the course, interpret 0:08:03.080 --> 0:08:07.840 static lane markings as in white or yellow lines provided 0:08:07.880 --> 0:08:12.080 with the Route Network Definition file that's R in DF, 0:08:12.600 --> 0:08:16.800 and behave in accordance with applicable traffic laws and conventions. 0:08:17.240 --> 0:08:19.400 Darp as intent is for the R and d F 0:08:19.720 --> 0:08:23.960 lane boundary descriptors to match the physical lane markings on 0:08:24.080 --> 0:08:27.440 the ground. DARPA cannot ensure that this will be the 0:08:27.480 --> 0:08:30.240 case in all areas, and as such, the R n 0:08:30.320 --> 0:08:33.920 DF will take precedence over the physical ground markings in 0:08:34.000 --> 0:08:38.440 conflicting areas. So, in other words, DARPA would provide to 0:08:38.520 --> 0:08:43.280 each team a digital file that would represent the area 0:08:43.640 --> 0:08:49.320 the features of the course. It wouldn't give information about 0:08:50.080 --> 0:08:54.280 where anyone was expected to travel, but it would give 0:08:54.520 --> 0:08:57.920 a layout of all the different roads in that course, 0:08:58.600 --> 0:09:04.240 and according to that file, you would program your autonomous 0:09:04.240 --> 0:09:08.679 cars behaviors to follow within those guidelines, and if the 0:09:08.720 --> 0:09:11.080 car were to come upon part of the road that 0:09:11.200 --> 0:09:15.240 was not in accordance with that same digital file, it 0:09:15.280 --> 0:09:17.880 would defer to the digital file as opposed to the 0:09:17.920 --> 0:09:22.240 actual conditions the road. So if in one case a 0:09:22.320 --> 0:09:26.680 lane is is shown as being in one orientation and 0:09:26.720 --> 0:09:29.480 it turns out it's slightly off in reality, you go 0:09:29.559 --> 0:09:33.360 with the file, not with what reality is. This reminds 0:09:33.400 --> 0:09:36.600 us that this is in fact a simulation, because that 0:09:36.679 --> 0:09:39.520 does not fly in a real world situation. You have 0:09:39.600 --> 0:09:42.880 to conform to what is really there for all drivers. 0:09:43.280 --> 0:09:47.520 Back to the rules, the vehicles to exhibit context dependent 0:09:47.640 --> 0:09:52.280 speed control to ensure safe operation, including adherents to speed limits. 0:09:52.640 --> 0:09:56.560 It's to exhibit safe following behavior when approaching other vehicles 0:09:56.559 --> 0:09:59.680 from behind in a traffic lane, which includes maintaining a 0:09:59.760 --> 0:10:04.160 say following distance. Is to exhibit safe check and go 0:10:04.360 --> 0:10:07.760 behavior when pulling around a stopped vehicle, pulling out of 0:10:07.760 --> 0:10:11.680 a parking spot, moving through intersections, and in situations where 0:10:11.720 --> 0:10:14.640 collision is possible, and is to stay on the road 0:10:14.960 --> 0:10:18.320 and in a legal and appropriate travel lane while en route, 0:10:18.800 --> 0:10:24.080 including around sharp turns, through intersections, and while passing. The 0:10:24.160 --> 0:10:28.480 route definition or Route network definition file will specify the 0:10:28.520 --> 0:10:31.800 GPS coordinates of the stop signs. The R and d 0:10:31.920 --> 0:10:34.840 F specifies the location of stop lines on the ground. 0:10:35.280 --> 0:10:38.640 On paved areas. Such stop lines will be represented by 0:10:38.679 --> 0:10:42.080 a painted stop line on the pavement. Physical stop signs, however, 0:10:42.160 --> 0:10:45.319 may or may not be present at the stop line locations. 