1 00:00:00,520 --> 00:00:03,840 Speaker 1: Welcome to Brainstuff from how stuff Works dot com where 2 00:00:03,840 --> 00:00:15,120 Speaker 1: smart Happens. Hi. I'm Marshall Brain with today's question, how 3 00:00:15,160 --> 00:00:19,479 Speaker 1: do self driving cars work? Who wouldn't like to have 4 00:00:19,520 --> 00:00:22,840 Speaker 1: a car that drives itself. The idea of being able 5 00:00:22,880 --> 00:00:26,240 Speaker 1: to sleep or to use your laptop during the commute 6 00:00:26,239 --> 00:00:29,600 Speaker 1: to work is certainly appealing. The idea of having your 7 00:00:29,640 --> 00:00:32,000 Speaker 1: car drop you off at the door of a store 8 00:00:32,560 --> 00:00:35,839 Speaker 1: and then go park itself is also attractive. And then 9 00:00:35,880 --> 00:00:39,520 Speaker 1: there's the safety thing. Human beings just are not that 10 00:00:39,600 --> 00:00:42,600 Speaker 1: good when it comes to driving, with millions of automobile 11 00:00:42,640 --> 00:00:46,760 Speaker 1: accidents causing damage, injury, and death every year. If we 12 00:00:46,840 --> 00:00:50,320 Speaker 1: all had self driving cars, it would be great. Two 13 00:00:50,360 --> 00:00:54,200 Speaker 1: recent announcements have made self driving cars seem a lot 14 00:00:54,240 --> 00:00:57,720 Speaker 1: closer to reality. The first comes from the Army, which 15 00:00:57,760 --> 00:01:01,200 Speaker 1: is now using self driving vehicle to guard a large 16 00:01:01,240 --> 00:01:04,920 Speaker 1: military facility and nuclear waste dump in Nevada. And then 17 00:01:05,000 --> 00:01:08,959 Speaker 1: there's Google, which recently announced that it's developed self driving 18 00:01:09,000 --> 00:01:12,319 Speaker 1: cars that have logged over a hundred thousand miles on 19 00:01:12,440 --> 00:01:16,759 Speaker 1: real roads in real traffic. What's going on inside these 20 00:01:16,800 --> 00:01:19,640 Speaker 1: cars to make it possible for them to drive themselves. 21 00:01:20,040 --> 00:01:22,240 Speaker 1: We can take a look at it. By looking at 22 00:01:22,400 --> 00:01:26,200 Speaker 1: Google cars or some of the cars that have participated 23 00:01:26,280 --> 00:01:31,280 Speaker 1: in DARPST Grand Challenge, you could, in theory, turn any 24 00:01:31,319 --> 00:01:34,320 Speaker 1: car into a self driving car. The first thing you 25 00:01:34,319 --> 00:01:36,880 Speaker 1: would have to do is make it possible for a 26 00:01:36,920 --> 00:01:40,280 Speaker 1: computer to control the car. This is not quite as 27 00:01:40,319 --> 00:01:43,920 Speaker 1: easy as it sounds, but it's relatively straightforward. The computer 28 00:01:44,000 --> 00:01:46,720 Speaker 1: needs to be able to turn the steering wheel, push 29 00:01:46,760 --> 00:01:50,200 Speaker 1: the accelerator and brake pedals, move the gear shift control, 30 00:01:50,320 --> 00:01:54,480 Speaker 1: and start the engine. These tasks, at least in experimental 31 00:01:54,520 --> 00:01:58,640 Speaker 1: self driving cars, are often accomplished with electric motors. A 32 00:01:58,680 --> 00:02:01,800 Speaker 1: motor is mounted so it can turn the existing steering 33 00:02:01,800 --> 00:02:05,240 Speaker 1: wheel for example. Another motor is able to put pressure 34 00:02:05,280 --> 00:02:08,519 Speaker 1: on the accelerator and brake pedals and so on. It's 35 00:02:08,560 --> 00:02:11,880 Speaker 1: not a particularly elegant solution, but it gets the job done. 36 00:02:12,280 --> 00:02:16,639 Speaker 1: As production cars become more advanced, drive by wire systems 37 00:02:16,639 --> 00:02:20,040 Speaker 1: are becoming a lot more common, So in a like 38 00:02:20,080 --> 00:02:23,840 Speaker 1: a Prius a Toyota Prius, if you push the brake pedal, 39 00:02:24,280 --> 00:02:27,079 Speaker 1: it might go to a computer and the computer actually 40 00:02:27,320 --> 00:02:31,960 Speaker 1: activates the brakes. There's no direct mechanical linkage between the 41 00:02:32,000 --> 00:02:36,000 Speaker 1: brake pedal and the brake mechanism that's called drive by 42 00:02:36,160 --> 00:02:38,720 Speaker 1: wire when the computer is sitting there in the middle. 