Well, they certainly didn't under-sensor the thing. The big model Velodyne LIDAR, a small LIDAR looking upward, stereo cameras in all directions, and probably some radars.
The big rotating Velodyne thing has got to go. But so far, nobody who claims to have a suitable solid-state 3D LIDAR is actually shipping. Quanergy claimed a solid-state LIDAR suitable for automotive use [1][2] and showed it at CES, but never shipped. Their web site hasn't been updated since 2015. They were aiming at a price point around $100.
Advanced Scientific Concepts has a suitable LIDAR that costs far too much. Space-X uses an ASC unit on the Dragon capsule for docking. ASC announced they were getting into automotive in 2012 but never shipped.[3]
Somebody is going to do this soon. The technology works. It's a cost problem. Once the cost comes down, cars can have multiple LIDARs (at least fore and aft) near the top of the windshield and rear window. This gets rid of all that topside gear.
Quanergy raised $90 million on a $1.5 billion valuation last month. They announced products at $250 /unit and $100 / unit starting production in Q4 with full capacity Q1 2017.
November 17, 2015 09:00 AM Pacific Standard Time
SUNNYVALE, Calif.--(BUSINESS WIRE)
* Quanergy Systems, Inc., will introduce in early 2016 the world’s first solid state LiDAR (Light Detection and Ranging) sensor for self-driving cars for less than $1,000 per car, it was announced today at the Los Angeles Auto Show’s Connected Car Expo.*
They're already six months late. The press releases keep coming out, and the hype continues, but where's the hardware?
American Scientific Concepts has working units which do what Quanergy claims to do and more, but they seem content to sell them to DoD and Space-X at a high price.
"Quanergy’s LiDAR sensors are being used by five automakers that are developing automated vehicles."
That sounds like shipping to me. You don't ramp up volume production as an automotive supplier until your customers are ready. Naming Flex and Sensata/TI as manufacturers is a fairly specific and strong indicator that they are ramping,
unless you're suggesting securities fraud?
It looks more like a way to simplify manufacturing. Put all the sensors in one factory-built topside unit, then bolt that to the car. Less labor intensive than hanging stuff all over the car,
cheaper in small volumes than integration with the car body,
and you can align everything during manufacturing.
This sort of add-on thing is temporary. Another generation and cars will have an equally good sensor suite, but it will barely show.
I was thinking the exact same thing. It's not a beauty contest as long as Uber is owning the prototype fleet. Plus, I think it will clue in other drivers to: "Dude, that was one of those 'driverless ubers' right?"
I wonder if having all the gear will have a positive effect. Once drivers learn it means that those are self-driving cars it can begin to acclimate drivers to having such cars on the road. And as they see how (hopefully) predictably those cars drive it will increase their comfort level until they may be willing to purchase a self-driving car for themselves. I could also see a small rise in rear end collisions as people think that they can safely tailgate a self-driving car believing that it will not stop suddenly, that it will not act in the unpredictable ways that normally make us cautious of other drivers/cars on the road.
I still can't believe how under-sensored Tesla's "Auto Pilot" is. It's dangerous marketing it as self driving and not forcing drivers to pay attention.
Objectively, tonnes of people are using it right now, have been for months (is it a year yet?) and there has been one incident.
In my opinion, it was developed by engineers probably a lot smarter than you or I, has enough sensoring for what it does, and despite not being flawless, might be better than no autopilot.
This is a ridiculous response. The known fatality from Tesla's Autopilot was caused by an existing sensor problem that has no remedy, besides more sensors. The thing will essentially drive into overhanging white billboards and who knows what else. The fact that this isn't a wide disclosure is criminally reprehensible.
I was wondering if they were including all of those sensors in order to maximize how much data they gather during the trial, especially when they're trying to improve failure conditions. From that, they'll be able to optimize to just the sensors they find necessary based on their training data.
1) They're basically duplicating the early Google cars with mostly the same sensor suite bolted on top and integrated in the side panels. This is good because it means that unlike Tesla, they're planning for fully autonomous.
2) Since the human driver is still responsible for switching lanes, it looks like this is very early stage testing and data collection.
3) I think it's great that they're taking the approach of giving customers a view of what the car is doing via the back seat tablet, and even encouraging them to share. This is a great way to get people comfortable with the tech.
4) It still saddens me that everyone is keeping their autonomous data proprietary. How much faster could development go if Google let anyone have access to their million+ miles of driving sensor data instead of needing to collect it all from scratch?
>if Google let anyone have access to their million+ miles of driving sensor data instead of needing to collect it all from scratch?
You basically suggesting Google to give up their(only?) leverage. Open source is all good, but if they can't make money out of these projects, you will see less of moon shots.
Nothing stops them from actually selling their finished automation software.