0:10:46.000 --> 0:10:49.400 Navigate safely in areas where GPS signals are partially or 0:10:49.559 --> 0:10:53.839 entirely blocked. Uh there to follow paved and unpaved roads 0:10:53.880 --> 0:10:56.320 and stay in the lane with very sparse or low 0:10:56.360 --> 0:11:00.960 accuracy GPS way points very important because there for a 0:11:01.000 --> 0:11:06.600 long time GPS receivers were of questionable reliability, especially if 0:11:06.600 --> 0:11:10.040 you're going, like through a city with a lot of skyscrapers, 0:11:10.040 --> 0:11:13.600 for example, you might have signals blocked and then you 0:11:13.640 --> 0:11:18.040 would lose connectivity with GPS receivers. This still happens to 0:11:18.120 --> 0:11:20.560 this day, but not as frequently, at least not in 0:11:20.559 --> 0:11:24.400 my experience. It turns out that um whether there are 0:11:25.440 --> 0:11:28.680 receivers that are using other systems to supplement that data, 0:11:29.240 --> 0:11:32.480 or the actual receivers are just more sensitive, it seems 0:11:32.480 --> 0:11:36.079 to work more reliably these days. The vehicle was to 0:11:36.200 --> 0:11:39.199 change lanes safely when legal and appropriate, such as when 0:11:39.240 --> 0:11:42.040 passing a vehicle or entering an opposing traffic lane to 0:11:42.080 --> 0:11:46.040 pass a stopped vehicle. Vehicles must not pass other vehicles 0:11:46.200 --> 0:11:50.079 queued at an intersection. Very important. You can go around 0:11:50.080 --> 0:11:52.400 the car that's stopped in the middle of the road 0:11:52.440 --> 0:11:56.760 because of some possible mechanical failure. But you can't go 0:11:56.800 --> 0:11:59.040 around the car just because it's come to a stop 0:11:59.080 --> 0:12:01.400 at an intersection, because it might be waiting for its 0:12:01.400 --> 0:12:05.080 turn to go. Merge safely with traffic moving in one 0:12:05.240 --> 0:12:08.440 or more lanes. After stopping at an intersection, the vehicle 0:12:08.480 --> 0:12:10.400 would be able to pull across one lane of moving 0:12:10.440 --> 0:12:13.400 traffic to merge with moving traffic in the opposing lane. 0:12:14.000 --> 0:12:16.800 It was supposed to stop safely within one meter of 0:12:16.840 --> 0:12:19.600 the stop line at a stop signed intersection and proceed 0:12:19.679 --> 0:12:23.640 without excessive delay, so less than ten seconds delay according 0:12:23.679 --> 0:12:27.920 to intersection precedents rules. It was to exhibit proper que 0:12:28.000 --> 0:12:31.840 behavior at an intersection, including stopping at a safe distance 0:12:31.880 --> 0:12:34.760 from other vehicles and stop and go procession to the 0:12:34.800 --> 0:12:38.720 stop line without excessive delay. It was to navigate a 0:12:38.760 --> 0:12:42.680 destination toward a destination in a large open area where 0:12:42.720 --> 0:12:45.920 minimal or no GPS points are provided, as in loading 0:12:45.920 --> 0:12:49.440 dock areas or parking lots. These areas may contain fixed 0:12:49.480 --> 0:12:53.960 obstacles such as parked vehicles and moving obstacles, including other vehicles. 0:12:54.480 --> 0:12:57.120 It was to safely pull into and back out of 0:12:57.160 --> 0:13:00.640 a specified parking space in a parking lot. It was 0:13:00.720 --> 0:13:04.160 to safely execute one or more three point turning maneuvers 0:13:04.160 --> 0:13:07.760 to affect a U turn. And it was to dynamically 0:13:07.840 --> 0:13:11.160 replan and execute the route to a destination if the 0:13:11.240 --> 0:13:14.839 primary route is blocked or impassable. These are all things 0:13:14.880 --> 0:13:17.840 that human drivers can do, and these are all things 0:13:17.880 --> 0:13:20.680 that human drivers do on a regular basis. But as 0:13:20.679 --> 0:13:23.120 you sit there and think about what it's needed to 0:13:23.880 --> 0:13:26.560 program a machine to be able to do these things, 0:13:26.640 --> 0:13:30.160 you start to recognize the challenges here because you have 0:13:30.200 --> 0:13:34.120 to be able to detect these scenarios and then you 0:13:34.120 --> 0:13:37.480 have to be able to react appropriately