From Mumbai: Test these cars on Mumbai roads. All the scenarios will be covered within an hour. No need to wait for 5 years.

Honestly I trust self driving cars more than humans because there are so many reckless and horrible drivers. I do understand people disagreeing with me though.

They need to be exposed in india,delhi environment......if they worked there then they are safe in every scenario Thanks to all who liked🙂🙂

people in 1936: In 2005 we sure will have flying cars people in 2019: Why dont we have self driving cars

What I'd like to see in the short term are cars with an emergency self-driving system that can take over if a driver falls asleep at the wheel, for example, and can safely pull the car over to the side of the road and stop. I think that would be doable in a few years.

Instead of self-driving car technology, I wish the United States would have high-speed rail.

Everyone could have a self driving car tomorrow. It just involves getting rid of all normal cars and not letting anybody on or near the roads

Just imagine when every car accident leads to a $10M lawsuit against the vehicle OEM...

Brilliant! Instead of making a self-driving car, keep raising money and saying that you will.

Seems like a lot of tesla hate in this video

When they do a self-driving that can run in India, then I'll put faith in this technology.

The problem with self driving cars is those edge cases in which the intelligence of the computer is just too limited. For example, imagine a signalized intersections in which the traffic lights broke, and there is a cop managing the intersection with hand gestures. A computer just can’t handle such a scenario.

Once we reach Level 5, we still have a few more problems to solve. Having the car reliably figure out the maze of where downtown is a good place to park. Under what conditions might the car decide on its own to move and to find itself a better parking space? Perhaps some big truck needs more room to maneuver? Somebody blowing a horn? Some stranger acting a fool or molesting the car? Also, cars should have at least 3 modes: off, standby, and on. I once had a microwave oven turn on by itself, and since then I really do not trust a computer to be in charge of a high-powered appliance. By placing it on a surge suppressor power strip with a manual off switch, I can know that it can not possibly turn itself on while I am away. How do you know that one day your fully autonomous car will not lose its mind and just drive off by itself? It needs a standby mode so that it can be summoned from an app on your smart-phone. But in off mode, it should require physical presence to switch it into standby or on mode. A software glitch should not allow it to just drive off. Key control will also have to be smarter. Who exactly is allowed to ride in it, and who is or is not allowed to delegate or to invite friends to ride with them? A self-drive car could easily have a dozen or more authorized users. It probably should be smart enough to notice, Joe has not ridden in this car for the last 5 months. Do you want to revise his authorization status?

9:29: "Most consumers are never going to own..." Remember when IBM said ...a personal computer

I bought a Tesla knowing the self-driving promise is most likely years away.

The Uber fatality results have come out and the blame lies with everyone apparently. The woman was on meth and illegally crossing the high speed road. The Uber safety driver was streaming video entertainment and not watching the road. The Uber car safety system failed and the government itself has no regulations on self driving testing.

"Why don't we have self driving cars yet?" Me: have you been to Miami, bro?

Why is this employee at navigant research speaking? Navigant is a consulting company, I doubt they have authority to talk about self driving vehicles.

Recently went from a 2010 to a 2019. Love all the new bells and whistles. GPS on a built in dashboard display, instead of my phone clipped to the dashboard. Lane keeping assist that "sees" the lines on the road and gives a little nudge on the steering wheel if it detects you drifting. Cruise control that senses everything around you and adjusts it's self accordingly. And a fully hands off mode that, combined with cruise control, seems fine for steering through slow turns on an empty highway. Even slowing down when it detects another car a couple of hundred feet away. It's not going to flip a blinker on and change lanes, or stop at traffic lights, not that I tried it, but it seemed ok for long distance highway driving. I've only tested the hands off cruise control a couple of times, and only for a few minutes on the highway. Seemed to work well. Honestly though, I'm not comfortable giving up control of my vehicle to it's on board computer. I trust my self waaay more than a device thats thinking components are less complex than those of a worm. Lane keeping assist , I like. If it notices me drifting too far to either side of the road, it gives a small tug on the wheel and a dummy light on the console. Something I would like to see in the next generation is a device that automatically shuts my blinker off if it realizes its been on for 10 minutes while I'm driving down a straightaway looking like an idiot.