This seems like the kind of domain where the government can be useful. Negotiate terms to buy the aggregate data off everyone experimenting with self-driving vehicles and make it all publicly available for them all to use. Its a safety issue more than anything else, because we have borderline genocide going on every year with tens of thousands of roadside deaths, making cars safer is not just some side project for profits sake alone.
> This seems like the kind of domain where the government can be useful. Negotiate terms to buy the aggregate data off everyone experimenting with self-driving vehicles and make it all publicly available for them all to us
Why should the tax-payers subsidize private actor's R&D? I'm sure Google is amenable to partnership/sharing their data with anyone for the right amount or under the right terms; free hand of the market and all - no need for government "meddling".
You could claim the data is a public good. If the data is useful for companies beyond Google or Uber, or even other industries, it is plausible that it confers a public benefit, a benefit that (at least theoretically) is not being accounted for when Uber and Google do their calculus about how much to invest in data collection. In that sense, government would be the perfect actor to support -- and perhaps even subsidize -- this data collection in exchange for having it shared with the broader community.
Because then you keep kicking the problem down the road. Groups with cash treasure troves can reap data, and certain actors will never release the data, and in practice you won't see many organizations negotiating such data because it is valuable but its value is highly volatile based on what comes of the R&D. I cannot imagine almost any department willingly just selling it under current conditions, or else they may face the reality they effectively sold the necessary work to get the first fully automated system to market for what becomes pennies on what the reaper of that gold mine receives.
Because this data is a matter of survival, where you can easily forecast at least several million vehicular fatalities in the next decade worldwide, it becomes relevant that the various states (I'm extending this globally since several European car companies are also engaged in automation development) it is paramount it be made accessible in the same way we seem to be finally having the epiphany that should have happened at least two decades ago that scientific research should be made openly available.
> Groups with cash treasure troves can reap data, and certain actors will never release the data, and in practice you won't see many organizations negotiating such data
You make it sound like it is such a travesty or a great injustice, and I cannot for the life of me figure out why. Generating this data is not rocket science; it is fairly well-documented, the components are commodities and the process repeatable, one of the big challenges is that it is capital-intensive to scale. If the data is valuable as you postulate, raising the required capital should be easy, capitalism has been solving this problem for centuries.
> Because this data is a matter of survival, where you can easily forecast at least several million vehicular fatalities in the next decade worldwide,
No, it really isn't a matter of survival (at nation-state level you've elevated the discussion to). There are things that kill more people than vehicular accidents, so if your angle is "saving lives", then world governments really need to step-up their health game and meaningfully fight heart disease and cancer; hell, maybe they could even buy all the pharmaceutical research data and open that up to the public while there are at it (good luck getting a good price from GSK, Pfizer et al).
Just because Uber starting field trials with passengers first, doesn't mean they are ahead of the game. Google have been doing it for years and you know how good at they are at this. So, Google will not sell the data, rather they will improve their offerings and just partner with car companies.
My take on Google X projects is to do cool things that may find alternate revenue streams to selling ads (or just do cool things, without necessarily having a business plan up front). Sure Google makes most of their money from ads, but they are trying lots of different things to make money elsewhere. Self driving cars, Ara, Google Glass, and others weren't trying to push ads, they were trying to find new products to make money in some other way (maybe).
Google's push to expand their Cloud offering is also showing them trying to pursue non-ad based revenue.
Google knowing where you go could make their models of you more valuable. Now they can specifically target ads based on where you go and when. Sitting in the cab waiting to get where you are going what are you going to do? Pull out your smart phone and see an ad offering a sale at a competitor to where you told the cab to go.
I don't see the connection. The core code and data of ad-supported companies like Google and Facebook are closed, while the open source companies and projects like Red Hat and OpenStreeMap are not usually ad-supported.
I can eventually see regulations forcing companies to share particular types of data, such as that associated with accidents. Especially more unique accident scenarios. It'd be pretty limited in scope, but sharing that sort of data would have a pretty clear public safety rational even if they're just using it to build up simulations. No doubt there would be a lot of difficulty in just finding a way to make it portable.
I don't have the familiarity with what they're collecting (and how) to make any actual guesses on how you'd go about it.
Beyond that, no company is going to want to share their general data. There's too much time and expense behind its collection, and just sharing it wouldn't be enough; to make sense of it they'd have to share a lot of insights into their proprietary systems as well.
On the flip side, the different biases of different manufacturers might not go so well together. Toyota couldn't get the software for an accelerator pedal right, so I have zero confidence that the 2030 self-driving Prius won't just e.g. drive 15 mph under the speed limit everywhere, camp out in the passing lane, stop for at least 30 seconds when switching from reverse to forward, and gridlock in shopping center parking lots.
Then again, as someone who won't buy a self-driving car for as long as possible, it might be fun to learn the driving styles of different self-driving cars and use them to my advantage. E.g. If I know that tailgating a Prius which is going the speed limit in the left lane will cause it to move over in self-driving mode, or that a Mirai will always brake-check itself when it gets cut off, it might be possible to exert some control over them if it's advantageous in traffic to do so.
