WEBVTT 00:12.080 --> 00:16.560 What's up everybody? Koiwire here with a 00:14.160 --> 00:18.160 special edition of CNN 10. Today we're 00:16.560 --> 00:20.560 taking a deep dive into the world of 00:18.160 --> 00:23.199 artificial intelligence and self-driving 00:20.560 --> 00:25.119 cars. Today we're going to Curiosity Lab 00:23.199 --> 00:27.359 in Peach Tree Corners, Georgia to 00:25.119 --> 00:31.439 experience this tech firsthand. This is 00:27.359 --> 00:32.719 my first time in a Whimo. Let's go. 00:31.439 --> 00:35.280 You could call this one of the most 00:32.719 --> 00:37.760 advanced smart cities in the world. 00:35.280 --> 00:40.399 Tomorrow's technologies being tested and 00:37.760 --> 00:42.399 created today. The color printer was 00:40.399 --> 00:45.600 invented here in Technology Park. The 00:42.399 --> 00:48.079 Haze modem, too. Now, autonomous or 00:45.600 --> 00:51.200 self-driving vehicles are being steered 00:48.079 --> 00:54.079 to new frontiers. Air package delivery 00:51.200 --> 00:57.440 systems via drones. Mobile delivery 00:54.079 --> 00:59.039 robots tested and perfected. Where are 00:57.440 --> 01:00.160 we? What is this place? I feel like I 00:59.039 --> 01:02.239 stepped into the future. 01:00.160 --> 01:04.080 >> This is Curiosity Lab. We are a 01:02.239 --> 01:06.240 nonprofit just inside the city of Peach 01:04.080 --> 01:08.080 Tree Corners. And what we do is we help 01:06.240 --> 01:10.640 companies test and deploy their 01:08.080 --> 01:12.799 technology on public infrastructure. So 01:10.640 --> 01:14.400 that meant always sounds super cool, but 01:12.799 --> 01:16.000 it is. I mean, think about all the stuff 01:14.400 --> 01:18.240 that public infrastructure entails, 01:16.000 --> 01:19.920 right? Roadways, intersections, even 01:18.240 --> 01:21.119 airspace. Um, a lot of different 01:19.920 --> 01:22.320 technologies work on those things. 01:21.119 --> 01:23.840 >> And that's very important because we're 01:22.320 --> 01:25.680 living at a time where we're seeing more 01:23.840 --> 01:28.880 and more of it, hearing more and more 01:25.680 --> 01:31.280 about it. They might be taxi cabs that 01:28.880 --> 01:33.119 are flying through this city. So how 01:31.280 --> 01:34.560 well all do you help companies prepare 01:33.119 --> 01:37.119 us for that type of future? 01:34.560 --> 01:38.799 >> Sure. So when it comes to an environment 01:37.119 --> 01:40.799 like what is the ideal environment for 01:38.799 --> 01:42.079 testing? So you always start in a closed 01:40.799 --> 01:43.680 environment, right? There's there's no 01:42.079 --> 01:45.200 things coming in in terms of variables. 01:43.680 --> 01:46.960 You want to make sure the product works. 01:45.200 --> 01:48.240 So in the case of a vehicle on a 01:46.960 --> 01:50.720 roadway, that would be like a closed 01:48.240 --> 01:51.920 track. But we also know that from there, 01:50.720 --> 01:53.759 how do you get into an urban 01:51.920 --> 01:55.200 environment? How do you get into a real 01:53.759 --> 01:57.600 city? And you've got to have that middle 01:55.200 --> 02:00.159 ground. We have a closed three mile loop 01:57.600 --> 02:02.799 inside a 500 acre ecosystem so that 02:00.159 --> 02:04.560 cars, technology, drones, whatever it 02:02.799 --> 02:06.479 is, can test in a real environment where 02:04.560 --> 02:08.720 there's real pedestrians, real cars, 02:06.479 --> 02:10.319 