This stuff seemed so obvious even years ago.

Elon musk: LIDAR will be doomed

I made a living for 47 repairing high technology that failed to operate as designed , what do I think lots of people and property will be damaged.

Many live in big cities with much annoying traffic congestion, noise and bad air quality. Now, if one wants to take the train, one needs to walk half a km to the bus stop, then take a bus "on time" and then hope traffic is not jammed so one can take a train "on time" It's not comfortable, it gives stress, it's slow even if the train itself takes only 50 min because of all the delays and one better takes a good margin. Imagine a small and light (plastic? ) automatic electric vehicle can be ordered right before the house. These small vehicles could bring one exactly timed (normally) to public transport or even to a normal size share car (not automatic) outside the city waiting next to a big road. I can understand the present automatic car gets confused by all the other vehicles, pedestrians crossing, bikers falling etc. So we must make the situation much simpler and safer. These small vehicles should have their own physically separated lane but if needed bikes could drive there too and pedestrians might cross. In that case an immediate stop must be realizable and even with a collision, nobody should get hurt. Oh yes, and with luggage and a big family, we order three of them of course.

It's very interesting how different the perception of how advanced the technology is and how the timeline is going to look like is so different in the US. I'm studying computer science in Germany and wrote my bachelor's thesis 2 years ago about technology in autonomous driving and back then my supervisor who had a lot of contacts with German car manufacturers said there was some kind of consensus that HAD will definitely take quite some years. And then I heard these predictions by companies like Tesla that seemed completely misjudging. I can't imagine that their engineers were that illusive. I think it was just a publicity stunt or maybe the work of illusive marketing people

To make it work it will be necessary to have all vehicles on the road connected to a supercomputer control system. Then all pedestrians, horses and cyclists will have to be banned. It threatens personal liberty and freedom.

I'm just wondering how these cars will operate in places where aggressive driving is the norm. Just imagine these cars trying to get around in places in India or Vietnam or even a parts of Africa or south America

I'm a Domino's delivery driver that drives manual and drifts an RX8. The only car that'll be self-driving for me is my funeral cab.

My concerns with Self-Driving Vehicles is making sure the AI in the vehicle recognizes traffic lights, road signs, etc. Also, what if the sensors on the vehicle, fail, there has to be a feature where the person can manually drive the car. Aside from that, I think that Self-Driving vehicles are a great idea. Especially for those humans who just can't seem to put down that smartphone and just have to be texting and/or talking while they drive, despite knowing that it's dangerous.

No driverless cars can survive in India's traffic

Why not interviewing Elon Musk?

Reminds me of the guy who bought a new Winnebago. He decided to try out the cruise control that the salesman had demonstrated. He was in the back making coffee when the accident happened. The vehicle was a total loss. His lawyers successfully sued Winnebago because the user documents didn't state he couldn't go do something else!

Having been in the tech industry for a number of years, I'm very skeptical of completely autonomous vehicles. Developers have become so accustomed to releasing imperfect software and fixing bugs in new releases that I'm not sure that they can develop a flawless system that won't fail at many touchpoints. Look at how often the technology fails on high-end cars right now. I know numerous people who own Mercedes, Audi, and BMW cars who constantly have their cars in the shop for electrical malfunctions and technological flaws. I think the technology is useful in certain applications, but I'm not totally convinced that it can be widely used on the same roads as unpredictable regular drivers.

The way this is framed is very reminiscent of FUD: So much fear of misuse of self-driving features, emphasis on uncertainty of when FSD will be achieved, and doubt about the readiness of the technology. Not really a balanced presentation, nor useful information for investors, consumers, or decision makers.