> Toyota couldn't get the software for an accelerator pedal right, so I have zero confidence that the 2030 self-driving Prius won't just e.g. drive 15 mph under the speed limit everywhere, camp out in the passing lane, stop for at least 30 seconds when switching from reverse to forward, and gridlock in shopping center parking lots.
All the things you describe are as likely to cause accidents as wrong turns or over-acceleration. If Toyota can't get this right (my bet is they will) then they won't and shouldn't be doing autonomous cars, despite their "change of heart".
Number 3 I totally agree. I don't agree with number 4 though. I think if you spend millions of dollars, time and energy to amass data then you shouldn't share it with your competition. You'd just make everyone's job easier.
1,300,000 people die each year in car wrecks and another 20,000,000 - 50,000,000 more are injured and/or disabled.
Google's motto is "Don't be evil" after all. Is valuing short term profits over millions of lives not "evil"?
>Don’t be evil. We believe strongly that in the long term, we will be better served — as shareholders and in all other ways — by a company that does good things for the world even if we forgo some short term gains.
Who's to say that opening up the data will save more lives than Google continuing on their current course?
If google opens the data (that I would argue is still incomplete), and irresponsible actors are given a means (via incomplete LIDAR maps, algorithms, etc) to mislead people already screaming TAKE MY MONEY, that would likely lead to a TERRIBLE image for self-driving cars if and when those race-to-the-bottom companies cause injuries and deaths.
I could foresee that those irresponsible actors could then scapegoat Google's data as the source of issues, whether or not that is true.
BUT I suppose injuries, deaths, and ensuing lawsuits are what prompts stricter licensing and oversight committees. I believe however that strict enforcement of self-driving cars will be an eventuality, no homebrew kits I'm afraid.
Killing a dog is a direct action. Yes it is wrong, at least unless there was a shortage of food.. then I guess it beats cannibalism.
Running a private experiment to gather data.. that data is private. Sharing that data is opt-in, not opt-out. Google has no obligation to release it. (If they do, cool. But not evil of them not to).
Now it is valid to say the government should run more studies on self driving cars and share that data. But I am not sure any business would exist in this country if every private R&D effort they did got open-sourced.
What if there was a middle ground where you could run your self-driving software in a Google simulation data set pulled from real world situations, so Google could essentially hide their data behind a testing black box? (Think codingame, etc). They could charge companies for the privilege of testing their software against that data.
It's interesting to consider Google's data from another perspective:
> “Uber logs as many miles in 24 minutes as Google’s autonomous cars have logged in their existence,” they wrote. “While none of these miles are fully autonomous today, we just point out the scale of experience that can accelerate the development of the AI, mapping and learning for autonomous cars.” [1]
>It's interesting to consider Google's data from another perspective:
>> “Uber logs as many miles in 24 minutes as Google’s autonomous cars have logged in their existence,”
Google's data is not limited to its autonomous driving project, despite what the cleverly-worded sentence implies. Off the top of my head: Google also has 3D LIDAR data from human-driven StreetView vehicles, GPS and accelerometer data from millions of mobile Waze and GMaps users.
The 24 min quote is strange, they only have 14 Ford Fusions equipped with sensors, so probably not more than Google, so I am not sure they are comparing LIDAR data. It seems this is about smartphone GPS timestamped coordinates of their taxis versus the google LIDAR data, which seems a strange comparison to me. Google also has a billion of unsuspecting people having their smartphone working for them, but the data quality is probably lower than Uber GPS coordinates.
c'mon, that the race for competition ! The great competition that drives the world forward ! I for one welcome my new competition overlord, blah blah blah
Based on the article, it sounds like the driver is only responsible for changing lanes when all normal lanes are blocked: i.e. a truck is blocking the only lane on a 2 lane road, and the driver has to cross over into oncoming traffic to get around it. While this is disappointing, it's very understandable that they haven't automated this process yet.
As the technology matures, all Tesla vehicles will have the hardware necessary to be fully self-driving with fail-operational capability, meaning that any given system in the car could break and your car will still drive itself safely. It is important to emphasize that refinement and validation of the software will take much longer than putting in place the cameras, radar, sonar and computing hardware.
Even once the software is highly refined and far better than the average human driver, there will still be a significant time gap, varying widely by jurisdiction, before true self-driving is approved by regulators. We expect that worldwide regulatory approval will require something on the order of 6 billion miles (10 billion km). Current fleet learning is happening at just over 3 million miles (5 million km) per day.
Here Elon mentions a milestone of 6 billion miles. At their current rate, Tesla will hit that in 5 years, but clearly we can expect the rate to increase rapidly.
What struck me as interesting about Tesla's approach is they are using a much cheaper sensor suite than Google/Uber. If they succeed it means they'll have developed algorithms (patents) to achieve a self-driving car on cheap hardware. Giving them 2 competitive advantages.