based upon the 0:13:37.520 --> 0:13:41.360 information that's available to you. So it's not just you know, 0:13:41.520 --> 0:13:44.720 if this, then that there are a lot of other considerations, 0:13:45.559 --> 0:13:48.840 and it gives you a hint at the challenge that 0:13:48.880 --> 0:13:51.520 was ahead of these teams. However, DARPA did point out 0:13:51.559 --> 0:13:54.439 a few things that were outside the scope of this 0:13:54.520 --> 0:13:57.520 challenge that teams would not need to worry about for 0:13:57.559 --> 0:14:01.640 the purposes of this competition that included the recognition of 0:14:01.679 --> 0:14:05.040 external traffic signals like traffic lights and stop signs through 0:14:05.080 --> 0:14:08.160 the use of sensors. You didn't have to worry about that. 0:14:08.360 --> 0:14:13.360 So because they were providing information about where stop signs 0:14:13.400 --> 0:14:17.319 were based on GPS coordinates, you could program that directly 0:14:17.320 --> 0:14:20.760 into your system, so that your car quote unquote knows 0:14:21.440 --> 0:14:24.360 to stop at a certain point because it quote unquote 0:14:24.440 --> 0:14:27.479 knows there's a stop sign, there doesn't have to detect it. 0:14:27.480 --> 0:14:30.520 It's been programmed with that info, so there was no 0:14:30.560 --> 0:14:33.080 need to develop any kind of optical system that would 0:14:33.120 --> 0:14:36.760 detect a stop sign or a traffic light. This works 0:14:36.840 --> 0:14:41.240 great if you're operating in a very limited area, right 0:14:41.280 --> 0:14:44.920 in this case, in this simulated urban environment, if you're 0:14:45.120 --> 0:14:49.200 operating in something where you've got to find border, and 0:14:49.280 --> 0:14:53.560 you know quote unquote everything about that's everything that's inside 0:14:53.560 --> 0:14:56.640 that border, you can program this kind of stuff into 0:14:56.640 --> 0:15:02.240 your vehicle directly. But this approach it's more and more 0:15:02.280 --> 0:15:06.480 impractical the larger your area of services. So while you 0:15:06.600 --> 0:15:08.880 might be able to program this in for an area 0:15:08.920 --> 0:15:13.360 that might consist of a few simulated blocks of a city, 0:15:14.000 --> 0:15:15.960 it doesn't really work so well for a full city. 0:15:16.440 --> 0:15:18.240 You would have to go in and program the GPS 0:15:18.320 --> 0:15:22.000 coordinates for every single stop sign and every single traffic signal. 0:15:22.280 --> 0:15:24.920 Not only that, but for traffic signals, you would also 0:15:24.960 --> 0:15:27.040 have to figure out a way for your vehicle to 0:15:27.280 --> 0:15:30.840 understand when the signal had changed. But for the purposes 0:15:30.840 --> 0:15:34.080 of the competition, they said this isn't necessary. They also 0:15:34.120 --> 0:15:37.600 said it's not necessary to be able to recognize pedestrians 0:15:37.920 --> 0:15:41.760 or build in pedestrian avoidance in your vehicle because there 0:15:41.800 --> 0:15:45.680 were not going to be any pedestrians on the simulated course. Clearly, 0:15:45.720 --> 0:15:48.320 that would also be something that would be necessary if 0:15:48.360 --> 0:15:51.800 we were to use driver lest cars in everyday life, 0:15:51.800 --> 0:15:55.600 but for the purposes of this challenge not not pertinent. 