real people, real intersections. So you 02:08.720 --> 02:12.160 can start to get the data and learn how 02:10.319 --> 02:13.840 these technologies work in real time 02:12.160 --> 02:15.840 with real elements so that you know, 02:13.840 --> 02:18.959 hey, we're ready to get into the city. 02:15.840 --> 02:19.760 >> I just saw a robot fly by. I don't know 02:18.959 --> 02:20.959 what it was, but 02:19.760 --> 02:22.879 >> oh my gosh, I think it might have been 02:20.959 --> 02:24.319 Cheetah. She's a cute little thing. 02:22.879 --> 02:26.319 Basically, she's a helping hand 02:24.319 --> 02:27.840 essentially. She is a a robot that 02:26.319 --> 02:29.520 follows a subject in front of it. So, 02:27.840 --> 02:31.200 it's not a self-driving one. It's not 02:29.520 --> 02:32.640 like it's doing delivery, but it's more 02:31.200 --> 02:33.360 like, you know, a robotic dog that 02:32.640 --> 02:34.400 follows you everywhere. 02:33.360 --> 02:34.879 >> You don't need to carry a backpack 02:34.400 --> 02:36.319 anymore. 02:34.879 --> 02:39.040 >> It's exactly what it is. It's a rolling 02:36.319 --> 02:40.400 robotic backpack. You open up the uh vat 02:39.040 --> 02:41.599 inside of it, and you know, they have 02:40.400 --> 02:43.920 different inserts. I mean, you can make 02:41.599 --> 02:44.959 it follow you like a tailgate. Um, but, 02:43.920 --> 02:47.599 you know, it also increases 02:44.959 --> 02:49.200 accessibility for things like disabled 02:47.599 --> 02:50.720 folk or even elderly and, you know, 02:49.200 --> 02:52.239 walkable communities if you're at the 02:50.720 --> 02:54.800 airport and you need help carrying 02:52.239 --> 02:57.200 stuff. Um, so it's cool. The fun fact 02:54.800 --> 02:59.440 there is that its design was based off 02:57.200 --> 03:01.440 of the droids in Star Wars. So they're 02:59.440 --> 03:04.000 they're quite cute, I think. 03:01.440 --> 03:06.239 >> Now you're speaking my language. Where's 03:04.000 --> 03:08.239 my lightsaber? I feel cars whizzing 03:06.239 --> 03:09.680 behind me. So tell me what's happening 03:08.239 --> 03:11.120 here. What are we seeing? What are we 03:09.680 --> 03:13.360 learning from what we're seeing? 03:11.120 --> 03:16.640 >> Yeah, absolutely. So we talk about, you 03:13.360 --> 03:18.239 know, the roadway. Um that is a very 03:16.640 --> 03:19.680 loaded thing to think about. There's so 03:18.239 --> 03:21.680 many things going on at all different 03:19.680 --> 03:23.360 times on the roadway. And as drivers, 03:21.680 --> 03:24.800 you know, we learn to get used to these 03:23.360 --> 03:26.720 things. You learn when someone's going 03:24.800 --> 03:28.159 to cut in front of you. You learn how to 03:26.720 --> 03:29.519 interact with different lights and 03:28.159 --> 03:30.799 different signals and different signs. 03:29.519 --> 03:32.720 You know, what we're doing here is we're 03:30.799 --> 03:34.560 using uh cameras and we're putting 03:32.720 --> 03:36.319 analytics over those cameras to show you 03:34.560 --> 03:38.480 what's happening in here. So, um this 03:36.319 --> 03:39.760 can be a forum of AI, right? So, it 03:38.480 --> 03:41.120 could be doing things like object 03:39.760 --> 03:42.959 detection. As you can see here, it's 03:41.120 --> 03:45.360 looking at vehicle vehicle. It could 03:42.959 --> 03:46.799 clock a, you know, a bicyclist or a 03:45.360 --> 03:49.360 pedestrian, which we're seeing a couple 03:46.799 --> 03:51.120 pedestrians come right there. So, it 03:49.360 --> 03:52.640 says unknown and that's a part of 03:51.120 --> 03:54.239 testing, right? Is that, you know, it's 03:52.640 --> 03:56.080 still learning. It says person now as 03:54.239 --> 03:57.439 it's getting closer to the camera. 