A million miles? That doesn't sound like a lot. Uber currently only has 14 self-driving cars but they are approaching 200,000 drivers. In 2-3 years, they could require a certain level of "automation" for all their drivers in order to gather data. They'll have hundreds of millions of miles.
How many Tesla cars have with their self-driving feature? 20,000 cars? They'll have 100,000 cars RSN.
By 2020, we will have 2 orders of magnitude more data.
> By 2020, we will have 2 orders of magnitude more data.
Sure, but so will Google. Their latest claim is > 1.5m autonomous miles and thanks to their Streetview project they have a ton more detailed mapping data than that.
Btw, I'm not saying that Google didn't do a great job, and make valuable progress. The game, however, is about to take a huge leap. Google was an early investor in Uber and many thought they were going to cooperate.
First, tesla collects zero lidar data.
So no, they don't, because they don't collect the same kind of sensor data.
The second assumption is that the sensor data google is collecting, that is similar to tesla's, can only be collected by google's self driving fleet ,and not any of the other ways google has to collect data.
Because, as above, it's not the same kind of sensor data, this is also false.
And Uber? i didn't make any claims about the current status, but a few years out. Tesla is using radar, instead of lidar, and cameras. Their goal is fully autonomous and they believe they can do it with their current set up. Their equipment could change, of course, but they will soon be shipping 100,000 cars a year.
While you're picking nits and calling things wrong, note that a lidar-equipped Tesla has been seen at Tesla HQ. So it's not zero, but that's OK, I won't refer to your statement as being "false", just over-confident.
One could say something like 'Reddit produces more words in 24 minutes than Shakespeare produced in his career', but there is a substantial qualitative difference between those two datasets.
This is the only way i can see Uber can survive in the future.
Its business model was sacrificing someone's income in the middle to be cheaper than ordinary taxis. But with self driving cars, they can minimize the costs without hurting anyone.
Really really smart and huge move at the same time. Whoever pushed this idea in the company deserves a huge credit.
Exactly. And that is something we human beings adapt to without noticing it. Telephone operators, newspaper boys, lumberjacks, meter readers have all disapeared. There were thousands of people who earned their life with it but eventually switched to new careers.
This will happen sooner or later and for sure there will be companies and individuals that will be affected by the change, but that is not enough reason to stop the development. And this is not Uber specific topic either. If not Uber, some other company will one day work on this.
What they are actually eliminating is work, which is very admirable. After that it becomes a question of income redistribution. Ideally the income should be only partly eliminated (in a way that balances welfare with motivation to find a new job). But that's a societal choice the US would have to make.
> But doesn't this business model entirely eliminate someone's income to be cheaper?
Think about this problem from Uber's perspective. More than two thirds of a fare go to the driver in most cases. Let's say eliminating the driver allows them to cut prices by 50% to account for purchasing a crapload of cars. Now imagine someone else builds self-driving technology (or licenses it) and copies the Uber business model.
A competitor building self-driving cars for-hire is instantly cheaper by a ridiculous margin. Uber would be overtaken by this competition in no time at all. This is unquestionably inevitable. Uber can't say "pay double (or more) so we can take the moral high ground and employ a bunch of fallible humans" because a.) humans are worse drivers than computers and b.) humans just cost more and nobody would pay to have a human driver.
Additionally, if Uber ever goes public, they're legally obligated to maximize return for investors. You can't keep human drivers around as a purely feel-good exercise.
> But doesn't this business model entirely eliminate someone's income to be cheaper?
A lot of business models do. Horse carriages eliminated income of (most of) manual porters, and cars eliminated income of (most of) horse drivers.
Unless you are a worshipper of Ned Ludd, that's how technologies work and will always work. Something new replaces something old, and people whose income depended on the old stuff have to look for different income.
Except now those people who would normally be drivers now get zero dollars from Uber. I do agree with you that this is the way for the future and I'm all behind it, however doing this does cut out jobs (eventually) for a lot of drivers. I don't think that's necessarily a bad thing, and I suspect there will always be at least some human drivers. I live in a small town and I really couldn't see them justifying the cost of this here. Where I live there's currently one 1 or 2 uber drivers anyway, and they aren't always active and I just have to call the regular taxi company.
It is better for those drivers to consider new jobs when there is a small push from a private company. Then there will be enough time for those drivers to adapt to a new environment. Think otherwise a government push for driverless cars. It would happen so quick that all those drivers would struggle until they find a new income.
The difference between the idea (fire the drivers) and the execution (develop decades' worth of cutting-edge AI which kills people if it fails) is huge, though. You don't expect your usual fancy-app startup to develop this kind of technology.
Drivers don't care! No one thinks they are going to be an Uber driver forever most think its a temporary thing.
I am pretty confident self driving cars are going to create more jobs than they destroy. Whole new businesses will be economically viable once the transportation costs are super low, traffic is less of a problem and when you don't need parking. On demand economy 2.0.
