Self-driving vehicles

[Norman Bates]

Originally Posted by AndyMEng

I look forward to not having stop and go traffic on the Queensway, when all cars are autonomous. Also no traffic lights, and no stop signs. Also no back-ups at left hand turn lanes, no waiting for lights (because all cars will accelerate at the same time!)

All cars on the road will not be autonomous in your lifetime.

[lrt's friend]

Every road has a capacity. Autonomous cars may give us some time but that is all. Obviously, pedestrians and bicycles will not be autonomous so does this all work? Or will we be creating pedestrian and bicycle hell with the roads so crowded.

[Jim613]

Originally Posted by roger1818

Tesla driver killed in Autopilot crash had praised safety of system

What a nightmare. I wonder what he was thinking as the car was headed straight for the tractor-trailer

[phil235]

Originally Posted by lrt's friend

Every road has a capacity. Autonomous cars may give us some time but that is all. Obviously, pedestrians and bicycles will not be autonomous so does this all work? Or will we be creating pedestrian and bicycle hell with the roads so crowded.

Exactly. No traffic lights and stop signs? How exactly do pedestrians cross the road then?

[Norman Bates]

Originally Posted by Jim613

What a nightmare. I wonder what he was thinking as the car was headed straight for the tractor-trailer

I've read a fair bit about that collision and the current best thinking was that he saw the trailer and had placed all his confidence in the system to react to it. But when he realized it would not react it was too late for him to intervene.

As well, it has been reported that he was watching a Harry Potter movie at the time.

[AndyMEng]

Originally Posted by phil235

Exactly. No traffic lights and stop signs? How exactly do pedestrians cross the road then?

In the future, every autonomous car is covered in sensors, and each car communicates to its neighbours, acting as a complete system. As a pedestrian approaches an intersection (or a biker travels along the road) each and every car is monitoring the obstruction, relaying the information to every other car on the road, actively avoiding the obstruction. The entirety of the traffic actively responds to wherever the pedestrian/biker/construction/obstruction occurs. Including automatically re-routing traffic to other roads to avoid busy sections and optimize the system. So, rather than your suggestion of a nightmare for pedestrians, you could sit down and have a picnic in the middle of the street if you wanted, traffic moving at full speed around you, without a problem.

Each Tesla is currently actively recording reams and reams of data every day, relaying the data back to the engineers who are working extremely hard at getting this system running smoothly.

As for the poor guy who was cut in half, I'm sure he was wishing that he'd been paying attention, just as the car warns you to do every time you turn it on. Pay attention, keep hands on the wheel, autopilot is in beta mode, not fail-safe. Also, don't watch Harry Potter, of all things.

Disruptive technology is developed in an exponential way. We've already started along the path, it's only a matter of time.

[AndyMEng]

Originally Posted by Norman Bates

All cars on the road will not be autonomous in your lifetime.

Hey Norman, guess it depends how old you are

http://www.driverless-future.com/?page_id=384

http://www.businessinsider.com/repor...-2020-2015-5-6

[acottawa]

It's hard to tell how fast a technology will be adopted. Jet packs, flying cars, maglev, space tourism have yet to be adopted in any significant way despite predictions going back almost a century, whereas smartphones became ubiquitous in less than a decade.

Given that there are zero driverless cars on the road right now, I would find it hard to believe there will be millions by 2020. I suspect this Tesla accident will discourage the use of semi-autonomous vehicles (where you are theoretically supposed to be alert and able to take control at any time, but in practice Harry Potter will command more attention).

[Norman Bates]

Originally Posted by AndyMEng

Hey Norman, guess it depends how old you are

http://www.driverless-future.com/?page_id=384

http://www.businessinsider.com/repor...-2020-2015-5-6

Autonomous cars are on the road today.

I was responding to your use of "all cars" on the queensway. And I stand by my point that all cars will not be autonomous in your lifetime.

[phil235]

Originally Posted by AndyMEng

In the future, every autonomous car is covered in sensors, and each car communicates to its neighbours, acting as a complete system. As a pedestrian approaches an intersection (or a biker travels along the road) each and every car is monitoring the obstruction, relaying the information to every other car on the road, actively avoiding the obstruction. The entirety of the traffic actively responds to wherever the pedestrian/biker/construction/obstruction occurs. Including automatically re-routing traffic to other roads to avoid busy sections and optimize the system. So, rather than your suggestion of a nightmare for pedestrians, you could sit down and have a picnic in the middle of the street if you wanted, traffic moving at full speed around you, without a problem.