I can't wait till we have self-driving cars.... Cause when I'm driving.....I have to occasionally look up from my wonderful, precious phone and I'll NEVER get that loss of quality phone time back...!!!!! With a self-driving car, I'll be able to look at my phone.....24/7...!!!!!!

I’ve been part of the driving environment in the San Francisco area for over 5 years. I believe we are 5 years or less away from moving 51% of the consumes in the San Francisco area. My recommendation to Uber/ Lyft Drivers is that you should start re-educating yourself. You can get a job driving with Zum, Waymo, Cruise or Zoox starting at $26 an hour plus benefits. The RideShare industry is about to be disrupted again.

God forbid God forbid !!! Why in the hell do need we driverless cars and trucks and who the hell are these people who are pushing the absolutely insane idea? Where the hell are these people from a different planet? Why do we need to have them and why is it going to be forced on us. I personally don't ever want to share the road with a driverless car and especially a 80,000 lb driverless truck. I will never stop driving my own car and no one is ever going to tell me I will have to. The technology will never be 100 percent and worse yet it will always be a threat by hackers. Take your driverless crap and shove it up your you know where!!!

Self Driving Cars!?? It's 2021 and they still haven't made good self driving Lawn Mowers! lol!!

CNBC: Why don’t we have self-driving cars yet? Tesla: Am I a joke to you? Also me: ty for everyone who like

I know I am not the only one who caught when the guy said "even though we can say that an autonomous vehicle is better than a human driver, it doesn't mean that the autonomous vehicle is better than a human driver PLUS all the advanced driver assist systems we have".

I’ve been wondering about this because I’ve been hearing that these will be here any day. Thank you for looking into it

I’ve been following the space since 2005. Unfortunately it seems like the “fun part” is over and reality is hitting most of these companies, which will inevitably consolidate. Also we must remember the goals of these cars - more safety, less traffic, less cost, less commute time. At least in the short term, not clear who is actually delivering this.

I'm still waiting for a development update for the flying car that folds into a briefcase, as shown by George Jetson in the opening credits of "The Jetsons".

I don’t want them to make it Illegal for people to drive cars. It would be nice for somewhere to drive a car for fun. Like racetracks or big spaces where you can drive around, not fully remove normal cars.

I review the crucial AI problems that need to be solved for full self-driving cars to become a reality. They are 1) intuitive physics (including causal reasoning,) 2) intuitive psychology, and 3) multi-step reasoning and planning. The article is here: www.bobblum.com/ESSAYS/COMPSCI/self-driving-cars.html The sensor suite needs to achieve pixel level RGB-depth (range) at least comparable to human binocular vision (stereopsis) including minimal degradation in bad weather. This will not be achieved by Tesla, Cruise, or Aurora (or anyone else) for several more years (such that the car can drive autonomously in heavy traffic and bad weather.) Nonetheless, I look forward to exciting progress and gradual widespread rollout of Level 4.

From my point of view, the full self driving cars will be available by the end of early 2020s.

And that is just for the states. Imagine one of those cars in a road in mexico..

8:57 I've taken a waymo to that same destination before!

How can a self driving car know to go extra slow if a child off the side of the road is playing with a ball? As a teen driver, I could have killed a child had I not been watching her and run off the road when she ran right in front of me at the very last second.

"Are we there yet?" "Nope"

Car:Where would you like to go? Me:Hell Car:On my way! Me:Wait.....

Business model is the biggest challenge, how to make money on self-driving cars? We need more use cases.

People don’t realize how far away we are from level 5 autonomy. Like 10+ years.

Would a driverless car,react,like a human,to an immediate change,or,would it just keep trying to go forwards?Would the route be unchanging?Would the car drive into a flood,or can it 'see'?