The introduction of self-driving cars would definitely displace a large number of jobs - especially truck drivers - but this is not exactly a new trend. Technological unemployment has been affecting the job distribution for hundreds of years, with inventions like the wheel, plow, tractor, weaving machine, etc having a huge impact. In the 1900s, 41% of the workforce was in agriculture, by the 30s it was nearly half that at 21.5%, and in 2000 it was 1.9% [1].
I'm definitely interested to see how this iteration plays out over the next few decades.
Using Pittsburgh as a proving ground is a smart move - not just because that happens to be where CMU is located but if you can train a self-driving car to navigate Pittsburgh, you've pretty much trained it to navigate anywhere.
In Pittsburgh you have a haphazard street grid with unconventional and often poorly maintained road conditions, hills with poor visibility, narrow streets, weird unpredictable driver behavior like making left turns in front of oncoming traffic as a light turns green (my friends call this the "pittsburgh left"), every sort of weather condition imaginable including heavy rain snow and ice, etc etc. A self driving car that can navigate Pittsburgh without ever needing driver input is truly autonomous
This is a promotional video produced by Uber. It clearly shows the vehicle illegally entering a crosswalk containing pedestrians. That's not a great start, and not particularly good advertising.
It looks like they had only just entered the crosswalk. Sure there's a "letter of the law/intent of the law" distinction here, but I don't see this as a particularly egregious action, especially when compared to normal human behavior (human drivers do this often, and with much narrower margins).
(If the software was designed not to do this but failed to behave appropriately, then the situation deserves addressing.)
Also note Google's cars are explicitly programmed to exceed the speed limit in order to track with a lead car in order to get better info on road conditions ahead. [1] Technically illegal, but this both increases safety and is well within the bounds of normal human driving behavior.
>It looks like they had only just entered the crosswalk.
To me it looks like they were fully inside the intersection before the car began the turn, so it was a violation (if very technical and unenforced).
>Sure there's a "letter of the law/intent of the law" distinction here, but I don't see this as a particularly egregious action, especially when compared to normal human behavior (human drivers do this often, and with much narrower margins).
Absolutely -- and as a pedestrian, I would prefer drivers punch through if they have sufficient clearance, rather than pointlessly sit there and watch me walk through.
But there's a difference between
a) "everyone has to do it even though it's not technically allowed", vs
b) "hey, let's showcase this as a prime example of our advanced technology in action"
I think it was wrong to do it as b), since it looks like some elderly people were cut off.
Nope, it's a violation. In a crosswalk, the pedestrians have the right-of-way, and motorists must wait until they have walked through the area the car wants to turn through to go. Yes, this means that in cities, you often can't turn right until right at the end of the signal, when you can pass behind the pedestrians completing their crossing, but there isn't enough time for any pedestrians to start crossing and claim a right-of-way.
I was always taught that from the moment the pedestrian steps off the curb, you cannot turn in front of them. They have the right-of-way.
I believe "irrational" is the word you're looking for.
If the intersection is wide enough that you can turn while maintaining a safe distance, then what purpose would such a law serve? Why increase congestion for little or no benefit?
Once, at 2:30 AM in Arizona, I executed a legal right-on-red via a rolling stop. The streets were empty, and I endangered no one. This is not a contested fact.
But an officer, unidentifiable in the dark some hundred feet behind me, cited me for "not stopping at a red light" due to the fact my vehicle did not come to a complete stop.
Arizona has a state law, named after an unfortunate little girl killed by an asshole who blew through a red light, which makes any "red light violation" an automatic 3 points on your license and applies some certain minimum fine. (8 points triggers a suspension.)
Now despite the fact that there is an entire world of difference between (A) running a red light into cross traffic and (B) performing a California stop at an empty intersection, when I showed up to contest the ticket, the judge mocked me for my suggestion that, as I had endangered no one and not been reckless, I had not violated the spirit of the law and that the ticket should be dismissed. I had to pay the fine and accept the points. State law.
"Irrational" may well be the word I should have used, but it changes nothing. What purpose, indeed, was served by such a strict interpretation of the law?
"And I say to any creature who may be listening, there can be no justice so long as laws are absolute. Even life itself is an exercise in exceptions."
-- Jean-Luc Picard
I agree, a law that serves no purpose shouldn't exist (the only purpose that one served was probably political). That's part of the reason why I find it hard to believe there are laws stating you must wait for pedestrians who are on the far side of the road. I searched but couldn't find them... and since such laws would do little or nothing to enhance safety while significantly increasing congestion (which is dangerous in itself), I think it's reasonable to conclude there are none.
People are probably interpreting the word 'yield' far too generously.
I saw a car in San Francisco on Monday that had an Uber label on it. It was an SUV looking thing with sensors and devices all over the outside of it. As others have said, it looked like the early Google self driving cars.
