141–150 of 175 posts
Earlier quoted context omitted.
An oft-repeated adage in the self-driving engineering circles is "the last 10% is 90% of the work"
Isn't that just the 90-90 rule? https://en.wikipedia.org/wiki/Ninety-ninety_rule
Keeping the safety drivers in the beginning, scaling up, and getting a year or more worth of actual real world use, before slowly scaling back the drivers is what I'd call a very responsible choice. IMHO, to do less would be reckless showboating that could do unimaginable damage to the AV industry, as well as end up endangering lives. In the tech industry, we don't like to wait, we like moving fast, taking risks, shi…
My favorite way to tell if a car is a self-driving car is to look and see if it has a driver in it.
https://www.foxnews.com/auto/tesla-apparently-on-autopilot-p...
"After following the sedan for about seven miles trying to get him to respond to lights and sirens, the officers guessed that the Tesla’s semi-autonomous Autopilot driver assist system was engaged, so they pulled in front of it and started to slow down, according to a police report.
The sedan did the same, and the two vehicles safely came to a complete stop in the middle of the highway. With some difficulty, police roused the driver, who was identified as Los Altos Planning Commission chair ..."
One feature I like is the LCD displays for passengers, which (supposedly) show what the Waymo car is "thinking". So if the car stops unexpectedly, the display might show, "waiting for a crossing pedestrian". Ideally with a simple map diagram of what's around the vehicle.
Earlier quoted context omitted.
> It looks like they’re using Chrysler Pacificas which already have wheelchair ramp conversions from VMI and Braun. How common is this for other makes/models? If it is rare, I wonder if it influenced Google's choice of vehicle?
In terms of minivans, most American makes and models have conversions from at least one of the conversion manufacturers. Braun even has a converted Ford Explorer. The one issue I can see is that Waymo is using Pacifica Hybrids which can't be converted because the floor of converted vans is lowered ~12-14 inches which gets in the way of the batteries. But I can't imagine it would hurt to use a handful of gas-powered v…
Earlier quoted context omitted.
As excited as I am for self driving, it makes me sad that I get car/motion sick. I can't exploit self driving to do laptop work, reading, gaming, etc without getting sick - all I can do is sleep basically lol. It will be bittersweet in a self driving car realizing that since I don't have to pay attention, I can do so much more with my time!.. and then promptly stare out the window for an hour.
Is it the motion or seeing movement out of your peripheral vision?
In cars it's largely about how much I can see outside, if I'm facing the direction of movement, etc. Usually in common vehicles I do not suffer unless I read/phone/etc. However in a bus-limo thing on a work retreat we all sat sideways.. on windy roads.. I could still see out the big windows, and tried to face my body forward but it was a rough, rough ride regardless.
I've never investigated it much, as it's such an awful feeling I have a hard time debugging it.
I’m surprised to see, after what happened to Uber — the failure of their safety driver to stay engaged enough to stop the car before striking a pedestrian — that a safety driver who will be mostly disengaged from the driving experience is still thought to add an appropriate safety margin?
Earlier quoted context omitted.
That’s funny, a month ago my car was totaled by a guy who fell asleep and blew through a red light. That happened at a right angle but only at 30mph. Since then I’ve been concerned at every intersection about what the other cars will do. I’m not sure how much of it can be blamed on cell phone use, but I swear that human drivers have been getting substantially worse over the last few years (at least in the Bay Area).…
I think so much of it is just a scale issue. I grew up in a mid-sized suburb in the Northwest just a couple decades ago. Most road were only 2, or maximum 4 lanes. The idea of driving around without your headlights on in the dark would be unfathomable. And if you did, you'd get honked at, flashed, and presumably at some point within 5 or 10 minutes, a police officer would see you and stop you. That area's gotten bigg…
In Arizona you pretty much have to make the cops do their job unless they think they can get you for a DUI then they'll pull you over for the slightest infraction. The things I've seen people do when there's a cop right next to me at an intersection and not get stopped is mind boggling.
Earlier quoted context omitted.
>> Keeping the safety drivers in the beginning, scaling up, and getting a year or more worth of actual real world use, before slowly scaling back the drivers is what I'd call a very responsible choice. I'm sorry but where does this "one year" expectation come from? What says that Waymo needs a year (or two years, or ten, or three months, etc) to start "scaling back the drivers"? From what I can tell, the only way to…
No, the only way to know how well it's doing is for it to get enough "interesting" miles, and then compare the success rate of those miles to similar human-driven miles. Right now Waymo is past 8 million miles driven. Miles driven is a better and normalized metric, with comparable real world statistics, unlike time.
I know this is a heated debate, but it's interesting to see the divergent approaches that are being taken to achieve self driving. In one corner, it's lidar-based, equipping vehicles produced by other companies, organically and incrementally growing the navigation technology. In the other, it's video-based with now 100s of thousands of vehicles feeding data to a massive neural net. Each has very well thought-out argu…
In your two camps, I'd be curious to see why someone would specifically choose just video based? Ie, I'd expect (as a layman) neural nets to be able to consume many different data types. Why limit to just one? I wonder what is better about just video, rather than video+lidar+audio+x+y. Hell, one of the most neat things to me was how Teslas can scan multiple cars ahead to know of non-visible traffic blockages/etc. Usi…
other than cost? Teslas are consumer vehicles that are price sensitive.