43 00:02:39,320 --> 00:02:42,880 Speaker 1: In cars with drive by wire, a computer can hook 44 00:02:43,000 --> 00:02:46,440 Speaker 1: directly into that existing control system without having to have 45 00:02:46,560 --> 00:02:49,840 Speaker 1: these motors hanging out in the driver's area. The next 46 00:02:49,840 --> 00:02:52,960 Speaker 1: thing that a self driving car needs is sensors, and 47 00:02:53,040 --> 00:02:57,240 Speaker 1: here things get pretty complicated. Today, a human being drives 48 00:02:57,320 --> 00:03:00,440 Speaker 1: his car by using his or her eye is as 49 00:03:00,520 --> 00:03:04,440 Speaker 1: the soul sensor. The human visual system is amazing in 50 00:03:04,480 --> 00:03:07,959 Speaker 1: this regard because it's able to accurately judge the presence 51 00:03:07,960 --> 00:03:12,760 Speaker 1: of obstacles, their distances, their relative size based on distance, 52 00:03:12,800 --> 00:03:16,800 Speaker 1: their speed, and so on. A human being also recognizes 53 00:03:16,880 --> 00:03:19,560 Speaker 1: what he or she is seeing, So if a human 54 00:03:19,600 --> 00:03:23,520 Speaker 1: being sees a fence, you can accurately predict with high 55 00:03:23,560 --> 00:03:26,079 Speaker 1: certainty that that fence is not going to jump out 56 00:03:26,080 --> 00:03:28,200 Speaker 1: into the middle of the road and do something to you. 57 00:03:28,520 --> 00:03:31,000 Speaker 1: On the other hand, if there's a child playing with 58 00:03:31,040 --> 00:03:33,760 Speaker 1: a ball on a sidewalk, it's a completely different story, 59 00:03:33,840 --> 00:03:38,600 Speaker 1: and humans can recognize that instantly. Computer vision systems are 60 00:03:38,680 --> 00:03:42,000 Speaker 1: nowhere near this point in their development, so they rely 61 00:03:42,200 --> 00:03:46,800 Speaker 1: on extra sensors to provide more information. Self driving cars 62 00:03:46,880 --> 00:03:50,800 Speaker 1: do have camera based vision systems, and they use it 63 00:03:50,840 --> 00:03:55,040 Speaker 1: to see other cars, unexpected obstacles, road marking signs, and 64 00:03:55,080 --> 00:03:58,840 Speaker 1: so on. But in addition, self driving cars almost always 65 00:03:58,840 --> 00:04:01,440 Speaker 1: have GPS sense, or so they have a better idea 66 00:04:01,480 --> 00:04:04,760 Speaker 1: of exactly where they are and where they're pointing. They 67 00:04:04,800 --> 00:04:08,880 Speaker 1: also have light our systems laser scanners that can look 68 00:04:08,960 --> 00:04:12,880 Speaker 1: for nearby objects and accurately judge their distance. They also 69 00:04:13,000 --> 00:04:17,680 Speaker 1: have radar systems normally, and they may have infrared sensors 70 00:04:17,720 --> 00:04:21,719 Speaker 1: to improve night vision. These sensors all feed into a 71 00:04:21,760 --> 00:04:26,920 Speaker 1: powerful onboard computer, often multiple computers that are usually located 72 00:04:26,960 --> 00:04:29,599 Speaker 1: back in the back catch area of the trunk. They 73 00:04:29,680 --> 00:04:33,680 Speaker 1: could fill an entire trunk in fact, and these process 74 00:04:33,839 --> 00:04:37,279 Speaker 1: all the data that the sensors are gathering. The internal 75 00:04:37,279 --> 00:04:40,200 Speaker 1: computer will have access to a database of maps and 76 00:04:40,240 --> 00:04:44,880 Speaker 1: other relevant information as well. For example, engineers may pre 77 00:04:45,080 --> 00:04:50,880 Speaker 1: drive or route and pre catalog all the signs, road markings, curves, crosswalks, 78 00:04:50,920 --> 00:04:53,920 Speaker 1: traffic lights, and so on that the self driving car 79 00:04:54,000 --> 00:04:57,599 Speaker 1: will encounter along the road. This way, the car knows 80 00:04:57,680 --> 00:05:02,200 Speaker 1: what to expect and complan accordingly. Finally, the onboard computer 81 00:05:02,279 --> 00:05:05,960 Speaker 1: may also be communicating by radio, with bigger computers holding 82 00:05:06,040 --> 00:05:10,320 Speaker 1: even more data in nearby buildings. You might wonder why 83 00:05:10,360 --> 00:05:14,120 Speaker 1: the driving task, which seems effortless to just about any 84 00:05:14,160 --> 00:05:17,640 Speaker 1: normal human adult, would be so hard for a computer. 85 00:05:18,200 --> 00:05:22,160 Speaker 1: It's because the human brain is deceptive. Although it's small 86 00:05:22,200 --> 00:05:25,760 Speaker 1: and silent compared to a desktop computer, the human brain 87 00:05:25,960 --> 00:05:30,560 Speaker 1: still has more power than mankind's biggest fills a whole 88 00:05:30,640 --> 00:05:34,680 Speaker 1: building supercomputers. The massive amount of computing power in the 89 00:05:34,760 --> 00:05:37,560 Speaker 1: human brain is what makes driving appear to be so 90 00:05:37,640 --> 00:05:41,320 Speaker 1: easy for us. In thirty or forty years, our computers 91 00:05:41,320 --> 00:05:43,839 Speaker 1: will have that same kind of power in a small 92 00:05:43,920 --> 00:05:50,039 Speaker 1: package if things keep progressing at their current rate. Be 93 00:05:50,120 --> 00:05:52,760 Speaker 1: sure to check out our new video podcast, Stuff from 94 00:05:52,800 --> 00:05:55,640 Speaker 1: the Future. Join how Stuff Work staff as we explore 95 00:05:55,680 --> 00:06:01,080 Speaker 1: the most promising and perplexing possibilities of tomorrow. The Housuffork's 96 00:06:01,080 --> 00:06:04,359 Speaker 1: iPhone app has arrived. Download it today on iTunes.