While that may not be using autonomous cars outside of Pittsburgh yet I wouldn't be surprised if they are driving some cars around other cities to digitally map them in preparation.
It's a bit surprising we don't have that. People customize ringtones, you'd think they want to pay to customize Siri voices and so on. Maybe the companies in the position to do this (Google/Apple) don't view it as good enough revenue compared to potential harm to brand (looking trashy?).
I meant different patterns for acceleration/braking. More aggressive v.s. less aggressive. If it's all within the realm of a safe ride, they could eventually optimize the AI driver better to the passenger.
In my case, I like having a more aggressive driver as long as they keep things safe. Others prefer "10 and 2 while keeping it exactly at 55 mph".
I would guess there's a range of accelerations and relative to posted speeds that would be equally safe. Optimizing for passenger preference within that range is what I am referring to.
And I'm saying that there are people not inside your car that cares about how you're driving. Most of Uber's rides are in densely populated areas with plenty of other drivers, cyclists and pedestrians.
It'll be for sure interesting to see how this plays out, especially with other human aggressive drivers. If people know they can always cut off the robot car which annoyingly drives the speed limit everywhere, takes turns super slowly, and optimizes itself to stay in the car pool lane, I'd hate to be riding in one. I'm not sure how many people currently mess with the Google cars but I'm sure it's more than zero. "Hey, wouldn't it be fun to see what happens when I just sorta...drift over the line into the stupid robot car's lane?"
There will be mistakes. People will die at the hands of self-driving cars. There will be backlash. There will be people who "hunt" autonomous vehicles (look at drones, and consider how much people value airspace vs driving etiquette). GLHF.
I do think it's cool that they are letting anybody ride in these. It will help make the transition easier for rolling out this feature if people have already ridden in the cars.
From what I've read, it is actually invite-only, loyal Uber customers for now. When Uber makes the transition in 30 years to fully autonomous vehicles, I think people will be feeling swindled that they've had to hear about it for so long while still having a driver the whole time.
Wonder how much of it will succeed in developing nations, where many roads are not well laid out and most importantly not well mannered.. On the flip side, the jobs of drivers are protected until the AI tech improves to cover complex scenarios or the infrastructure improves drastically, both of which can take decades IMO.
Labor is so cheap in developing nations that automated anything is rare. It was not uncommon for venues to over staff by 50%-200% even on slow nights becuase labor is so cheap.
Ever been to a club where security outnumbers customers 5 to 1 or waiters huddle in a corner on a slow night in big groups of 10?
The complex senarios and infrastructure won't take decades. Google's driving car has performed insanely well so far for millions of miles safely.
In Brazil I took some buses out in the countryside where they had an additional employee to collect the fare from passengers entering, sitting just behind the driver who did nothing while the bus was stopped.
A "bus conductor" was very common decades ago in wealthy countries, before being replaced with random ticket inspections or tickets being sold by the driver.
I've seen them in most developing countries I've visited. Even on ~15 seat minibuses, where as well as selling tickets the conductor can help with luggage, and keep a lookout for potential passengers.
>Wonder how much of it will succeed in developing nations, where many roads are not well laid out and most importantly not well mannered
Not to mention that the difference between cost of AI equipment (fixed) compared to driver salary (order of magnitude lower) is not even close to developed nations.
There's a cost you're missing in your equation. Automobile accidents are a major cause of death, particularly for individuals in their prime productive years (unlike, say, cigarettes which still typically wait until you're toward the end of your career to kill you). If automated cars can achieve even a fraction of the reduction in automobile deaths that they're predicted too, any developing nation would be stupid not to heavily subsidize widespread deployment of this tech.
Sorry but that's ridiculous, were talking about countries where a decent % of population in rural areas doesn't even have basic sanitary conditions and you are suggesting they should burn cash on AI cars ?
Generally, safety is prioritized less in developing nations. How many developing nations have building codes, workplace laws, environmental regulation comparable to America? And it's not just because we love litigation.
Most do. These days, most building codes are international and places like Turkey (where I lived up until very recently) now have earthquake codes on par with California due to recent experience with a devastating quake (https://en.wikipedia.org/wiki/1999_İzmit_earthquake). Granted, enforcement may not be on par with the US, but that's more a question of corruption as a general problem, not limited or specific to safety.
That said, Turkey has a stupidly high traffic accident death rate.
So, yes, as the 19th biggest economy in the world, with 70+ million people, I think heavily subsidizing AI cars would definitely be in the country's best interest (and well within its means...Turkey already has at least one world-famous robotics laboratory). I'd wager economies number 20-50 are probably in similar situations. Yes, truly impoverished nations and regions of nations likely have more important things to worry about, but I'd still wager at least ⅕ the world's population falls into the "can't afford AI cars individually, but would benefit from gov't subsidization of such".
We're drifting a bit, but perhaps our differences are due to the definition of a developing country.