0:15:56.600 --> 0:16:00.520 Also outside the scope or behaviors necessary for high a driving, 0:16:00.560 --> 0:16:03.040 such as high speed passing or high speed merge at 0:16:03.080 --> 0:16:05.800 an on ramp. Speed limits for the urban challenge will 0:16:05.840 --> 0:16:08.960 be thirty miles per hour or less, because again, it 0:16:09.080 --> 0:16:12.440 was supposed to be going through surface streets in a 0:16:12.520 --> 0:16:16.280 city environment, not on a highway. Driving in difficult off 0:16:16.400 --> 0:16:19.000 road terrain is outside the scope of the program. Off 0:16:19.120 --> 0:16:22.880 Road navigation in an unpaved area, travel along roads with potholes, 0:16:23.160 --> 0:16:27.000 and travel along a dirt road are within scope, so 0:16:28.000 --> 0:16:32.280 no off road travel would be necessary, But they didn't 0:16:32.280 --> 0:16:35.040 guarantee that all the roads would be in perfect condition 0:16:35.680 --> 0:16:39.680 and that a vehicle might have some difficult terrain that 0:16:40.040 --> 0:16:44.080 that a regular driver might encounter if uh, the area 0:16:44.160 --> 0:16:47.000 they are in has different types of roads like gravel 0:16:47.080 --> 0:16:53.360 roads and that kind of thing. No, uh, obviously we 0:16:53.400 --> 0:16:56.960 would need more than what DARPA was rolling out to have, 0:16:57.200 --> 0:16:59.800 like a driverless car that that a person could safely 0:17:00.000 --> 0:17:03.440 eide in. But you have to crawl before you can walk. 0:17:04.680 --> 0:17:08.680 We'll talk a little bit more about the Urban Challenge, 0:17:08.680 --> 0:17:11.760 but first let's take a quick break to thank our sponsor. 0:17:19.280 --> 0:17:21.600 All Right, so I've talked about the objective of the 0:17:21.720 --> 0:17:25.040 Urban Challenge, but but what about the actual procedure. Well, 0:17:25.119 --> 0:17:27.960 every team would receive a detailed layout of the test 0:17:28.000 --> 0:17:31.960 area that would be that digital file I mentioned that 0:17:31.960 --> 0:17:34.480 would have all the accessible roads, all the parts of 0:17:34.520 --> 0:17:37.520 the course that the vehicles could drive through without penalty. 0:17:38.240 --> 0:17:41.399 That would help teams define exactly where their vehicles could 0:17:41.480 --> 0:17:43.480 and could not go. So this is sort of like 0:17:43.520 --> 0:17:46.920 getting a really really good map of a specific part 0:17:46.960 --> 0:17:50.159 of a city that you're working with there's no information 0:17:50.320 --> 0:17:53.920 about the route on that map. It's just a map 0:17:53.960 --> 0:17:58.399 of the area. Immediately before the test, vehicles would be 0:17:58.440 --> 0:18:01.600 given a series of check points that they were told 0:18:01.640 --> 0:18:06.439 they had to visit. The pathway between those checkpoints was undefined, however, 0:18:06.800 --> 0:18:10.560 so the vehicles could actually plot their own course from 0:18:10.640 --> 0:18:13.119 start to finish, which is kind of like getting in 0:18:13.160 --> 0:18:15.520 a friends car and you say, hey, let's go to 0:18:15.600 --> 0:18:18.560 that that pizza place that we like to go to, 0:18:18.960 --> 0:18:21.200 and you find out that they take a different route 0:18:21.240 --> 0:18:25.160 than you would to get to that specific destination. So 0:18:25.359 --> 0:18:27.680 each vehicle would be able to plot its own route 0:18:27.720 --> 0:18:30.679 that that made the most quote unquote sense to the 0:18:30.760 --> 0:18:33.800 respective vehicle, and they'd have to do this a few times. 0:18:33.800 --> 0:18:38.000 There'd be a few different checkpoints, but the actual uh 0:18:38.240 --> 0:18:40.840 direction that the vehicle would travel would be up to 0:18:40.880 --> 0:18:44.359 the vehicle. In addition, there could be road blockages on 0:18:44.400 --> 0:18:47.480 the course and they would not be indicated by the 0:18:47.680 --> 0:18:50.399 r n d F information. They wouldn't be on that map, 0:18:50.920 --> 0:18:52.600 so you wouldn't get a map that said, oh, by 0:18:52.640 --> 0:18:55.639 the way, this little road here is actually going to 0:18:55.720 --> 0:18:58.800 be blocked so you can't go down it. And that 0:18:58.920 --> 0:19:01.199 was important. They want to have the blockages there, but 