What's so cool about self driving cars? I want to drive my car myself

This was a really good video showing where we are at. In essence still developing and not ready... As one in the computer science industry, I wish CNBC would get a bit more into the "brains" or "AI" part of self driving cars, as there is widespread misunderstanding or gaps of knowledge of so called "AI" within and without this industry. The general public and lawmakers deserve to know the underlying fundamentals of this technology and the risks involved. The self driving car industry is keen on hiding, denying, or underplaying the underlying flaws of an AI based self driving car. I'm going to generalize and simplify things significantly to keep the focus on the impact. So underneath the buzzword Artificial Intelligence as it is currently, is simply a learning "program" or algorithm that makes its own decisions. It needs to make mistakes in order to learn and to make better decisions. This is implemented using a "training algorithm". The term for this learning technique is called reinforcement learning. You can learn more about reinforcement learning in detail online or in books; the idea is nothing new and has been around for at least a decade. So the way it works is much like how a child would learn through knocking over a vase that it is an outcome that is quite undesirable. The same type of process needs to happen for these "self driving cars". These cars need to hit things in order to get better at not hitting things. Which in the video they allude to through the use of simulation. Researchers developing these cars do have a degree of control to try to coax the AI into doing what they want, but in the end these machines are not pre-programmed to do what we want them to do. In fact these cars are independent decision makers. What everyone deserves to know is that these self driving cars are not yet prepared for every traffic situation possible in this world. Which is why companies are testing them on real roads, but sooner or later these "15 year old barely licensed" self driving cars are bound to cause accidents. And many many accidents need to happen whether in real life or not before these machines get really "good" at driving in our world.

When I first heard of autonomous vehicles being implemented into society a few years ago, I initially thought this type of technology was way ahead of its time before being trustworthy practical everyday vehicles.

Let's see autonomous racing cars on a track first. Not just educational but fun too.

People have this unconditional love for Tesla cars. They think Elon Musk can do no wrong. Period. That's fanatical.

As a software developer, I have no illusions about building complex systems that don't mess up in spite of our best efforts (Boeing 737 MAX). I'm certainly not surprised that the hype is dying somewhat. As far as driverless trucks and taxis are concerned, the drivers who are put out of work may not take it lying down. It would not be difficult for them to learn how to disable autonomous vehicles from the roadside by jamming the sensors. Then we will be looking at a brand new class of crime - interfering with the robots. In a world where machines replace people, this is predictable. In the science fiction fantasy world, Frank Herbert's "Dune" trilogy, includes a background story called the "Butlerian Jihad" - a rising up of the people against the machines. This is no longer science fiction, I think it is arguably inevitable.

Good to hear a bit of reason rather than just hype. Still not if but when. The big motor companies are in too deep to turn back now.

Yeah but electric vehicles are safer.. They get in much less accidents even when you count for the amount of them relative to traditional vehicles. If we don't want things on the road until they're definitely safe, we should be taking most people off the road.

This video should be in 2016.

The problem with self-driving cars is that the cameras only look directly in front of the car. I remember being in driver's ed at age 16, and we had to be concerned with the distance in front of the car as well as up to the sidewalk. We needed to have the awareness that somebody off the sidewalk might at any moment come in front of you. The problem with self-driving cars is that it wall only be aware of speed limit and just continue driving according to road signs only as well as the immediate path in front of the car. Self-driving cars need very wide lens cameras for sensing the path and the sides of the car. Long story short, a computer will never surpass the human brain in capabilities.

Self-driving is the latest gimmick from the car industry to increase profits and divert from the fact that their cars are designed to self-destruct; so you have to buy another one; the moment you're finished paying for them.

It will never happen without dedicated road systems. So it will never happen. Driver assistance, sure.

I don't even buy cars with OBD2. No one is gonna be controlling me.

This will put an end to the shenanigans that occur daily: 3 point turns, turning left from the right lane & vice versa, speeding, running red lights, running stop signs at pedestrian lights, disobeying traffic lane signals, failing to yield, driving the wrong way on one way street, random lane changes, random U turns, etc...

Yeah right... You left out the part that the Uber crash was the pedestrians fault. She was jaywalking at night on a dark road. I think that many people would have the same result on that situvation. On the pedestrians defence. Uber car noticed her on the road but ignored her as an "object", before the crash was immediate.

