The future race car -- 150mph, and no driver
-
0:01 - 0:02So, how many of you have ever
-
0:02 - 0:04gotten behind the wheel of a car
-
0:04 - 0:09when you really shouldn't have been driving?
-
0:09 - 0:11Maybe you're out on the road for a long day,
-
0:11 - 0:13and you just wanted to get home.
-
0:13 - 0:15You were tired, but you felt you could drive a few more miles.
-
0:15 - 0:16Maybe you thought,
-
0:16 - 0:19I've had less to drink than everybody else,
-
0:19 - 0:20I should be the one to go home.
-
0:20 - 0:25Or maybe your mind was just entirely elsewhere.
-
0:25 - 0:26Does this sound familiar to you?
-
0:26 - 0:29Now, in those situations, wouldn't it be great
-
0:29 - 0:31if there was a button on your dashboard
-
0:31 - 0:37that you could push, and the car would get you home safely?
-
0:37 - 0:39Now, that's been the promise of the self-driving car,
-
0:39 - 0:42the autonomous vehicle, and it's been the dream
-
0:42 - 0:45since at least 1939, when General Motors showcased
-
0:45 - 0:49this idea at their Futurama booth at the World's Fair.
-
0:49 - 0:51Now, it's been one of those dreams
-
0:51 - 0:55that's always seemed about 20 years in the future.
-
0:55 - 0:57Now, two weeks ago, that dream took a step forward,
-
0:57 - 1:01when the state of Nevada granted Google's self-driving car
-
1:01 - 1:04the very first license for an autonomous vehicle,
-
1:04 - 1:07clearly establishing that it's legal for them
-
1:07 - 1:08to test it on the roads in Nevada.
-
1:08 - 1:12Now, California's considering similar legislation,
-
1:12 - 1:14and this would make sure that the autonomous car
-
1:14 - 1:17is not one of those things that has to stay in Vegas.
-
1:17 - 1:20(Laughter)
-
1:20 - 1:23Now, in my lab at Stanford, we've been working on
-
1:23 - 1:27autonomous cars too, but with a slightly different spin
-
1:27 - 1:31on things. You see, we've been developing robotic race cars,
-
1:31 - 1:35cars that can actually push themselves to the very limits
-
1:35 - 1:37of physical performance.
-
1:37 - 1:40Now, why would we want to do such a thing?
-
1:40 - 1:42Well, there's two really good reasons for this.
-
1:42 - 1:46First, we believe that before people turn over control
-
1:46 - 1:49to an autonomous car, that autonomous car should be
-
1:49 - 1:52at least as good as the very best human drivers.
-
1:52 - 1:56Now, if you're like me, and the other 70 percent of the population
-
1:56 - 1:58who know that we are above-average drivers,
-
1:58 - 2:01you understand that's a very high bar.
-
2:01 - 2:03There's another reason as well.
-
2:03 - 2:07Just like race car drivers can use all of the friction
-
2:07 - 2:08between the tire and the road,
-
2:08 - 2:11all of the car's capabilities to go as fast as possible,
-
2:11 - 2:15we want to use all of those capabilities to avoid
-
2:15 - 2:16any accident we can.
-
2:16 - 2:18Now, you may push the car to the limits
-
2:18 - 2:20not because you're driving too fast,
-
2:20 - 2:22but because you've hit an icy patch of road,
-
2:22 - 2:24conditions have changed.
-
2:24 - 2:27In those situations, we want a car
-
2:27 - 2:31that is capable enough to avoid any accident
-
2:31 - 2:33that can physically be avoided.
-
2:33 - 2:38I must confess, there's kind of a third motivation as well.
-
2:38 - 2:40You see, I have a passion for racing.
-
2:40 - 2:43In the past, I've been a race car owner,
-
2:43 - 2:45a crew chief and a driving coach,
-
2:45 - 2:49although maybe not at the level that you're currently expecting.
-
2:49 - 2:52One of the things that we've developed in the lab --
-
2:52 - 2:53we've developed several vehicles --
-
2:53 - 2:56is what we believe is the world's first
-
2:56 - 2:58autonomously drifting car.
-
2:58 - 3:00It's another one of those categories
-
3:00 - 3:03where maybe there's not a lot of competition.
-
3:03 - 3:04(Laughter)
-
3:04 - 3:08But this is P1. It's an entirely student-built electric vehicle,
-
3:08 - 3:10which through using its rear-wheel drive
-
3:10 - 3:12and front-wheel steer-by-wire
-
3:12 - 3:14can drift around corners.
-
3:14 - 3:16It can get sideways like a rally car driver,
-
3:16 - 3:18always able to take the tightest curve,
-
3:18 - 3:21even on slippery, changing surfaces,
-
3:21 - 3:23never spinning out.
-
3:23 - 3:25We've also worked with Volkswagen Oracle,
-
3:25 - 3:28on Shelley, an autonomous race car that has raced
-
3:28 - 3:32at 150 miles an hour through the Bonneville Salt Flats,
-
3:32 - 3:36gone around Thunderhill Raceway Park in the sun,
-
3:36 - 3:39the wind and the rain,
-
3:39 - 3:44and navigated the 153 turns and 12.4 miles
-
3:44 - 3:45of the Pikes Peak Hill Climb route
-
3:45 - 3:49in Colorado with nobody at the wheel.
