Self Driving Cars?

[Midpack]

Did you really read what I wrote about Waymo? Not enthusiastic enough for you?

Why is it that anytime that a progress is made, I should be expected to jump up/down saying that a truly autonomous car is imminent?

Well, it could be and well within current technology, but it is going to be expensive because of all that redundancy.

And when I said some people still would not mind paying for such an expensive car because it suits their needs and they can afford it, you act offended. I wasn't offended at the idea at all, which post #? IIRC - Several people said they would buy their own autonomous cars vs car/ride sharing without acknowledging how expensive fully autonomous cars might be, at least early on. Sure some people would be willing and able to pay for any premium to own their own, but IMO it's not likely to be the same cost as a car today (inflation adjusted). Someone floated $150-200K, might have been you? That might be closer to the choice buyers might actually face.

What's your problem?

No problem, it finally dawned on me what your MO is on this thread. I'm probably the last to recognize it.

 

[NW-Bound]

Some of the back-forth I see here in this thread is reminiscent of some discussions I see on the 'green/environmental' sites...

In the past, I have seen posters (non-engineer types), because they will be old and infirm soon, getting excited about the real self-driving cars coming soon and being only slightly more expensive than the normal cars.

A lot of them read these threads but not post. Hence, I try to tell people how these things are not getting there that fast. I cite my work in aerospace where we have major advantages, and it is still tough.

1) The aircraft environment is much simpler than that of a car. Nuf said.

2) People are willing to spend a great deal of money on a jet liner. For cars, they want a lot of features dirt cheap.

3) Airplane maintenance is rigidly controlled. Not so for cars.

4) Pilots are professionals. And they are trained for a particular aircraft. Cars are treated as commodity. Just hop on and drive off.

5) Pilots do not mess with the "stuff". Car owners fool around and try to cheat the system by disabling its safety features, as I saw on a car discussion forum.

I can think of some more, but those are on top of my head. And whenever I mention that, it is a compliment to auto engineers as I appreciate the difficulties of their job. And I was thought of as arrogant, no matter how often I repeated myself as a parrot.

 

[NW-Bound]

Some of the back-forth I see here in this thread is reminiscent of some discussions I see on the 'green/environmental' sites.

When an engineer/scientist comes along on one of those sites, and points out that some things being discussed are just impossible (you can't make a solar panel 10x more efficient than the commonly available ones today - that takes you past 100%), or that some things are unlikely, or will be prohibitively expensive, they are shouted down as being anti-environment, unable to 'think outside the box', luddites, have no vision, work for the oil companies, etc...

I have not been on any such site. Have no desire, and certainly not now.

 

[eroscott]

Eroscott, thanks for sharing your experience with the Tesla. It makes sense that most of the time, the car would work very well. It would not have its supporters otherwise.

You are welcome. I thought it can be useful to hear first hand experience. I've driven that same trip (less 100 miles) in our non-Tesla car and I can tell you I felt quite a bit more tired/drained manually driving where you need to stay much much more focused.

How about the rare instances where it has, shall we call, a hiccup? From what I have seen on the Web, it would result in a serious accident if the driver was not alert and reacted in time. You alluded to one instance yourself, in an earlier post. Of course these were always caused by some unusual environmental or road factors. Most accidents by humans are the same way.

Now, if that could be detected and the driver alerted before hand, that would be great. Of course the computer did not know, else it would not continue on a straight path to a collision, or veered off road as I saw in another instance. So, let me change "detected" to "defined".

Let me be 100% clear on my opinion. Tesla drivers that do NOT keep their hands on wheel are idiots. The car reminds you to do that every*time* you engage AutoPilot. If you trust it too much (hands off) it will crash. I think it definitely enhances my driving safety and gives me a more relaxed driving experience BUT I've had to take over for strange lane painting, construction areas, very close merging situations, and peeled off semi-truck tires in the middle of my lane! I think I've gotten 20+ Over-The-Air (OTA) updates to the software in the past year including the 2nd generation AutoPilot.~~ It is pretty solid now for normal usage. Still yesterday in some rural two-lane road driving I was going up over small hill and it moved out of the center lane close to the line ('drunk'). Just like a human it can[not] see through the hill so may have lost track of the projected path. It did correct itself but it was not smooth and my passengers noticed (~~ for a few days just after the 2nd gen software if you were on the highway in the rightmost lane and a car was on the shoulder, the Tesla would occasionally slow down from like 65 to 50 thinking the car was too close or may move into your lane. A software update fixed this while I was halfway through a roadtrip. I got an OTA update when I was at an AirBNB and then that problem disappeared).