Another commenter mentioned countries where a significant portion of its citizens don't have basic sanitary conditions, and that's what I'm picture. And in the case of Turkey, I wouldn't call it a developing country using this definition.
> I'd still wager at least ⅕ the world's population falls into the "can't afford AI cars individually, but would benefit from gov't subsidization of such"
> In the 2016 edition of its World Development Indicators, the World Bank made a decision that there is no longer distinguished between “developed” countries and “developing” in the presentation of its data. Nobody has ever agreed on a definition for these terms in the first place.
So much for the vagaries of human language! ¯\_(ツ)_/¯
Going back to Turkey, I really enjoyed my time there and it didn't fit into the developing country stereotype since the middle class seemed to be doing alright and there were plenty of outdoor leisure activities (my personal marker of how "developed" a country is). If the political situation was better, I wouldn't mind living there at all.
I disagree, with your wording at least. Safety is generally prioritized everywhere as a key goal.
It is just that if you live in rural India and are part of the 10% of the population that has no access to clean water Cholera is much more likely to kill you than an reckless driver.
Focusing on securing water in these regions is prioritizing safety. People are smart to focus on things more likely to kill them.
Yeah, I agree my original wording could have been better. However, I think both of us are in agreement that if you were that person in a developing country, you'd chance drinking potentially infected water over dehydrating. Just as one would work for pennies at a sketchy job site instead of not working and not being able to buy food at the end of the day.
And going back to the original discussion, automobile accidents are an abstraction, a probable cause of death, versus something more definite like starvation.
Pittsburgh's weather isn't exactly mild. And its roadways are profoundly complex, both due to very steep topography and due to being a city trying to keep up extensive infrastructure after losing half its population.
Who knows about visibility problems but new vehicles in the US are already required to hand control of traction to a computer, because the computer is better at it than at least the majority of humans.
I can't find good info on their ridiculous sensor suite.
I read in the spring their lone early prototype had 22 'coffee cup sized' cameras on the vehicle. By my count there are six 8 or 16 channel Lidar pucks positioned around the car not counting the big 64 channel spinning velodyne on top. There are at least 2 radar units, and likely also ultrasonic proximity sensors. Who knows what else, or what kind of compute they're packing in the trunk to manage all that data.
looking at the pictures of the cars, I'm starting to realize (fear) these images will be used for data collection as well! The cars are roving 360-degree cameras!
Not that I'm paranoid personally, but these will give the owners a lot of information about peoples' movement patterns, a competitive advantage, an asset. For example, they could tell how many people wait at bus stops and sell that info to municipalities... They could look for trends in fashion on the sidewalks and sell it to clothing companies...
Even if they don't achieve autonomy, just putting cameras any vehicle and re-selling the footage data-miners could be profitable.
I'm not looking forward to more business models built on the "let's collect data and sell it to people who hope to profit, probably through advertising" thing.
Putting together this story and the story about the self-driving car engineer nano degree from Udacity, I am envisioning a future where every self driving car has an accompanying engineer. Much as Steam engines needed an engineer to operate, self driving cars will need Engineers to be monitored.
Pretty much defeats the whole purpose for developing the self driving car, no? Why not just have the guy drive then, if you still have to pay him? And now that he needs a degree, he'll probably cost more than a driver. Not to mention, there's likely nothing for him to do once cars are autonomous.
I don't understand. Switching lanes is something that even AutoPilot can do. Is Uber deploying a lane-keeping system and calling it an autonomous vehicle?
I explained this above, it sounds like this is only an issue when the available lanes are blocked and they have to drive into oncoming traffic lanes, i.e. to get around an unloading truck or a taxi dropping someone off.
2 issues I have not seen addressed by this article or others:
1/ interference. Having one lidar-beaming car going down the road works. What happens when there are many? I suppose it just means more illumination, but could there be an interference or "blinding" situation? What about pedestrian/pet eye safety? I sincerely hope that all are being considered now that this self-driving race is on.
2/ if Uber skirts around taxi regulations by being "just" an app that connects independent drivers with riders, how do they justify researching, producing, and deploying these cars with uber employees aboard?
1. is a solved problem, at least at Project Chauffeur (Google's autonomous driving effort).
I live in Mountain View and see self driving cars every day, often driving next to or past each other.
2. Long term, it's not clear to me that taxi laws would apply to cars that drive themselves. Short term, re this pilot project w Uber employees in the driver seat: taxi laws are local. Maybe they got permission from the city.
Why can't Uber purchase a fleet of vehicles and let humans drive them? They can still have some auto-pilot functionality.
Let's say at time T=0, a few million Uber cars are parked or on the road. Users enter their travel plans into the app, and also indicate whether they are will to make an extra stop or more, whether they can drive a leg, or whether they desire a solo/express route. Then they either walk out to the parking lot and start driving, or wait to be picked up. So now, everyone in the Uber cars exists on this spectrum:
Express Passenger -> Driver for my leg -> Driver plus a little extra -> Driver plus a lot extra -> Full time Uber driver.