-
3:49 - 3:50(Laughter)
-
3:50 - 3:56(Applause)
-
3:56 - 3:59I guess it goes without saying that we've had a lot of fun
-
3:59 - 4:00doing this.
-
4:00 - 4:04But in fact, there's something else that we've developed
-
4:04 - 4:07in the process of developing these autonomous cars.
-
4:07 - 4:11We have developed a tremendous appreciation
-
4:11 - 4:15for the capabilities of human race car drivers.
-
4:15 - 4:19As we've looked at the question of how well do these cars perform,
-
4:19 - 4:22we wanted to compare them to our human counterparts.
-
4:22 - 4:28And we discovered their human counterparts are amazing.
-
4:28 - 4:32Now, we can take a map of a race track,
-
4:32 - 4:34we can take a mathematical model of a car,
-
4:34 - 4:37and with some iteration, we can actually find
-
4:37 - 4:39the fastest way around that track.
-
4:39 - 4:41We line that up with data that we record
-
4:41 - 4:43from a professional driver,
-
4:43 - 4:47and the resemblance is absolutely remarkable.
-
4:47 - 4:50Yes, there are subtle differences here,
-
4:50 - 4:53but the human race car driver is able to go out
-
4:53 - 4:56and drive an amazingly fast line,
-
4:56 - 4:58without the benefit of an algorithm that compares
-
4:58 - 5:01the trade-off between going as fast as possible
-
5:01 - 5:03in this corner, and shaving a little bit of time
-
5:03 - 5:05off of the straight over here.
-
5:05 - 5:08Not only that, they're able to do it lap
-
5:08 - 5:10after lap after lap.
-
5:10 - 5:13They're able to go out and consistently do this,
-
5:13 - 5:17pushing the car to the limits every single time.
-
5:17 - 5:21It's extraordinary to watch.
-
5:21 - 5:23You put them in a new car,
-
5:23 - 5:27and after a few laps, they've found the fastest line in that car,
-
5:27 - 5:30and they're off to the races.
-
5:30 - 5:32It really makes you think,
-
5:32 - 5:36we'd love to know what's going on inside their brain.
-
5:36 - 5:41So as researchers, that's what we decided to find out.
-
5:41 - 5:43We decided to instrument not only the car,
-
5:43 - 5:45but also the race car driver,
-
5:45 - 5:48to try to get a glimpse into what was going on
-
5:48 - 5:50in their head as they were doing this.
-
5:50 - 5:54Now, this is Dr. Lene Harbott applying electrodes
-
5:54 - 5:55to the head of John Morton.
-
5:55 - 5:58John Morton is a former Can-Am and IMSA driver,
-
5:58 - 6:00who's also a class champion at Le Mans.
-
6:00 - 6:04Fantastic driver, and very willing to put up with graduate students
-
6:04 - 6:06and this sort of research.
-
6:06 - 6:08She's putting electrodes on his head
-
6:08 - 6:10so that we can monitor the electrical activity
-
6:10 - 6:13in John's brain as he races around the track.
-
6:13 - 6:16Now, clearly we're not going to put a couple of electrodes on his head
-
6:16 - 6:20and understand exactly what all of his thoughts are on the track.
-
6:20 - 6:23However, neuroscientists have identified certain patterns
-
6:23 - 6:27that let us tease out some very important aspects of this.
-
6:27 - 6:29For instance, the resting brain
-
6:29 - 6:31tends to generate a lot of alpha waves.
-
6:31 - 6:35In contrast, theta waves are associated with
-
6:35 - 6:38a lot of cognitive activity, like visual processing,
-
6:38 - 6:41things where the driver is thinking quite a bit.
-
6:41 - 6:42Now, we can measure this,
-
6:42 - 6:44and we can look at the relative power
-
6:44 - 6:47between the theta waves and the alpha waves.
-
6:47 - 6:49This gives us a measure of mental workload,
-
6:49 - 6:53how much the driver is actually challenged cognitively
-
6:53 - 6:54at any point along the track.
-
6:54 - 6:57Now, we wanted to see if we could actually record this
-
6:57 - 7:00on the track, so we headed down south to Laguna Seca.
-
7:00 - 7:02Laguna Seca is a legendary raceway
-
7:02 - 7:05about halfway between Salinas and Monterey.
-
7:05 - 7:07It has a curve there called the Corkscrew.
-
7:07 - 7:10Now, the Corkscrew is a chicane, followed by a quick
-
7:10 - 7:12right-handed turn as the road drops three stories.
-
7:12 - 7:16Now, the strategy for driving this as explained to me was,
-
7:16 - 7:18you aim for the bush in the distance,
-
7:18 - 7:21and as the road falls away, you realize it was actually the top of a tree.
-
7:21 - 7:24All right, so thanks to the Revs Program at Stanford,
-
7:24 - 7:26we were able to take John there
-
7:26 - 7:27and put him behind the wheel
-
7:27 - 7:29of a 1960 Porsche Abarth Carrera.