03:56.080 --> 03:58.640 >> Um, but that's part of it, right? Is 03:57.439 --> 04:00.159 there are times where a truck comes 03:58.640 --> 04:01.760 through and it and you know, it might 04:00.159 --> 04:03.280 say person and you're like, well, that's 04:01.760 --> 04:04.720 definitely not a person. But that's why 04:03.280 --> 04:06.879 you need the testing environment so they 04:04.720 --> 04:08.959 can train their AI models to do better. 04:06.879 --> 04:10.959 >> Okay. So, we've spoken about the cities, 04:08.959 --> 04:12.560 the roadways, the intersections of 04:10.959 --> 04:14.640 tomorrow. 04:12.560 --> 04:16.400 Also, there will be the cars of tomorrow 04:14.640 --> 04:18.239 that work better with that. And I 04:16.400 --> 04:19.840 understand there's a garage where we can 04:18.239 --> 04:20.639 maybe check out one of these cars. Put 04:19.840 --> 04:22.479 me to work. 04:20.639 --> 04:24.400 >> All right, let's do it. Let's go. 04:22.479 --> 04:26.479 >> The company testing out their fleet of 04:24.400 --> 04:29.199 autonomous vehicles at Curiosity Lab 04:26.479 --> 04:31.440 during our visit is called May Mobility. 04:29.199 --> 04:33.120 These cars use similar technology to the 04:31.440 --> 04:35.919 brands we may be familiar with like 04:33.120 --> 04:38.560 Whimo. But this company is particularly 04:35.919 --> 04:40.400 interested in making AVs that are more 04:38.560 --> 04:42.400 accessible for passengers who use 04:40.400 --> 04:43.440 wheelchairs, scooters, or need other 04:42.400 --> 04:45.520 accommodations. 04:43.440 --> 04:47.199 >> In order for an autonomous vehicle to 04:45.520 --> 04:49.040 work, you have to have a bunch of 04:47.199 --> 04:51.600 different technology, right? So, you've 04:49.040 --> 04:52.800 got to have the hardware, and I'll walk 04:51.600 --> 04:54.160 through that in a minute. And you have 04:52.800 --> 04:56.240 to have the software, which is pretty 04:54.160 --> 04:58.800 much the brains of the technology. So 04:56.240 --> 05:01.199 the AI component, you're using 04:58.800 --> 05:03.199 human-like reasoning, which is basically 05:01.199 --> 05:04.720 the software component, and combining it 05:03.199 --> 05:07.440 with all the things you're gathering 05:04.720 --> 05:09.360 from the hardware so that the technology 05:07.440 --> 05:11.840 can make thousands and thousands and 05:09.360 --> 05:12.639 thousands of scenarios and decisions in 05:11.840 --> 05:13.039 real time. 05:12.639 --> 05:15.039 >> Mhm. 05:13.039 --> 05:16.880 >> So you'll see here different 05:15.039 --> 05:20.160 technologies. A lot of people say this 05:16.880 --> 05:22.160 looks like a cup holder. It is not. It's 05:20.160 --> 05:24.880 actually technology. 