Skip to main content

tv   Panel on Autonomous Vehicles at The Atlantic Festival  CSPAN  December 28, 2018 11:51pm-12:10am EST

11:51 pm
education, law enforcement, first responders, fire departments, civic groups, faith-based loops, business groups, recovery community, the treatment community to my youth athletics -- you name it. .veryone has a stake in this when you rally all of them together, it is amazing what you get done if you can leverage the best practices out there. there are good nonprofits out there. we took an early philosophy that this is not a competition. the worst things can do is let your ego get attached and politicize it. we decided we will work with any fellow nonprofits. lift all boats if we can. it is a multidimensional effort to get this done. host: think you for being with us and thank you for your story. admiral winnefeld: my pleasure. [applause] [applause]
11:52 pm
you started your career and autonomous technology by helping google build their autonomous version. and then you decided to branch out on your own to start aurora. what is aurora doing differently? i have been working in self driving cars for 15 years. i had a chance to spend years that google, and at the end of it it became time to move on. i spent a few months trying to figure out what to do next, looking at everything from lying on a beach to do something with self-driving car's. as i look at the moment, there was a unique period where you and ibuild a company,
11:53 pm
look at the opportunity, leaping in and trying it where worst case handed over to somebody else. if i chose not, i would look back 20 years and regretted. it seemed like a clear choice. we have an amazing pool of talent. the three of us who founded it, i have been working in the space for a long time, sterling anderson, one of our cofounders helped launch autopilots. we have shipped stuff in the space. with that court, the 160 people we have hired since, we are building a team that is mission driven and once to work to bring this technology to market. worth making a confession to you in the audience, my interest in
11:54 pm
autonomous vehicles is self-serving. i love road trips, but i am the person who does not want to drive for over 45 minutes. when i am on a road trip, and we hit traffic, when you get reason it, the saddest there was a car wreck or accident, and the more likely reason is something silly happened, like one lane is closed. we are trying to merge, half want to drive all the way up and cut in early and it is causing backup. , anyoneakes me incensed driving with me has to listen to a tirade about how so many humans should not be driving because we are equally responsible, and we are not paying attention, and it goes on and on. when are we getting to the point where i can hop in and
11:55 pm
autonomous vehicle and drive across country, not have to feel bad for not wanting to drive for long times, but not have to worry about human errors? [laughter] what is incredible is that we are in the moments. today, if you go to a suburb of phoenix, arizona, the team i used to lead has a small number of vehicles on the road today driving with nobody behind the wheel. that is exciting. it is not commercially viable yet, and it is not all of phoenix. over the next five years you will see more broad capabilities where they can try lamar street's. -- where they can try on more streets. over the next 50 years, you will see this technology come to market to scale where it will profoundly impact the way we get
11:56 pm
around. uber crash in an phoenix, and other crashes with autonomous vehicles. how as an industry do you deal with that? and how much do you think that sets you back? how does it set back the idea of a driverless car? chris: this comes back to our to deliver theon technology safely and broadly. we take those words carefully, we think that is the stakes. 40,000 people killed every year on american roads, and worldwide is 1.3 million people. we think there is a profound opportunity to make that better.
11:57 pm
what happened in phoenix is a tragedy. failure thatural led to software that should not have been driving, and operational mistakes that led to a tragic outcome. as an industry, we have to look at how we are being thoughtful about bringing this technology to market, how we are testing it so we will be respectful that we are on other people's roads. and put the first part of the mission first safety first. us understand what the limitations of the .echnology are when accidents happen, things go insights a fear people have about robots and robotics, and the tendency that we should throw the baby out with the bathwater. chris: that is one of the worst
11:58 pm
things we can do. it is easy to forget how bad the status quo is. 40,000 americans who die every year is incredible. we accept that as the cost of getting from a to b. the economic impact is something drag. trillion dollar there will be mistakes, that systems will be fallible. we cannot let perfect be the enemy of way better. technology,p this we think about what happens when systems fail. how we understand weird events on the road, and how do we design for that. that is what trips up human drivers. paying attention to what you were doing, you are actually quite good. times forhappen 1.15 100 million miles.
11:59 pm
an incredible number. when you are looking at your phone, distracted, fall asleep, too much to drink, that is when bad things happen. what is exciting about this technology, all of those >> one of the things i think about a lot when i write about the change in technology, especially with things people do all the time is how they will be regulated. this systemave where technology gets a way out in front and then some of the crack's in the foundation, when regulation starts to pop up and regulation tries to catch up. what regulations make sense to you? this as almost a feature, not a bug. i think this is what has allowed the united states over centuries to be a technology leader. system here is
12:00 am
permissive. there is always a backstop. whether it is the recall and they, local police ability to say something is unsafe and taken off the road. or whether it is the legal system. all of that acts as a back stop to protect the public. he contrasts that with the european system, which is anything you want to do you have to ask permission first. au have to ask permission of bureaucrat who is incentivized to say no. it makes it hard to have a vision. said, we are talking about something that is going to be safety critical. i think they have done a very thoughtful job of putting out their own framework for anyone who has worked in this space to operate within it and share that information.
12:01 am