Disruptive technology is developed in an exponential way. We've already started along the path, it's only a matter of time.

I think a lot more work has to go into consideration of the pedestrian environment and how self-driving cars will impact that. Having a picnic with traffic moving around you at full speed is not particularly appealing. I think we are a long way from removing designated crossings from our residential streets.

This is more than disruptive technology. This would be a fundamental change to our urban environment. Not saying that it's not going to happen, but we do have a habit of rushing ahead with these changes before we understand the full impact. If you look at a model city of the future from the 50's, with its elevated expressways and segregated pedestrian walkways, you can see why it looked great to engineers of the time. But it wasn't.

[1overcosc]

Widespread availability of self-driving cars by 2020 is simply not going to happen. I doubt fully autonomous cars will even be available to the public by then.

Artificial Intelligence is a very finicky field; progress comes in fits and starts and many seemingly basic problems remain unsolved. I suspect it will be quite some time yet for self-driving cars to become universal.

Now, electric cars.. that's coming, and probably a lot sooner than people think. The technology is already viable, it's just a matter of scale and pricing. Electric cars will be like smartphones. For quite some time they'll be around, but only as a niche product, then all of a sudden everyone will have one. I suspect the breakthrough is no more than 10 years off.

[roger1818]

Originally Posted by Norman Bates

I was responding to your use of "all cars" on the queensway. And I stand by my point that all cars will not be autonomous in your lifetime.

Agreed. Unless human driving is banned, there will always be people wanting to drive, be it just for the joy of driving or wanting to drive their classic cars.

The real question is how many people will want this technology. Some great ideas flopped while others succeed. Marketing plays a huge role, making people believe they want it, but there are other untangibles that also play a role. We have had hybrid vehicles for well over a decade now, and still a small minority of vehicles on the road are hybrids. The average person just hasn't embraced the technology and seen the value for the money, and there aren't any trust issues with it.

If it succeeds, there are several factors that will determine how long it will take for most cars on the road to be drive automatically. First, how long does it take for most cars sold to have the feature and second, once they are, how long does it take for the old cars without to be retired. The former is hard to judge, but here in Ottawa, where salt use is high, the latter is probably quite short (in the 10 year time frame), but in warmer climates it is probably more like 15, 20 or even more years.

Unlike TVs, where many people threw away perfectly good CRT TVs to get a new flat screen HDTV because the cost was relatively low, cars are significantly more expensive and most people cant afford to just throw away a perfectly good one because it lacks a certain feature. Some will sell and buy new, but the old one is still on the road, just with a different owner. Only when the value gets so low that it isn't worth repairing does it get retired.

[Norman Bates]

Originally Posted by 1overcosc

Widespread availability of self-driving cars by 2020 is simply not going to happen. I doubt fully autonomous cars will even be available to the public by then.

Artificial Intelligence is a very finicky field; progress comes in fits and starts and many seemingly basic problems remain unsolved. I suspect it will be quite some time yet for self-driving cars to become universal.

Now, electric cars.. that's coming, and probably a lot sooner than people think. The technology is already viable, it's just a matter of scale and pricing. Electric cars will be like smartphones. For quite some time they'll be around, but only as a niche product, then all of a sudden everyone will have one. I suspect the breakthrough is no more than 10 years off.

I feel fairly certain that if your next car isn't electric, your next one after that will be.

As with manual transmissions there will always be hold outs. And things like pickup trucks, minivans and SUVs, which are legally classified as <light duty vehicles> as opposed to <passenger vehicles> will be among those holdouts.

[1overcosc]

Originally Posted by Norman Bates

I feel fairly certain that if your next car isn't electric, your next one after that will be.

As with manual transmissions there will always be hold outs. And things like pickup trucks, minivans and SUVs, which are legally classified as <light duty vehicles> as opposed to <passenger vehicles> will be among those holdouts.

Agreed. Most trucks and vans will probably be combustion-engine for quite some time. Sedans, though, will probably be mostly electric soon enough. Shortly after electric sedans become widespread, automakers will probably stop making combustion-engine sedans, figuring that anyone wanting a gas engine will just buy an SUV or pickup.

I do suspect that large electric vehicles will become economically viable soon enough--there is a big market interest in electric buses--but it will take longer for pickup & SUV drivers to want to switch.

[acottawa]

Originally Posted by 1overcosc

Now, electric cars.. that's coming, and probably a lot sooner than people think. The technology is already viable, it's just a matter of scale and pricing. Electric cars will be like smartphones. For quite some time they'll be around, but only as a niche product, then all of a sudden everyone will have one. I suspect the breakthrough is no more than 10 years off.