0:19:01.240 --> 0:19:04.199 they didn't want the teams to have advanced knowledge of 0:19:04.240 --> 0:19:07.679 which routes were blocked. Just like a static map, you 0:19:07.680 --> 0:19:10.720 wouldn't know if a road was temporarily closed because of 0:19:10.720 --> 0:19:13.639 a wreck or flooding or whatever it might be. So 0:19:13.680 --> 0:19:16.760 when you encounter that sort of obstacle, if you're a 0:19:16.840 --> 0:19:18.920 human driver and let's say you've just got a paper map, 0:19:18.960 --> 0:19:23.000 you don't have a GPS receiver, you're forced to reevaluate 0:19:23.040 --> 0:19:25.399 your plan. You're forced to look at the map and 0:19:25.520 --> 0:19:28.320 consult it and figure out a different way to get 0:19:28.359 --> 0:19:30.320 to where you're going, and you have to change things 0:19:30.320 --> 0:19:32.760 on the fly. Well, that would happen in the Urban 0:19:32.840 --> 0:19:37.199 Challenge as well to these autonomous cars. In addition to 0:19:37.240 --> 0:19:40.840 the Route Network Data File, DARPA would provide each team 0:19:40.880 --> 0:19:44.520 with the Mission Data File MDF. This is the one 0:19:44.560 --> 0:19:47.520 that would actually have the information about the checkpoints that 0:19:47.600 --> 0:19:50.280 the vehicle was meant to visit, as well as information 0:19:50.280 --> 0:19:54.679 about the minimum and maximum speed limits for each road segment, saying, 0:19:55.400 --> 0:19:58.119 along this section of this road, you can go up 0:19:58.119 --> 0:20:00.560 to thirty miles per hour. For example, on the a section, 0:20:00.840 --> 0:20:04.200 twenty miles hours your maximum. That kind of thing and 0:20:04.680 --> 0:20:07.240 as the rules had stated, DARPA expected all vehicles to 0:20:07.280 --> 0:20:10.240 operate within those speed parameters for each road segment. Actually, 0:20:10.560 --> 0:20:15.159 to be more specific, each car was to behave according 0:20:15.320 --> 0:20:23.080 to the driving laws of California. Vehicles could be teleoperated, 0:20:23.520 --> 0:20:27.080 that is, they could be controlled remotely, but only for 0:20:27.160 --> 0:20:30.000 the purposes of staging the vehicle at the start of 0:20:30.040 --> 0:20:34.200 the competition, so a team could take over manual control 0:20:34.320 --> 0:20:36.960 of the vehicle in order to move it to its 0:20:37.040 --> 0:20:41.880 starting point, but then they would have to disengage from control, 0:20:42.160 --> 0:20:44.600 and at that point, each vehicle had to be under 0:20:44.680 --> 0:20:49.119 complete autonomous control with no input from the teams until 0:20:49.160 --> 0:20:52.480 the completion of the course or if the vehicle experienced 0:20:52.560 --> 0:20:55.720 some sort of failure that would remove it from consideration. 0:20:56.280 --> 0:20:58.119 All vehicles had to be built on top of a 0:20:58.200 --> 0:21:01.879 chassis that had a documented safety record, so you couldn't 0:21:01.960 --> 0:21:05.320 just build an entirely new vehicle on a custom chassis 0:21:05.600 --> 0:21:09.119 and be eligible for the competition. Most teams would retrofit 0:21:09.240 --> 0:21:13.240 existing vehicles for their entries. In addition, all vehicles had 0:21:13.320 --> 0:21:17.000 specific parameters they had to fall inside, including weight requirements. 0:21:17.160 --> 0:21:19.600 Um they had to be at least two thousand pounds 0:21:20.240 --> 0:21:23.680 that's about nine seven rams. They could be no heavier 0:21:23.720 --> 0:21:28.760 than a whopping thirty thousand pounds or thirteen thousand, six 0:21:28.880 --> 0:21:33.879 hundred krams. Now, for reference, the heaviest consumer car on 0:21:33.920 --> 0:21:37.040 the market that I could find is the Mercedes may 0:21:37.080 --> 0:21:41.560 BACHS six hundred Pullman Guard at eleven thousand, two hundred 0:21:41.680 --> 0:21:46.520 forty four pounds or five thousand one ms. Still very 0:21:46.600 --> 0:21:49.200 shy of that thirty thousand mark. If you want to 0:21:49.240 --> 0:21:52.879 go with pickup trucks, that's the heaviest class of commercial, 0:21:53.000 --> 0:21:55.639 well consumer vehicle. I should say not commercial. I mean 0:21:55.680 --> 0:21:58.880 you've got massive dump trucks and stuff. They're super heavy. 