-
7:29 - 7:33Life is way too short for boring cars.
-
7:33 - 7:35So, here you see John on the track,
-
7:35 - 7:37he's going up the hill -- Oh! Somebody liked that --
-
7:37 - 7:39and you can see, actually, his mental workload
-
7:39 - 7:42-- measuring here in the red bar --
-
7:42 - 7:44you can see his actions as he approaches.
-
7:44 - 7:47Now watch, he has to downshift.
-
7:47 - 7:48And then he has to turn left.
-
7:48 - 7:52Look for the tree, and down.
-
7:52 - 7:55Not surprisingly, you can see this is a pretty challenging task.
-
7:55 - 7:58You can see his mental workload spike as he goes through this,
-
7:58 - 8:00as you would expect with something that requires
-
8:00 - 8:03this level of complexity.
-
8:03 - 8:06But what's really interesting is to look at areas of the track
-
8:06 - 8:09where his mental workload doesn't increase.
-
8:09 - 8:10I'm going to take you around now
-
8:10 - 8:11to the other side of the track.
-
8:11 - 8:14Turn three. And John's going to go into that corner
-
8:14 - 8:16and the rear end of the car is going to begin to slide out.
-
8:16 - 8:18He's going to have to correct for that with steering.
-
8:18 - 8:21So watch as John does this here.
-
8:21 - 8:23Watch the mental workload, and watch the steering.
-
8:23 - 8:27The car begins to slide out, dramatic maneuver to correct it,
-
8:27 - 8:30and no change whatsoever in the mental workload.
-
8:30 - 8:33Not a challenging task.
-
8:33 - 8:36In fact, entirely reflexive.
-
8:36 - 8:40Now, our data processing on this is still preliminary,
-
8:40 - 8:42but it really seems that these phenomenal feats
-
8:42 - 8:44that the race car drivers are performing
-
8:44 - 8:46are instinctive.
-
8:46 - 8:49They are things that they have simply learned to do.
-
8:49 - 8:52It requires very little mental workload
-
8:52 - 8:54for them to perform these amazing feats.
-
8:54 - 8:58And their actions are fantastic.
-
8:58 - 9:00This is exactly what you want to do on the steering wheel
-
9:00 - 9:04to catch the car in this situation.
-
9:04 - 9:07Now, this has given us tremendous insight
-
9:07 - 9:10and inspiration for our own autonomous vehicles.
-
9:10 - 9:12We've started to ask the question:
-
9:12 - 9:14Can we make them a little less algorithmic
-
9:14 - 9:17and a little more intuitive?
-
9:17 - 9:19Can we take this reflexive action
-
9:19 - 9:21that we see from the very best race car drivers,
-
9:21 - 9:23introduce it to our cars,
-
9:23 - 9:25and maybe even into a system that could
-
9:25 - 9:27get onto your car in the future?
-
9:27 - 9:29That would take us a long step
-
9:29 - 9:31along the road to autonomous vehicles
-
9:31 - 9:33that drive as well as the best humans.
-
9:33 - 9:36But it's made us think a little bit more deeply as well.
-
9:36 - 9:39Do we want something more from our car
-
9:39 - 9:41than to simply be a chauffeur?
-
9:41 - 9:45Do we want our car to perhaps be a partner, a coach,
-
9:45 - 9:48someone that can use their understanding of the situation
-
9:48 - 9:53to help us reach our potential?
-
9:53 - 9:55Can, in fact, the technology not simply replace humans,
-
9:55 - 10:00but allow us to reach the level of reflex and intuition
-
10:00 - 10:03that we're all capable of?
-
10:03 - 10:05So, as we move forward into this technological future,
-
10:05 - 10:08I want you to just pause and think of that for a moment.
-
10:08 - 10:12What is the ideal balance of human and machine?
-
10:12 - 10:13And as we think about that,
-
10:13 - 10:15let's take inspiration
-
10:15 - 10:18from the absolutely amazing capabilities
-
10:18 - 10:21of the human body and the human mind.
-
10:21 - 10:23Thank you.
-
10:23 - 10:27(Applause)
- Title:
- The future race car -- 150mph, and no driver
- Speaker:
- Chris Gerdes
- Description:
-
Autonomous cars are coming -- and they’re going to drive better than you. Chris Gerdes reveals how he and his team are developing robotic race cars that can drive at 150 mph while avoiding every possible accident. And yet, in studying the brainwaves of professional racing drivers, Gerdes says he has gained a new appreciation for the instincts of professional drivers. (Filmed at TEDxStanford.)
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDTalks
- Duration:
- 10:47
Thu-Huong Ha edited English subtitles for The future race car -- 150mph, and no driver | ||
Thu-Huong Ha approved English subtitles for The future race car -- 150mph, and no driver | ||
Thu-Huong Ha accepted English subtitles for The future race car -- 150mph, and no driver | ||
Thu-Huong Ha edited English subtitles for The future race car -- 150mph, and no driver | ||
Morton Bast added a translation |