The technology is viable (assuming cost is no object) in very limited applications (urban areas, limited use) but they make no sense for people who want to drive long distances, people who drive a lot (uber, contractors) or people who drive beyond their supply lines. I can't see it being more than a niche market for a long time unless there is a major breakthrough in battery technology.

[roger1818]

Originally Posted by Norman Bates

And things like pickup trucks, minivans and SUVs, which are legally classified as <light duty vehicles> as opposed to <passenger vehicles> will be among those holdouts.

Why would pickup trucks be holdouts? One of the key things trucks need is torque and electric motors can provide maximum torque at ALL speeds, something gasoline engines just ca not do, which is why cars with gasoline (or diesel) engines need transmissions. There is good reason why trains use electric motors for traction (with diesel generators). A long train would never be able to start moving if engine was directly coupled to the wheels.

[Norman Bates]

Originally Posted by roger1818

Why would pickup trucks be holdouts? One of the key things trucks need is torque and electric motors can provide maximum torque at ALL speeds, something gasoline engines just ca not do, which is why cars with gasoline (or diesel) engines need transmissions. There is good reason why trains use electric motors for traction (with diesel generators). A long train would never be able to start moving if engine was directly coupled to the wheels.

Well you tell me.

With the exception of the <Strong Hybrid> technology developed jointly by GM and BMW, and applied in select few GM pickups and full-size SUVs from 2009-14, I can't think of any other use of electric power in those applications.

[lrt's friend]

Originally Posted by acottawa

The technology is viable (assuming cost is no object) in very limited applications (urban areas, limited use) but they make no sense for people who want to drive long distances, people who drive a lot (uber, contractors) or people who drive beyond their supply lines. I can't see it being more than a niche market for a long time unless there is a major breakthrough in battery technology.

I totally agree with you. Electric cars have been around since the early days of automobiles, yet their major weakness, battery storage has not been solved. These cars do not have the range and recharging is too slow, so it will remain a niche market. I cannot afford to have two vehicles, one for city use, and another to allow me to travel further. I think this applies to most people. And like most batteries, they have a finite life. Do these become throw away cars because of the high cost of replacing the batteries?

[1overcosc]

Originally Posted by acottawa

The technology is viable (assuming cost is no object) in very limited applications (urban areas, limited use) but they make no sense for people who want to drive long distances, people who drive a lot (uber, contractors) or people who drive beyond their supply lines. I can't see it being more than a niche market for a long time unless there is a major breakthrough in battery technology.

We've already hit 400km range and 30-60 minute recharge times; that alone is enough for at least 90% of the home market. Very few home users of cars regularly need to travel more than 400km in a day; and when they do, they're typically not doing it all at once, so there's plenty of opportunity for a half hour of down time in there. Even for a road trip to Toronto; just stop at a rest stop and grab a burger, not that big of a pain to have to kill 30 minutes.

Moreover, there's critical mass of improvement now. Billions of dollars in development and major upgrades well on their way in the pipeline; ranges of up to 1000km, improved battery compositions, substantially reduced capital costs, dramatically expanded network of fast charging stations (there will be hundreds of them in Ontario alone by the end of this decade).

The story of electric cars in 2016 is looking a lot like the story of smartphones in 2005 or desktop computers in 1985: a niche product that is rapidly destroying obstacles to widespread use. Back in 1985 only like 5% of households in North America had a desktop computer and pace of adoption was slow with many questioning why households would want to buy one... but anyone watching the trends thirty years ago could have seen it coming. There's a reason IBM invested billions in the development of the x86 PC in the early 1980s, just like how companies across the world are investing billions in electric cars right now.

[1overcosc]

Originally Posted by lrt's friend

And like most batteries, they have a finite life. Do these become throw away cars because of the high cost of replacing the batteries?

Replacing batteries is not cheap, but bear in mind that electric cars more than make up for that by requiring far less maintenance & servicing over the course of their life span. Electric cars don't need oil changes or engine fluids. Furthermore, you're not likely to run into nearly as many repair bills, as electric cars have far less than can go wrong with the engine, and the much smaller size of the engine allows for much more padding of the body meaning accidents will cost far less in body work costs. For these reasons, electric cars will also reduce potential insurance liabilities; most insurers already give pretty healthy sized discounts to Tesla drivers.

Over the 8 year lifespan of a battery pack I bet the savings on these things, on average, cancel out the cost of replacing the battery pack.

One of the largely as-of-yet undiscussed impacts of electric cars will be far less employment available for car mechanics.