So, some driver education about the limitation of the system would be helpful. But when I mentioned that, a poster got violently upset.

Even if driver education, or shall we say user familiarity, is "acceptable" how do we deal with that in a rental car, where anyone unfamiliar with the system can drive it? Shall the rental car company sit the customer down for a 1-hour training session?

In the current Tesla systems, I think driver education and training should be required. I got an overview of the controls when purchased it but I had already watched various videos. I taught my wife and two boys to use while I was in the passenger seat. I would not have felt comfortable just telling them how to engage it and sending them on their way by themselves. I'm a little paranoid though because it was an expensive purchase for me.

Of course did drivers need training to use cruise control, do they now for adaptive cruise control? If you FULLY UNDERSTAND that you need to keep your hands on the steering wheel AT ALL TIMEs *and* eyes on the road just like you were manually driving, THEN I think an experienced (3+ years) driver can pretty quickly use Tesla AutoPilot. After all, just like with cruise control, braking or turning the steering wheel OVERRIDEs it immediately. I think that analogy holds water.

 

[NW-Bound]

I wasn't offended at the idea at all, which post #? IIRC - Several people said they would buy their own autonomous cars vs car/ride sharing without acknowledging how expensive fully autonomous cars might be, at least early on. Sure some people would be willing and able to pay for any premium to own their own, but IMO it's not likely to be the same cost as a car today (inflation adjusted). Someone floated $150-200K, might have been you? That might be closer to the choice buyers might actually face.

It was on the post where you polled people's acceptance of the self-driving car. I pulled out the above number out of a hat, simply based on an uneducated wild guess.

Of course a lot would come down with volume production. I have very little experience in the area of commercial products. But assuming that readers indulge me, I will tell a little anecdote.

We had a problem with a particular component on an aircraft. The subcontractor had been making them for us for years, and the volume was low, perhaps in the order of 10,000's of units. Neither the subcontractor nor the aircraft maker could figure out what was wrong. It passed the individual unit production test, but failed when installed on an aircraft. And that was how I got involved, and got to visit the subcontractor, looked over their manufacturing process and reviewed test procedures, and test results, etc...

It is a long story, but the gist of the story is this: I was surprised to learn this unit cost $100K a piece. Holy cow! And a variant of this unit was made for other aircraft makers, so some parts were shared. Each unit was serialized, individually tested, and calibrated records meticulously kept. But still...

How could it cost that much? I was not privy to the cost breakdown (no need to know), nor had a real desire to, but still remember this.

These high-performance sensors that Waymo built, what kind of test do they require? I would think they have to be carefully tested one by one on an assembly line, and that will be automated. Most stuff in aerospace are all hand tested, unit by unit, so it is not the same thing.

So, I do not really know what a level-4 car would cost compared to a normal car. I have no idea, other than it had to be quite a bit more.

 

[growing_older]

Some of the back-forth I see here in this thread is reminiscent of some discussions I see on the 'green/environmental' sites.

When an engineer/scientist comes along on one of those sites, and points out that some things being discussed are just impossible (you can't make a solar panel 10x more efficient than the commonly available ones today - that takes you past 100%), or that some things are unlikely, or will be prohibitively expensive, they are shouted down as being anti-environment, unable to 'think outside the box', luddites, have no vision, work for the oil companies, etc.

Did you just say that anyone who doesn't agree with you is obviously wrong and anti-science? This does tend to be a polite and respectful site, but that statement is not very polite.

 

[growing_older]

Of course did drivers need training to use cruise control

It seems so obvious how to work cruise control, but then every so often you see stories like the guy who turned on cruise control in his RV, then left the driver seat to go make himself a sandwich. Any kind of mostly-automatic control that requires driver attention and interaction will be prone to being misunderstood and will require adequate training to use properly.

 

[NW-Bound]

In the current Tesla systems, I think driver education and training should be required. I got an overview of the controls when purchased it but I had already watched various videos. I taught my wife and two boys to use while I was in the passenger seat. I would not have felt comfortable just telling them how to engage it and sending them on their way by themselves. I'm a little paranoid though because it was an expensive purchase for me.