I’m fine with just really good assist like new Mercedes or Tesla. It’s so amazing on long drives

We also need to develop the roads to support the self-driving cars so that it’ll be more safer and more functional. The society can’t fully switch to driverless car unless we transform the roads.p

The one dynamic I've yet to see is the result weather has on self driving cars. I'm looking through a snow storm with heavy snow ruts on the roads right now, how can the lack of traction be simulated?

Why don't we make the streets smart? It would cost a lot, but it would make implementation of self driving cars a lot easier.

The more important question is who cares. I'm glad I won't be alive when the only car you can get you can't actually drive because it is 100% automated. What a sad day that will be when the last driving car is discontinued.

Why the push for driver-less cars???? Automate washing my car, and changing the oil. I like driving it.

"Fleet applications" is essentially a euphemism for larger organizations that can manage the liability aspect of the technology application. So much of the discussion about "autonomous" vehicles revolves around hardware, software, development and deployment. What we REALLY want to know is : who pays when an autonomous system inflicts harm -- autonomously...? Certainly not the owner, as they're not operating the vehicle. How are the manufacturers going to dodge this MAJOR hurdle...?

Because there are people like me who don't want to risk the car malfuncting & taking me to the middle of no where, crash, etc. It's just mentally safer to be in control.

I suggest everyone watch VICE's "future of work" after this

"Welcome to Johnny cab" What is the allure of a self driving car? Use the bus, or train or plane. What next, self feeding food stations?

Six years ago my supervisor was so excited to think that she will soon be chauffeured to and from work by an all-electric autonomous vehicle. All she knew is what she was told. I said, No. It ain't gonna happen before you retire. She will retire next year and she's furious that I predicted it. But it was OBVIOUS! Every day I drive to work, I see situations that a computer simply will not handle reliably every time. And sure enough, that's what is happening. This really is the "trough of disillusionment."

I see some people here saying that some people like to drive. True, especially fopr those raised in car culture. But not only are young folks today less interested in cars and driving; when others realize they can go on their digital devices while autopilot gets them where they're going, I expect many who suddenly DON'T want to drive.

The real question is why don’t we have flying cars yet??

Final test: Welcome to Saigon

The infrastructure has to be developed along side the autonomous vehicles as well. There are many areas where the roads change, where you have falling debris, wet leaves, and other unpredictable scenarios. The best place to implement this are in cities with a defined grid system, and highways that are well traveled devoid of the most extreme inclement weather. Also, you will have people will naturally fear giving up any control to autonomous system because that psychological transition will take a few generations. I'm optimistic we'll get there, but if your in your mid 30s as am I, we probably won't see this fully accepted and partially implemented when we are in our mid 50(s).

Processing power is almost there, but I think another 5 years the CPUs will be better equipped. The problem is other drivers and pedestrians. Self driving cars can drive just fine, but when you add other people that can't drive or walk in front of cars expecting them to stop, it throws the computer off. It can not predict people. The same thing happened when cars were invented. A lot of people died from getting hit by the first cars, because people were not use to the speed of cars and walked right in front of them. I think self driving cars should be used the same as cruise control is used today. Mostly used on Interstates and not in town.

Simple answer. Not.everyone wants everything done for them. People like to drive

steering the wheel, shifting the gear and paddling break, clutch and accelerator is exercise?

Back when I was young the goal was flying cars. Perfecting self-driving technology is a little bit easier than that.

People need to think about what they say about autonomous driving. It is not necessary that autonomous driving works 100% of the time for it to be a success. IMHO, if autonomous driving proves to be accurate 90% or more of the time regulators will not be able to resist the public demand that ALL cars have autonomous driving. Remember if 10% of all accidents involve cars with autonomous driving that does not mean that faulty autonomous driving was the CAUSE of these accidents. First, was the autonomous driving feature ENGAGED when the accident occurred? (Of course eventually the driver will NOT be given the option of disengaging autonomous driving.) Second, could the car with engaged autonomous driving have AVOIDED the accident? For example, if a driver who is texting rear-ends a Tesla stuck at a stoplight with autonomous driving that would NOT be the fault of the car with autonomous driving. Insurance companies and autonomous taxis will solve this problem. We will reach a point wherein insurance companies will charge such a higher premium for any car that does NOT have autonomous driving that every driver will be forced to either a get a car with autonomous driving or to use autonomous taxis. No one will be allowed to drive a car with our autonomous driving. Most importantly, very soon - within one to three years - the number of annual deaths and injuries from traffic accidents will be reduced to a tiny fraction of what those numbers are today due to autonomous driving.