0:21:59.040 --> 0:22:00.359 But if you wanted to go out and by the 0:22:00.359 --> 0:22:03.320 heaviest pickup truck out there, you would look for the 0:22:03.400 --> 0:22:07.359 International x T, which weighs in at fourteen thousand, fifty 0:22:07.400 --> 0:22:12.280 one pounds or six thousand six d. You're still less 0:22:12.320 --> 0:22:16.280 than half of what the upper weight limit is at 0:22:16.320 --> 0:22:20.320 that point. All other heavy consumer vehicles like SUVs and 0:22:20.400 --> 0:22:24.080 vans are actually in between those two extremes. So it 0:22:24.119 --> 0:22:26.720 would have to be a really hefty vehicle to max 0:22:26.840 --> 0:22:30.600 out the weight limit of DARPA. But this reminds us 0:22:30.640 --> 0:22:33.840 again the ultimate golfer DARPA is to develop technology to 0:22:33.880 --> 0:22:38.639 devote toward automating military combat ground vehicles, which tend to 0:22:38.640 --> 0:22:42.000 be a bit on the hefty side. In addition, the 0:22:42.119 --> 0:22:45.800 vehicles had to have a minimum wheelbase of seventy two inches, 0:22:45.800 --> 0:22:48.639 which is about one point eight meters from the front 0:22:48.640 --> 0:22:51.879 axle to the back axle, and it could have a 0:22:51.920 --> 0:22:55.560 maximum width of nine feet or two point seven four 0:22:55.680 --> 0:23:00.520 meters and a maximum height of twelve feet or about 0:23:00.560 --> 0:23:04.080 three point seven meters. Vehicles had to be able to 0:23:04.119 --> 0:23:08.240 move autonomously, both forward and in reverse. They had to 0:23:08.280 --> 0:23:09.920 be able to make that you turn. They had to 0:23:09.920 --> 0:23:12.480 be able to turn on a typical urban street which 0:23:12.480 --> 0:23:15.840 would be about thirty feet wide or nine meters, and 0:23:15.920 --> 0:23:19.360 not climb up on the curb on either side. They 0:23:19.359 --> 0:23:23.520 had to be vehicles that traveled on tires. There it 0:23:23.560 --> 0:23:26.200 could be no treads, no tracks, nothing like that. Had 0:23:26.240 --> 0:23:29.960 to be on tires. Uh. They were not to damage 0:23:30.000 --> 0:23:32.480 the surface of a street with their passage, so it 0:23:32.520 --> 0:23:36.040 couldn't be a means of getting around that could actually 0:23:36.040 --> 0:23:39.119 tear up the road. And all the vehicles, sensors, and 0:23:39.160 --> 0:23:42.560 technologies had to be self contained. The teams would not 0:23:42.600 --> 0:23:45.280 be allowed to set up any sort of additional sensors 0:23:45.359 --> 0:23:49.200 in the area to aid in the vehicle navigation or operation. 0:23:49.800 --> 0:23:53.640 All data processing would similarly have to take place inside 0:23:53.680 --> 0:23:56.000 the vehicle, So it was against the rules for the 0:23:56.080 --> 0:24:00.119 sensors to send data to some external computer for processing 0:24:00.440 --> 0:24:03.760 and then have that external computer beam back instructions. And 0:24:03.800 --> 0:24:06.640 when you consider the fact that vehicles need to make 0:24:07.160 --> 0:24:12.720 split second decisions to avoid a potential accident, you probably 0:24:12.760 --> 0:24:15.320 come to the same conclusion that this is the most 0:24:15.400 --> 0:24:18.639 logical approach. You don't want to insert delay if you 0:24:18.640 --> 0:24:21.960 can avoid it. It was also against the rules to 0:24:21.960 --> 0:24:24.720