Of course did drivers need training to use cruise control, do they now for adaptive cruise control? If you FULLY UNDERSTAND that you need to keep your hands on the steering wheel AT ALL TIMEs *and* eyes on the road just like you were manually driving, THEN I think an experienced (3+ years) driver can pretty quickly use Tesla AutoPilot. After all, just like with cruise control, braking or turning the steering wheel OVERRIDEs it immediately. I think that analogy holds water.

I think the problem arises when people become overconfident in the system, more so than the car builders. The dilemma with this stuff is that the better it works, the more complacent users become. Waymo guy stressed this over and over, and did not trust anybody other than their own trained drivers.

 

[NW-Bound]

It seems so obvious how to work cruise control, but then every so often you see stories like the guy who turned on cruise control in his RV, then left the driver seat to go make himself a sandwich. Any kind of mostly-automatic control that requires driver attention and interaction will be prone to being misunderstood and will require adequate training to use properly.

I heard this story, but has it been confirmed?

Some guy paid $2 million for a Prevost, and thought to himself, surely if a $100K Tesla has this feature, I should have it in something costing 20x more?

 

[ERD50]

Did you just say that anyone who doesn't agree with you is obviously wrong and anti-science? This does tend to be a polite and respectful site, but that statement is not very polite.

No, not at all. I was only saying that when someone gets 'hooked' on a certain 'solution', that they sometimes can tend to sort of shut out any criticism/critique.

I used the word 'reminiscent ', it tends to reminds me of (dictionary definition) - I didn't say 'this is exactly like....'.

Your statement that I wasn't polite isn't very polite!


-ERD50

 

[Midpack]

It seems so obvious how to work cruise control, but then every so often you see stories like the guy who turned on cruise control in his RV, then left the driver seat to go make himself a sandwich. Any kind of mostly-automatic control that requires driver attention and interaction will be prone to being misunderstood and will require adequate training to use properly.

I heard this story, but has it been confirmed?

Some guys paid $2 million for a Prevost, and thought to himself, surely if a $100K Tesla has this feature, I should have it in something costing 20x more?

It's not true. Several sources beyond snopes. Ever ask yourself why the story is even still circulating after 30 years? And some people wonder why fake news proliferates.

Cruise Control as Auto Pilot

 

[ERD50]

eroscott, I also appreciate the first hand experience you shared. But some of it seems a little contrary or just confusing to me - can you explain a little more fully?

First you say:

... I've driven that same trip (less 100 miles) in our non-Tesla car and I can tell you I felt quite a bit more tired/drained manually driving where you need to stay much much more focused. ...

Then you also add:

Let me be 100% clear on my opinion. Tesla drivers that do NOT keep their hands on wheel are idiots.

... I've had to take over for strange lane painting, construction areas, very close merging situations, and peeled off semi-truck tires in the middle of my lane!

If you FULLY UNDERSTAND that you need to keep your hands on the steering wheel AT ALL TIMEs *and* eyes on the road just like you were manually driving, THEN I think an experienced (3+ years) driver can pretty quickly use Tesla AutoPilot. After all, just like with cruise control, braking or turning the steering wheel OVERRIDEs it immediately. I think that analogy holds water.

So what I'm having trouble grasping, is if you have to maintain this same level of alertness ("hands on the steering wheel AT ALL TIMEs *and* eyes on the road just like you were manually driving,"), how is the drive any less fatiguing? I think it was discussed earlier, if I'm constantly wondering if the system is going to need assistance, couldn't that be more fatiguing? It would seem to me you'd wait a little before deciding it wasn't acting properly, so then you have to react even faster, due to that lag. That just doesn't sound relaxing to me, but I haven't actually experienced it either.

I tried to make the connection to (regular) cruise control - and I do find that more relaxing. So why? OK, for one, I have confidence in it (it's not complex, I have lots of experience with it), so I can rely on it to keep me from inadvertently speeding, if my attention is on other things, sometimes I may not realize I've sped up by another 10 mph. It allows me to move my feet and leg around a little, which keeps me more relaxed, less fatigued. And defeating it is simple, at the most basic level, hitting the brake like I would do in a 'exciting' situation is the exact same thing I would do anyhow. No delay, no change in thinking. Seems different to me.

-ERD50

 

[NW-Bound]

Earlier, I said "Pilots do not mess with "stuff"". Just now recall an incidence where one airline pilot did mess with stuff, trying to prove his knowledge of aerodynamics to his copilot. If I recall correctly, he caused the aircraft to tumble and lost a few thousand feet before he recovered.