"Past peak hype, in the trough of disillusionment" (;-) important comment: design for the worst - edge cases the toughest, most problematic. #VisionZero

CNBC: "Why no driverless car yet?" Also CNBC: "HEH Tesla trying driverless car but they sux and have 5 accidents out of hundreds of thousands of runs. Everyone point n laugh at dem!"

The reason we dont have it yet is that they havent figured out a way to keep people from suing the manufacturer when there is a wreck. Not if, but when. No matter how perfect the systems are there will be wrecks. If there is no one behind the wheel who do you blame? What would the purpose of insurance be? The manufacturers can not take the blame for every accident. Me, personally, I dont think self driving is any different than people driving. No one is paying attention on the road today anyway. So, whats the difference? Sensors fail. People are binge watching Netflix. Programs dont read inputs. People are drinking and driving. Mechanical failures. People swerve for something in the road. The only difference is that when a person wrecks there is someone to blame and with self driving............ no one is to blame.

the car industry never got to where it is today overnight neither did the airplane. why are people so impatient? autonomous cars only started a few years ago and they already want to see full autonomous cars? it will still take years and years of research and development before we see on the streets. and honestly, i would trust an autonomous car versus a human. car, when programmed to drive, will just itself. a human, when behind the wheel, can do a million things and can change their minds a million times. how many times have we seen, drivers driving on the left most side of the road will suddenly swerve to the right to exit the freeway? and that usually causes a crash because other drivers have no idea what that driver is thinking. on autonomous cars. when a car will exit the freeway, it will send a signal (talk) to surrounding cars that it intends to exit the freeway so other cars will give way. you would think that would happen with humans? when a car signals to exit, what will other drivers do, speed up so to avoid having the car exit in front of them. i've seen countless drivers speed up and try to overtake as much cars as they can, before they make their exit.

This wasn’t to hate on Tesla. It was them attempting to dismiss Andrew Yang.

4:42 - pretty sure this dude shorted Tesla.

That video at 8:04 says Nov 20 2019, but it looks like footage from the 90s.

There is a lot of miss understanding in the economics of what makes this a reality. If a car will drive to my house, I get in, I drive, then I get out and it parks that’s all I need, we are very close to this.

1:19 I thought majority of those cases aren't the self driving cars fault? Just dumb people doing dumb things near the streets, Jwalk, appearing from blind spot, the usual stuff not even human drivers can predict, and probably perform worst with higher mortality rate.

The interviews are too negative for a good report.

As for deployment, current autonomous systems development seems to assume operation in a 100% distributed environment, where each vehicle is an individual cell in a vast sea. Call me old-fashioned, but I suspect difficulties inherent in the solo flyer model will give rise to larger control management systems, probably in more densely populated urban and sub-urban spaces. Civic, county, state and federal orgs already have transportation departments, and any formal regulation will organically flow back through existing management systems. Think "virtual traffic signs and road markings" and you're heading in the right direction. Perhaps even a virtual space where the cars co-exist in a game-like environment. Given the economic aspects of deployment and liability, first deployment will almost surely be in the long haul sector, during low-density traffic periods. The first fully "autonomous" vehicles running legally will probably be trucks hauling late-night cargo between ex-urban hubs.

It is very very far away, driving is a very complex task. Mainly because of the unpredictability of road conditions, mechanical/electrical failures, weather conditions, other drivers, animals, and pedestrians on the road.