05:22.160 --> 05:27.360 And you'll see cameras here and LAR 05:24.880 --> 05:29.600 sensors. We have a total of nine cameras 05:27.360 --> 05:32.320 all around the vehicle. And then we have 05:29.600 --> 05:35.440 five LARs and radars. The other big guy 05:32.320 --> 05:38.720 here, the top hat, is basically um our 05:35.440 --> 05:41.039 largest LAR. And that is able to look 05:38.720 --> 05:43.919 out further away so that we can see 05:41.039 --> 05:46.160 what's happening in real time. 05:43.919 --> 05:48.639 >> While autonomous vehicles are relatively 05:46.160 --> 05:50.720 new, lidar has actually been around for 05:48.639 --> 05:53.120 quite some time. It was developed in the 05:50.720 --> 05:55.759 1960s and it was actually used on the 05:53.120 --> 05:59.039 Apollo space missions to map out the 05:55.759 --> 06:00.960 surface of the moon. It uses lasers, 05:59.039 --> 06:02.800 shoots them out everywhere to measure 06:00.960 --> 06:05.120 distances and create an incredibly 06:02.800 --> 06:07.280 accurate picture of an environment. 06:05.120 --> 06:10.080 Here's how it works. The system fires 06:07.280 --> 06:12.319 millions of laser pulses 06:10.080 --> 06:14.240 every second in every direction. They 06:12.319 --> 06:16.720 bounce off surrounding objects to create 06:14.240 --> 06:18.720 a detailed 3D picture of its 06:16.720 --> 06:20.720 surroundings. That includes everything 06:18.720 --> 06:23.199 from buildings to other cars, people, 06:20.720 --> 06:25.440 and animals. It's similar in a way to 06:23.199 --> 06:27.759 echolocation, the system used by bats, 06:25.440 --> 06:30.319 whales, and dolphins to navigate and to 06:27.759 --> 06:32.560 hunt prey. But this system swaps the 06:30.319 --> 06:34.960 sound waves for light. You may have seen 06:32.560 --> 06:36.800 the rapidly spinning mechanisms on some 06:34.960 --> 06:40.000 of these autonomous cars. Well, that 06:36.800 --> 06:42.639 allows them to have a 360° field of view 06:40.000 --> 06:45.759 to help eliminate blind spots and safely 06:42.639 --> 06:48.000 navigate the world around them. 06:45.759 --> 06:50.160 Pop quiz hot shot. Self-driving cars 06:48.000 --> 06:51.919 often use lidar to scan surroundings. 06:50.160 --> 06:53.600 What does lidar stand for? Light 06:51.919 --> 06:55.440 detection and ranging. Laser 06:53.600 --> 06:57.280 identification and radar. Light 06:55.440 --> 07:00.880 identification and response. Looking 06:57.280 --> 07:03.520 into dangers around the route. Answer is 07:00.880 --> 07:05.280 light detection and ranging. Lidar is a 07:03.520 --> 07:08.000 remote sensing method used to examine 07:05.280 --> 07:10.160 the surface of the earth. So what you 07:08.000 --> 07:12.880 want to do is you want to use a 07:10.160 --> 07:15.680 combination of historical data, right, 07:12.880 --> 07:18.000 which it's accumulated over time based 07:15.680 --> 07:20.960 on different locations. So for example, 07:18.000 --> 07:24.080 let's say you're in Georgia, hot 07:20.960 --> 07:27.280 weather, right? You're in Phoenix, 07:24.080 --> 07:30.800 Arizona, hot but dry weather, and you're 07:27.280 --> 07:32.400 in Michigan where you have snow. So, you 07:30.800 --> 07:34.720 want to be able to take all of those 07:32.400 --> 07:36.880 elements and have the vehicle learn from 07:34.720 --> 07:38.720 all the different weather conditions so 07:36.880 --> 07:41.120 that when it's a location, if we happen 07:38.720 --> 07:43.199 to get snow in Georgia this year, the 07:41.120 --> 07:45.680 vehicle will know what to do. 07:43.199 --> 07:48.000 >> But I think about, you know, my 07:45.680 --> 07:50.080 daughters when they're old enough to be 07:48.000 --> 07:51.759 going places, 07:50.080 --> 07:54.080 I think that I would feel comfortable 07:51.759 --> 07:56.319 with them being in