Gadgetory


All Cool Mind-blowing Gadgets You Love in One Place

Riding a Waymo self-driving taxi

2018-12-05
- It's 2018, where's my self driving car. Oh, here it is. This is Waymo, but you probably know it as Google's self driving car project. Now Waymo has long been considered the leader of autonomous vehicles and that's because it's been working on this technology longer than any other company. It's vehicles have driven the most miles and it has the lowest rate of disengagements or instances when humans have to take control of the vehicle and it's all been leading up to this, a real robot taxi service right here in Phoenix Arizona. Waymo's been operating it self driving cars in Arizona for about a year and a half and during that time it's even run them without safety drivers behind the wheel. Waymo's cars collect a ton of data and Waymo feeds that into it's deep learning algorithm for labeling and processing and this allows Waymo to deploy what it calls the safest driver on the road. Now Waymo is ready to start taking on passengers and create a real business for itself. It works basically the same way as any ride sharing service. You pull out the app, you pick your destination and hail a ride. A few minutes later one of Waymo's self driving Chrysler Pacifica mini vans pulls up. These are covered in high powered sensors like Lidar, radar, and ultrasonic along with a bunch of cameras to provide a full 360 degree omnidirectional field of view. Waymo is calling it's taxi service Waymo One and initially we were going to show you a demo of the ride that we took through Chandler, but truthfully it was kind of long and kind of boring, but not entirely boring. There were these touch screens behind the head rest that present a live view of what the car sensors can see. Other vehicles are blue, pedestrians are white and your trip is shown as a green line that the Waymo follows. It was pretty cool, but overall the ride itself was uneventful as it should be. It was like being in the back seat with a very cautious student driver. And sometimes it would take longer than usual to make a decision like this unprotected left into a parking lot. Now this is gonna be a complicated turn right here. We're taking a left hand turn across traffic into this parking lot. So the car is sort of figuring out where it can go, when it can go, if there's no traffic signal or signage to tell us when to go, the car needs to make that determination itself. It needs to find that gap in the traffic. Waymo programs it's cars to be conservative, to be safe drivers. A human driver probably would've gone right then. That might be comforting to some people, other people might find that a little bit annoying and would want to see the car being a little bit more assertive and that's something that Waymo is working on. It has been sort of trying to develop it's software, tune it's software in a way that the car drives more organically, more like a human. There were moments when the care acted more aggressively than I expected, but it was the type of behavior that you wouldn't notice if you were being driven by a human driver. There were jerky moments sure, but it never felt unsafe and I never felt car sick for that matter. - We want a service that brings safety to all of roads. That's really the core mission and it's around making sure that we constantly are tackling this problem of road safety. - So the company thinks it can build a real business by offering a safer, more convenient and perhaps cheaper service than Uber or Lyft. To start out with Waymo1 will only be available in a few towns around Phoenix and it'll be limited to just a couple members of Waymo's Early Rider Program. - For Waymo1 we'll have an itterative approach that will start with Waymo try drivers and eventually we'll take those out. - And when that happens, someone hails a care, it is possible a car with no one inside of it may role up to pick them up? - That's our vision, that's where we wanna get to. Is somebody has that space to themselves. - Waymo One will be limited to only certain neighborhoods meaning it'll be geofenced, and the cars will feature safety drivers behind the wheel. Even though the company had said it intends to deploy fully driverless vehicles at launch. So as you can see Waymo's approaching this whole commercialization thing super cautiously and there's a good reason for that. In March 2018 a self driving Uber vehicle struck and killed a 49 year old woman in Tempe Arizona, which is just a few miles from here. The Uber crash brought those early inflated expectations about self driving cars to a screeching halt, but in the days after the crash Waymo's CEO John Krafcik said that it's cars would have been able to prevent it. - We have a lot of confidence that our technology would be robust and would be able to handle situations like that one. Cocky, sure, but now Waymo has to bend over backwards to prove that it's cars are the safest on the road and it needs to convert a skeptical public, who just doesn't trust self driving cars. I've seen the surveys, people just don't wanna get into these cars. But how do you make people believe that a self driving car is safer than a human driven one? - One of the ways I think is around experience. We are working on building the world's most experienced driver. We've now driven over 10 million miles on public road in autonomous mode. That really allows us to continue to learn new skills, vet our current functionality, and test out, with the most extreme and challenging situations. - So self driving cars are suppose to be the future of transportation, but is the solution to all our transportation problems really more cars? US cities are overwhelmingly dominated by personally owned vehicles, but that landscape is changing. Millennials want better public transportation and more options to bike and walk around and they're getting that with dockless bike and scooter programs growing like crazy. And 51% of millennials don't think that owning a car is worth the investment. Just take Phoenix for example, it's the 25th most congested city in the US and drivers here spend 34 hours a year stuck in traffic. The real danger is for people out walking around those cars because Arizona has the highest pedestrian death rate in the nation. Between 2014 and 2017, 271 pedestrians were killed in fatal collisions. In 2018, 30 people were killed. Are self driving cars really the solution? - Driverless cars do have the opportunity to reduce traffic fatalities 'cause they could reduce a lot of the human error that leads to that, but at the same time we don't really have any data that supports that just because the number of miles driven by driverless cars so far is so small compared to the data set that we have for human driven cars, but it also kind of side steps one of the bigger problems, which is how our streets are actually designed and how their designed to be unsafe for a lot of people. - So while Waymo's self driving cars haven't caused any fatal collisions, there have been reports of some minor fender benders and even a few injuries, and Arizona drivers have complained that these cars can be a little bit annoying to navigate around. These antidotes also underscore a really important point about full autonomy, cars that can drive themselves anywhere under any conditions. That it may take decades for this to arrive, if ever. Even Waymo's CEO John Krafcik admitted this. He said that, "Autonomy will always have constraints." Driverless vehicles will need to stay on the road almost around the clock to offset the cost of the sensors, and the hardware, and the software, and the computer chips and everything that makes them drive without human beings. That means keeping these vehicles on the road for hundreds and thousands of miles. That's a lot of driving and while Waymo's cars may not succumb to the same failings as human driven ones they'll still occupy the same amount of space on the road. Now Uber and Lyft have been shown to increase traffic congestion in certain cities. What will Waymo's contribution be? Well to start out, not that much. As the company eases it's way into becoming a real business it's main focus is gonna be keeping up with demand. I mean people are gonna wanna use this. This is a robot taxi service. It's the future, right? So would you ride in a self driving car? Leave us a comment in the comment section and if you're interested in cars or the future of transportation check out this awesome video that we just did about Tesla. It's on youtube.com/theverge. YouTube, You tub.
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites.