[OutlawX3M]

Originally Posted by Alfisti

Everyone is underestimating the how far the technology still has to go because no government is going to introduce this until it is fool proof.

The last 5% of development will take longer than the first 95%.

We are 20 years away, minimum.

And we're not even considering how patents and rivalries will hinder the advances in the next stages of this technology progression. Many companies are sharing or opening up some patents in the hopes their version of the technology wins out.

At the current stage, these autonomous vehicles can handle mundane situations IMO. They can not handle panic stops + avoidance such as a pedestrian jumping into your lane, and other "quick thinking" situations where more than one input is required (braking and steering).

Real people have a hard time dealing with these as well and I've yet to find a computer or computer system that can outthink a human every day, all day (yeah, Watson can beat a chess champion but have you seen the size of that machine!?).

What we'll get is pre-programmed reactions so we're back to the single most problematic issue -- who is responsible when the program is inadequate? Who's accountable?

[Flying Ace]

IMO, the technology is 98% there. What we have yet to figure out is the legal liability side of things and also how to operationalize the concept.

Government needs to step in and provide guidance. Then the industry will still have to sort itself out with who is the lowest cost provider is.

Remember just because you invented it, doesn't mean you'll reap the profits from the system, history is filled with examples of this...just look at the Compact Cassette tape...Phillips invented it, but Sony benefited from it.

[Torgus]

I can't wait.

There should be less traffic which means commutes speed up.

I can do work, netflix, whatever on my ride into work, even nap.

I can go out to bars and actually stay for more than 1/2 time of a game without fear of a DUI.

It is going to be awesome.

[Alfisti]

Originally Posted by Flying Ace

IMO, the technology is 98% there.

It's not even CLOSE. All they can do is cruise on the freeway and even that is questionable at best. They are miles and miles away from city driving sans collisions and fatalities.

The article above is a bit concerning, he is saying he'd release it when the technology is at par with a human driver. No fucking thanks, you're just penalising the better half of the driving pool.

It has to be foolproof and that is going to take eons.

[fecurtis]

Originally Posted by Flying Ace

IMO, the technology is 98% there. What we have yet to figure out is the legal liability side of things and also how to operationalize the concept.

Government needs to step in and provide guidance. Then the industry will still have to sort itself out with who is the lowest cost provider is.

Remember just because you invented it, doesn't mean you'll reap the profits from the system, history is filled with examples of this...just look at the Compact Cassette tape...Phillips invented it, but Sony benefited from it.

It's my understanding that they've a long way to go. In inclement weather, unmarked roads, etc they're very, very unreliable.

[Torgus]

Originally Posted by Alfisti

It has to be foolproof and that is going to take eons.

and inexpensive.

I agree we are a long ways off but I expect to see it in my lifetime.

[sirdaft1]

Originally Posted by Outlaw M3

And we're not even considering how patents and rivalries will hinder the advances in the next stages of this technology progression. Many companies are sharing or opening up some patents in the hopes their version of the technology wins out.

At the current stage, these autonomous vehicles can handle mundane situations IMO. They can not handle panic stops + avoidance such as a pedestrian jumping into your lane, and other "quick thinking" situations where more than one input is required (braking and steering).

Real people have a hard time dealing with these as well and I've yet to find a computer or computer system that can outthink a human every day, all day (yeah, Watson can beat a chess champion but have you seen the size of that machine!?).

What we'll get is pre-programmed reactions so we're back to the single most problematic issue -- who is responsible when the program is inadequate? Who's accountable?

Another interesting thought... how are the software engineers supposed to program the logic into the AV's in the case of an inevitable crash? For instance, what should it be programmed to do when a semi-truck swerves into the AV's lane and the choice is either: head-on crash with the semi, swerve into oncoming traffic and high chance of head-on crash, or swerve to the right where there are people walking on the sidewalk? Do they place a "life value" on the individuals involved in the scenario and take this into account in the logic? (ie: mother pushing baby in stroller has higher life value than convicted felon driving in AV???)

It may sound like is a highly unlikely event, but it's worst case scenarios like these that someone will have to program the logic on. And the more you think about it, the more it gets really messy.

[Alfisti]

Originally Posted by Torgus

and inexpensive.

I agree we are a long ways off but I expect to see it in my lifetime.

I turn 40 soon and am not sure i will, 20 years, minimum but another 40 would not surprise me. They are miles away.

[Torgus]

Originally Posted by Alfisti

I turn 40 soon and am not sure i will, 20 years, minimum but another 40 would not surprise me. They are miles away.

I'm 32 I'm guessing around when I turn 70 they will exist. but yeah 30-40 years is my estimate. Hopefully sooner but I don't see it.

I think it will start as Highway only and when you get on/off a ramp you will then be back in control fro around town driving. I think highway is the quickest and easiest for them to impliment and it will get people used to driverless cars. Then eventually around down will go driverless as well.

You know who will not like this? Bodyshops and insurance companies. Insurance companies have the money and are going to fight this I guarantee.

Also, UPS Fedex etc will still need a person. They have to pick up and drop off and go up and down in buildings and all over. I just don't see a robot doing all of that at this point.

[fecurtis]

Originally Posted by Torgus

Also, UPS Fedex etc will still need a person. They have to pick up and drop off and go up and down in buildings and all over. I just don't see a robot doing all of that at this point.

They could use a system similar to this,

[sirdaft1]

Originally Posted by fecurtis

They could use a system similar to this,

Ha, I was just watching that episode last night. I see zero flaws with this style of delivery method!

[Maestro]

Originally Posted by Outlaw M3

For certain industries, it's a game changer. It's my opinion self-driving cars are the real reason behind Uber's enormous valuation. Think about it, their greatest "cost" or in their case, their greatest outflow of revenue is the drivers. If they can eliminate the drivers, think of what their margins will be!

Another area is trucking (big game for Volvo and MB), eliminate truckers and the cost savings that trickles through the economy will be great. UPS, DHL, FedEx will see huge profit gains too. Just think how much of the cost of goods we purchase have transportation fees baked in... it has to trickle down no?

People won't feel comfortable with self-driving taxis and trucks unless it's baked into passenger cars as well. It's a mental thing -- oh, my car has it and I sometimes use it so it's not bad.

Personally, I love driving yet there is ONE scenario I wouldn't mind engaging auto-pilot for and that's rush hour traffic. In that instance, I wouldn't mind letting the car do the bumper-to-bumper stuff and I could relax.

Now one thing to consider -- what if autopilot eliminated the pesky, irrational behaviors of other drivers? You know, the left-lane hugging, change lanes without signaling, right-on-red w/out full stop moves that have us all banging our heads on steering wheels?

That list could go on and on... but you get the idea

Bold text is the area I see it happening first, today in the US it is hard to get people to drive trucks the average age of a truck driver in the US is 55, and there is a move to have mandatory retirement for truckers. Today people do not want to drive a truck since it is hard work and people are lazy today.

I personal do not see self driving cars until all cars are self driving. Why, the human driver is still the unpredictable part of driving today. A self driving car may not cause an accident but do you want to be in a self driving car when another human causes an accident. Until they can prove to me the car's prime directive to save the human in the car over everything else I am not going to allow a software engineer somewhere to make a decision to kill the deer or try and avoid it and kill me. The only right answers is to kill the deer.

I personally think the best proving ground for a self driving car, will be NASCAR. If a self driving car and avoid accidents on the track and win the race then maybe it know how to drive and keep it passenger alive.

The other issue with self driving cars all of them only drive on roads which are almost perfect. They have not spent any time on roads in the north east or in bad weather like snow and rain.

[Maestro]

Originally Posted by David70

My father in law and I still take out his Model A's (roughly late 1920's), we are welcome to drive them up to the speed limit (reality is to their top speed), they have no seat belts, horrible brakes, horrible tires, no air bags, no crash protection, no traction control, etc. If someone pulls in front of us we will hit them and the odds of being killed are massively greater than with a current car, and the modern car would go around the car or stop. The safest thing for us and the people around us would be to ban these cars but it has never even been considered. If we run into someone because we can't get around them or stop but we weren't drunk, driving too fast, etc. we might get a ticket but can probably take out a different Model T tomorrow.

I still own a 53 F-100, which I had since I was 16, the reason I removed it from the road was the fact that all the other cars on the road could stop way faster than 4000 lb truck with drum manual brakes. The only positive part was knowing if I did hit someone they would be dead since the truck was all steel and their cars was plastic and light sheet metal and most likely I would survive with all the steel around me.

This is a project vehicle for me, I am going to replace all the mechanicals with modern day tech like big rotors and 6 pistons calipers, long with other newer stuff, it just will not have ABS since no aftermarket stuff exists. This is the only way this truck will get back on the road.

[Maestro]

Originally Posted by Flying Ace

autonomous cars are the future. It doesn't mean 100% of the locations will have autonomous cars...but especially in urban congested areas, it's the only option to have a car in a city.

Many people don't understand the concept on why we need it...here's a fantastic article on the chief google lead on their project and his vision (btw, he's a car guy, so he's not trying to get rid of cars):

http://www.roadandtrack.com/car-cult...ik-google-car/

He is like every tech person out there, making a car drive itself is a very interesting problem to solve. They solved some of the problem be not all of them. What most Tech people fail to understand is solving interesting problems does not mean that people will use or want to use the solution they came up to for an interesting problem. Ask Google about Google glasses, yeah neat technology, but it made people look like an idiot to others.

Keep in mind, getting people from one place to another has been solved a long time ago. Mass Transit and walking. I spent time in places with good mass transit systems and it works well. The only issue which does not work for people in the US you have to travel on the Mass Transits schedule not your own. But with Mass Transit you will get where you expect when you expect to and most likely always alive.

There is a large group of people who will not give up the wheel and it is for reason which has nothing to do with solving an interest technical problem.

[sirdaft1]

Originally Posted by Maestro

He is like every tech person out there, making a car drive itself is a very interesting problem to solve. They solved some of the problem be not all of them. What most Tech people fail to understand is solving interesting problems does not mean that people will use or want to use the solution they came up to for an interesting problem. Also Google about Google glasses, yeah neat technology, but it made people look like an idiot to others.

Keep in mind, getting people from one place to another has been solved a long time ago. Mass Transit and walking. I spent time in places with good mass transit systems and it works well. The only issue which does not work for people in the US you have to travel on the Mass Transits schedule not your own. But with Mass Transit you will get where you expect when you expect to and most likely always alive.

There is a large group of people who will not give up the wheel and it is for reason which has nothing to do with solving an interest technical problem.

Good points. +1 to you

[Maestro]

Originally Posted by sirdaft1

Another interesting thought... how are the software engineers supposed to program the logic into the AV's in the case of an inevitable crash? For instance, what should it be programmed to do when a semi-truck swerves into the AV's lane and the choice is either: head-on crash with the semi, swerve into oncoming traffic and high chance of head-on crash, or swerve to the right where there are people walking on the sidewalk? Do they place a "life value" on the individuals involved in the scenario and take this into account in the logic? (ie: mother pushing baby in stroller has higher life value than convicted felon driving in AV???)

It may sound like is a highly unlikely event, but it's worst case scenarios like these that someone will have to program the logic on. And the more you think about it, the more it gets really messy.

That is my exact point I was making above without getting into the details like you did. A human making a mistake and possibly hurting or killing someone in a car accident can be forgiven in most cases, a machine making the same mistake is not a forgivable action.

The programing has to follow the same logic the person would do if they were driving which is, most everyone will put higher value on their own life. If you are not the cause of an accident (meaning you did nothing wrong) most likely you going to do everything can to prevent your death even if it means hitting someone else who was in fact the one at fault and killing them.

What most people do not understand is the fact a self driving cars is operating under a set of predictable control systems with known feedback mechanism and sets of rules, and they just placed it in a chaotic system with no rules or predicable behaviors with humans and asked them both to interact with one another in a predictable manor.

[David70]

Originally Posted by sirdaft1

Another interesting thought... how are the software engineers supposed to program the logic into the AV's in the case of an inevitable crash? For instance, what should it be programmed to do when a semi-truck swerves into the AV's lane and the choice is either: head-on crash with the semi, swerve into oncoming traffic and high chance of head-on crash, or swerve to the right where there are people walking on the sidewalk? Do they place a "life value" on the individuals involved in the scenario and take this into account in the logic? (ie: mother pushing baby in stroller has higher life value than convicted felon driving in AV???)

It may sound like is a highly unlikely event, but it's worst case scenarios like these that someone will have to program the logic on. And the more you think about it, the more it gets really messy.

Good point on this one. Given an infinite number of slightly different choices (how much the car turns or brakes), all with the outcome of some type of crash which will more than likely result in an injury or possible death, somehow the programming has to make a choice on which to do and after the incident it will end up in court where we then decide if it made the best choice.

Maybe all the people in the car die because going off the road and hitting something out of the cars vision and it was possible hitting the oncoming car was better or the other driver would have reacted and there wouldn't have been an accident at all. We currently are ok with the driver doing his best and it not being the best choice but I don't see us giving systems and programming the same break. I also wouldn't want to be part of trying to program the system to make the best choice when all of the outcomes result in a crash and possible death.

Originally Posted by Maestro

Bold text is the area I see it happening first, today in the US it is hard to get people to drive trucks the average age of a truck driver in the US is 55, and there is a move to have mandatory retirement for truckers. Today people do not want to drive a truck since it is hard work and people are lazy today.

I personal do not see self driving cars until all cars are self driving. Why, the human driver is still the unpredictable part of driving today. A self driving car may not cause an accident but do you want to be in a self driving car when another human causes an accident. Until they can prove to me the car's prime directive to save the human in the car over everything else I am not going to allow a software engineer somewhere to make a decision to kill the deer or try and avoid it and kill me. The only right answers is to kill the deer.

I personally think the best proving ground for a self driving car, will be NASCAR. If a self driving car and avoid accidents on the track and win the race then maybe it know how to drive and keep it passenger alive.

The other issue with self driving cars all of them only drive on roads which are almost perfect. They have not spent any time on roads in the north east or in bad weather like snow and rain.

The part in bold means that there will never be autonomous cars and I don't see it. I think the amount the car drives will slowly increase and if up to me I would start with only limited access highways or at really low speeds like stop and go traffic. Limited access highway have the lowest number of options with no cross traffic, no oncoming cars, mostly well marked roads with decent lines and cutoff points.

I do agree on putting an autonomous car in each race (not all autonomous as this would be far too controlled). If they can't beat the best drivers in a very controlled environment (all cars going the same direction and all well prepared for an accident) then I don't see people having much faith in the system on the road where the possibilities are far greater.

[Alfisti]

Originally Posted by Maestro

Bold text is the area I see it happening first, today in the US it is hard to get people to drive trucks the average age of a truck driver in the US is 55

The number of trucks on the road has exploded in the past 15 years, this is the single biggest cause of traffic congestion. Over here, the ads for truckers are relentless and all the new hires seem to be fresh off the boat.

I mean good on them for not wanting to bleed the system and to work but we have had a few incidents, including one last month that cost 4 lives.

[Alfisti]

Guys i think we are on different pages. i do not see it as possible to integrate autonomous cars, it'd be impossible. You'd need to go cold turkey and switch over everyone at once.

[Maestro]

Originally Posted by David70

I do agree on putting an autonomous car in each race (not all autonomous as this would be far too controlled). If they can't beat the best drivers in a very controlled environment (all cars going the same direction and all well prepared for an accident) then I don't see people having much faith in the system on the road where the possibilities are far greater.

Actually racing is not a controlled environment, the only constant is the track, not the drivers and that is my point. The issue is not being able to follow a course, it is dealing with the unknown which is the human. In NASCAR the strategy or lack of one is always changing depending on who is in front, behind, or beside you at any given time. Also what you do in a NASCAR race to avoid an accident defines logic some times. Drivers will say you need to drive at the accident happening in front of you because by the time you get there the cars have moved, if you think you should drive around you may drive into where they are going to end up.

Driving in the snow with cars sliding all over the place, you do not hit your breaks because you will end up in the mess, and some time you need to head where the cars are now because those cars most time will end up sliding off the road. I save myself a few time doing this in the winter.

How does a computer learn this, and most of these programmers near dealt with real world driving experiences in every part of the world.

[Maestro]

Originally Posted by Alfisti

The number of trucks on the road has exploded in the past 15 years, this is the single biggest cause of traffic congestion. Over here, the ads for truckers are relentless and all the new hires seem to be fresh off the boat.

I mean good on them for not wanting to bleed the system and to work but we have had a few incidents, including one last month that cost 4 lives.

Yeah I driving in Toronto lately, I am up there every few months, the Traffic is horrible and Canada is big on bring in people from all over the world to do things most people do want to do. The trucker in Canada are some of the worst I have experienced. I am Pennsylvania, and we shit between the north east and the western US and lots of truck come through PA so our Roads are jammed with trucks but most are good driver, because they mostly older and know how to drive.