a vehicle all to 07:54.080 --> 07:58.639 themselves. The tech is only going to 07:56.319 --> 08:00.720 get better, right? And you would think 07:58.639 --> 08:03.520 as there are more and more autonomous 08:00.720 --> 08:05.440 vehicles on the roadways that are now 08:03.520 --> 08:07.919 communicating with each other, they're 08:05.440 --> 08:10.400 like computer brains communicating with 08:07.919 --> 08:13.199 other drivers. And so you would think 08:10.400 --> 08:15.199 that would be even safer than kind of 08:13.199 --> 08:16.400 guessing and wondering what a human 08:15.199 --> 08:19.280 might do. 08:16.400 --> 08:21.440 >> There's a lot of um transparency and 08:19.280 --> 08:25.759 visibility into a vehicle. you have 08:21.440 --> 08:28.240 cameras everywhere and um so you you can 08:25.759 --> 08:30.000 see what's happening and so for them in 08:28.240 --> 08:32.399 a lot of ways they're probably a lot 08:30.000 --> 08:35.200 safer you know in the vehicle we all 08:32.399 --> 08:38.240 know that you know last year I think 08:35.200 --> 08:41.039 there were over 40,000 deaths in the US 08:38.240 --> 08:43.279 from um accidents road accidents it's 08:41.039 --> 08:46.880 things like you know falling asleep 08:43.279 --> 08:49.360 texting and driving intoxication so all 08:46.880 --> 08:50.640 of those things can happen you know with 08:49.360 --> 08:53.200 with your 08:50.640 --> 08:55.760 right for the future. So, and I would 08:53.200 --> 08:57.600 argue that for the future and maybe even 08:55.760 --> 08:59.440 today that these vehicles are for, you 08:57.600 --> 09:01.519 know, safe and hopefully that's their 08:59.440 --> 09:03.760 future. And then you've got to think 09:01.519 --> 09:06.320 too, there's others as well. I mean, you 09:03.760 --> 09:09.120 have kids, right? But there's um people 09:06.320 --> 09:11.200 with disabilities, great for them. And 09:09.120 --> 09:12.959 then, you know, people who are elderly, 09:11.200 --> 09:16.800 right? They can come in and use those 09:12.959 --> 09:19.680 vehicles. In um Minnesota, we have um 09:16.800 --> 09:22.560 one of our biggest groups of riders are 09:19.680 --> 09:23.040 seniors and the middle school kids 09:22.560 --> 09:24.320 because they 09:23.040 --> 09:26.480 >> not seniors in high school. 09:24.320 --> 09:28.240 >> No, senior citizens. Yes, that's senior 09:26.480 --> 09:30.640 middle schoolers and middle schoolers 09:28.240 --> 09:33.519 because you know a lot of times they 09:30.640 --> 09:36.080 don't have a way to get home um you know 09:33.519 --> 09:38.720 after after school activities and so 09:36.080 --> 09:40.240 this is a way for them to do it safely. 09:38.720 --> 09:41.760 >> What do you think? I want you to press 09:40.240 --> 09:44.000 pause for a moment. Discuss with your 09:41.760 --> 09:45.920 friends if a self-driving car service 09:44.000 --> 09:47.839 was widely available in your city or 09:45.920 --> 09:49.680 town. Maybe they already are. Would you 09:47.839 --> 09:51.279 use them? Would you trust AI over 09:49.680 --> 09:53.440 getting your own driver's license? What 09:51.279 --> 09:54.800 are the risks? What are the benefits? 09:53.440 --> 09:56.959 Thanks for taking a ride with me today 09:54.800 --> 09:58.640 on this special edition of CNN 10. I 09:56.959 --> 10:03.959 learned a lot today. I hope you did, 09:58.640 --> 10:03.959 too. Be kind, stay curious, and rise up.