Let the market and scheduling algorithms work out getting people and cars to the right places are the right time. Now I don't need to own a car and we don't need to have highly functional self-driving vehicles.
You got pretty good point, wait a year from today, couple of those 5 Driver profiles you mentioned above will be available from Uber .
I will take this to next step, By JUNE 2018 when you have 200 miles(battery) Electric cars are on roads (from Nissan, Tesla), Uber offer above all the 5 Driver profiles . At the end of the day Cars will be parked in your drive way ( if you signup ) and you take care of charging overnight( after 11 PM to get electricity at 1/4 cost) in return Uber will give you UBER Milage credits which you can use whenever you /family rides.
These Uber milage credits are especially attractive to many casual drivers ( non Full time) and also good for UBER as they give only Credits(keeping 100% of fare cash) and it also put pressure on FULL time Drivers cut which is 80% today as more CASUAL driver pool is available in this new scheme .
Parking cars on People garage is economical compared on centralized location As it takes extra time to get cars back to neighborhoods where people need morning rides and also Uber saves on Real estate cost . UBER may have some centralized locations for Cars maintenance/repairs etc..
When 100% self-driving is ready by 2020/2021 ( both from Govt. regulation and technology side), All you have is Electric Self-driving Software Powered Cars (ESS Cars) parked all over in your neighborhood people Parking LOTS ( People get Mileage Credit for allowing their parking LOTS) and unlock and drive TO YOU from your mobile Phone .
- Personal vehicles cost roughly 60¢/mile to own if you have a car payment, according to AAA -- or 2¢/minute, obviously keeping in mind that you pay for every minute you own the car, regardless of whether you use it.
- Taxis in NYC and Chicago run about $2.00/mi, or 50¢/min if you're going nowhere in traffic.
- Uber depends on the city, but let's say $1.00/mi, or 20¢/min. Both Uber and traditional taxis have other fees they tack on, as a reminder.
So, the cost will be competitive initially: likely no higher than $1.00/mi or 30¢/min
with it getting as cheap as someone wants it to be for the sake of driving adoption. When you hit critical mass with an autonomous fleet, you could likely see fees below 20¢/minute (minute will likely be the dominant metric) to run a profitable model as we know it today.
With Mass adoption of Electric Autonomous vehicles, if it is offered at 40 cents/mile that that is cheaper than OWNING your own Car
I live just a bit outside of pittsburgh, but I'm not in town often enough to need an uber. I'm still probably going to head down and try to get picked up by one, though.
3) There is still a human in it. Once the human is removed, I suspect the car will only work if it has a continuous connection to HQ, and that will give it a constant GPS signal.
Something tells me the people who live in places where Uber operates and possess the necessary skills to steal an autonomous car generally are not interested in stealing cars.
I'm wondering what the potential will be for increased crime/theft of the automated cars. I'm sure there will be a security fallthrough to track cars that have somehow been overridden off of autopilot.
I know someone who worked at the ATC over the summer, so I asked him. He said there is no sure fire way to get a self driving one. You basically have to get lucky with the time of day and your start and end points being within the zone they are testing. I'm pretty sure the zone is the downtown area and areas near it. But I don't have exact details.
From what I heard on NPR this morning, they selected a few thousand users to be in the pilot program and only those users are able to get the self driving. Im guessing they looked at their past route data and user rating to find users to include.
Too bad. Being in Columbus, 3 hours from Pittsburgh, I'd love to road trip there just to try a self-driving Uber. But not sure I would if you have no way of dependably hailing one.
Either it's the catalyst for basic income, or the catalyst for the demise of capitalism and the beginning of the erosion of the fundamental premise of skilled human time being a valuable and scare commodity eventually leading to many very negative scenarios and suffering around the world.
The pronouns he and him have been prescribed for use to describe a person of unknown gender for centuries. You have to willfully ignore the English language to find a misogynistic conspiracy in pronouns.
The big rotating Velodyne thing has got to go. But so far, nobody who claims to have a suitable solid-state 3D LIDAR is actually shipping. Quanergy claimed a solid-state LIDAR suitable for automotive use [1][2] and showed it at CES, but never shipped. Their web site hasn't been updated since 2015. They were aiming at a price point around $100.
Advanced Scientific Concepts has a suitable LIDAR that costs far too much. Space-X uses an ASC unit on the Dragon capsule for docking. ASC announced they were getting into automotive in 2012 but never shipped.[3]
Somebody is going to do this soon. The technology works. It's a cost problem. Once the cost comes down, cars can have multiple LIDARs (at least fore and aft) near the top of the windshield and rear window. This gets rid of all that topside gear.
[1] http://www.quanergy.com/products/ [2] http://on-demand.gputechconf.com/gtc/2016/presentation/s6726... [3] http://www.advancedscientificconcepts.com/applications/autom...