over my time at google and now at aurora we engaged with state regulators and allow them to be apprised of a we were doing and why we think this is so important and how we are pink awful in bringing it to market. not just the people who create the technology, the few people who would use it, we need automakers and at local leaders, i'm wondering how you think about getting by all of those various layers? >> it is hard. there are days where i lament on making web apps for a living. i think it should be. we are talking about something that is fundamentally important. transportation, everything, whether it is coming here this morning or the food that came to your table. all of it is moved through our transportation system and that is what we are talking about.
12:02 am
good news is i think the technology speaks for itself. i have done hundreds of demos over the years. will come ineople skeptical and assume there is some kind of smoke and mirror. riding few short minutes one of the vehicles, assuming it of beingg well, -- arm on the freeway in california not touching the steering wheel. 10 minutes in, the woman sitting next to me said is that all it does. [laughter] it is a self driving car, i don't know what you expected. this is one of those experiences we all have. we have all been in the cars that effectively drive themselves. whether it is as a toddler in the backseat or when my wife is driving. it is just part of our lives. >> i talked a little bit about
12:03 am
some of the challenges i foresee. are there any challenges that you guys are facing as you try and move ahead with the technology that we are not even thinking about? >> we talk about this a fair bit but one of the hardest problems is understanding what the other drivers on the road are doing. it turns out that most of the time, 99% of the time, the other drivers do exactly what you think. for those on the ones that are interesting, those are not the ones you are worrying about. as the ones that decided they needed to go to mcdonald's half a block ago and make a left turn. i took to theride dinner last night, the driver the road was too busy and cut across the median. that was exciting to me. those are the kind of events
12:04 am
where -- the autonomy system to be able to learn what normal behavior is, but also be flexible enough to deal with these kind of rare events and actually lead to the bad events. >> talk to me about how that works. i'm imagining where we are on the highway where it is half autonomous vehicles and half people on the highway doing whatever they do. how do you plan ahead for that future where we are not at the point where everyone is kind of in the jetson's vehicle? >> if we could snap our fingers and say there is no more human driven cars, my job would be so much easier. policymakersome that would like to get on that, that would be great. in reality, that will never happen. we take our vehicles out on the public road and this is why it is so important to be doing some of this with other people. it is because our vehicles can
12:05 am
observe how they behave. for the learning techniques we use, we can hold that experience and build a model of what to expect of other drivers when they are behaving normally, or abnormally. we can build a model of how we drive so that we do not surprise other drivers on the road. we talked a little bit about the astronomical number of people who suffer pay colonies -- fatalities on the road. the dream would be that number is zero. how low do you think that number can go? >> i think the autonomous vehicles we can take out 95% of those accidents. if you look at statistics today, 95% of the fatal accidents are caused by human error. 5% are due to other failures in the environment. if you look at aviation which is the gold standard for this, is
12:06 am
pretty much zero. centuries todes or come, that would be our aspiration. in the near term, i would be excited if we got to half as many. >> thank you so much for taking time to join us. [applause] [captions copyright national cable satellite corp. 2018] [captioning performed by the national captioning institute, which is responsible for its caption content and accuracy. visit ncicap.org] c-span, where history unfolds daily. in 1979 c-span was created as a public service by america's cable television companies. we continue to bring you unfiltered coverage of congress, the white house, the supreme court, and public policy events in washington dc and around the country. by yours brought to you cable or satellite provider. it has been seven days since the government shutdown began. bordertions over
12:07 am
security between congress and the white house remain at a standstill as president trump continues to call for $5 billion in border wall funding, which democrats are objecting to. at 10:00meets monday a.m. eastern but no votes are expected. you can follow the house live on c-span and the senate live on c-span two. >> the united states senate, a uniquely american institution. legislating in carrying out constitutional duty since 1789. >> please raise your right hand. >> wednesday, c-span takes you inside the senate. learning about the legislative body and its informal workings. >> arguing about things and
12:08 am
kicking them around and having great debates is a thoroughly american thing. are in theer you senate, the more you will appreciate the cooling nature. >> we will look at the history of conflict and compromise with original interviews, key moments in history, and unprecedented access. allowing us to bring cameras into the senate chambers during a session. follow the evolution of the senate into the modern era. to thevice and consent role in impeachment proceedings and investigations. senate, conflict and compromise. a c-span original production. explore the history, traditions, and roles of this uniquely american institution. 8:00years wednesday at p.m. eastern and pacific, on c-span. be sure to go online at c-span.org/senate to learn more about the program and watch
12:09 am
original, full-length interviews with senators, view farewell speeches from a long serving members, and take a tour inside the senate chamber, the old senate chamber, and other exclusive locations. next, farewell speeches from three senators who are leaving congress. we hear first from missouri senator claire mccaskill, who lost her reelection bid. she is followed by arizona republican jeff flake being about his retirement, after that nevada republican dean heller says goodbye to his colleagues and it reflects on his time in congress. sen. mccaskill: it probably won't surprise my colleagues to know that i don't like much the idea of a farewell speech. farewell speech. i haven't spent a great deal of time contemplating it over the years i have been here. i'm not a big fan of the concept. but i want to respect the

8 Views

info Stream Only

Uploaded by TV Archive on