No death in that incidence as I recall, but his passengers were in for a roller coaster ride.

 

[JoeWras]

On the left, what currently exists: just the roads themselves. On the right, what Tesla is aiming to do: mapping out every lane.

The lanes I saw the other night in the construction zone were a mess. No mapping will keep up with it, especially considering the wind turned around a few signs.

Construction zones may have to have set up a "do not enter" zone for some of these vehicles if they are relying on maps.

I think ultimately the mapping and exact positioning will be the way to go. But there may have to be times of exclusion enforced.

 

[JoeWras]

It dawned on me why this engineer, Joe, is cautious and wants to push out the timeframe. Or, at least one big reason why I approach with caution.

I've done a lot of "demos" in my lifetime. Demonstrations of products are fun. But we cut a lot of corners. Basically, you get 10% done, and then there is 90% left to do later, that is never seen, to really make it product ready. Yes, I know autonomous cars are zipping around in real world traffic, etc. There is still a demo factor, however.

The other thing is cost. Cost will come down, but it usually takes years. LCD monitors and TVs are a good example.

And then there is just society. That's a wildcard.

 

[NW-Bound]

It dawned on me why this engineer, Joe, is cautious and wants to push out the timeframe. Or, at least one big reason why I approach with caution.

I've done a lot of "demos" in my lifetime. Demonstrations of products are fun. But we cut a lot of corners. Basically, you get 10% done, and then there is 90% left to do later, that is never seen, to really make it product ready. Yes, I know autonomous cars are zipping around in real world traffic, etc. There is still a demo factor, however.

The other thing is cost. Cost will come down, but it usually takes years. LCD monitors and TVs are a good example.

And then there is just society. That's a wildcard.

Demos are necessary to satisfy management and investors that there is progress. Not just that the remaining pieces may be a lot tougher than what you have done, but of all the trial runs you can show just the successful one.

Demos are still good in that it shows something is not science fiction, and within the realm of current technology. So, I like to watch demos too, but still remind myself that it is a demo.

In the case of self-driving cars, engineers know that it is not science fiction. Nothing violates any laws of physics here. So, the questions always have been: 1) How well, 2) How reliable, and 3) How much.

About costs coming down, it is hard to tell until you try to set up an assembly line. Engineering prototypes are often hand-tweaked (of course they have to be at first), and you hope not having to do that in production. Again, one does not know until he tries.

 

[NW-Bound]

... The other thing is cost. Cost will come down, but it usually takes years. LCD monitors and TVs are a good example...

Speaking of TVs, I am amazed how cheap they are. And yes, in a way they are really "cheap" as in not lasting that long.

I bought my first HDTV in 2003. Visitors ooh and aah'ed when they saw it, but now you would not think it is really HD when comparing to the "real" ones. But the doggone thing still works. My brother is now on his 4th TV. Each time I visit him, he has a new one. They don't last. He buys a new one about every 3 years, and it bothers him because he is frugal, like me.

 

[NW-Bound]

...So what I'm having trouble grasping, is if you have to maintain this same level of alertness ("hands on the steering wheel AT ALL TIMEs *and* eyes on the road just like you were manually driving,"), how is the drive any less fatiguing? I think it was discussed earlier, if I'm constantly wondering if the system is going to need assistance, couldn't that be more fatiguing? It would seem to me you'd wait a little before deciding it wasn't acting properly, so then you have to react even faster, due to that lag. That just doesn't sound relaxing to me, but I haven't actually experienced it either.

We share the same question in mind, and eroscott has not answered. Not being a Tesla owner, but just from reading his posts, I venture the following theory.

When he said repeatedly that the drive is now more relaxed, then it is true. Is he complacent and just lucky that nothing happened? No, he is attentive and from what he wrote, he did catch a few mishaps and correct in time.

So, I think what is going on is that by being attentive, a conscientious Tesla owner experiences the scenarios that may cause trouble, such as construction zones, mismarked lanes, etc... and gets prepared the next time he encounters the same, and it is no longer a big deal. That apparently works, although it seems risky to me. I guess I will always be afraid that I may encounter something new and do not correct in time.

Maybe the car departure from the lane is not so violent, and that it is not too different than when you get blown by a sudden gust of wind, and I have experienced that quite often. You just have to be driving and not sleeping. And that is what he wrote twice.

Then, why is it more relaxed? Apparently monitoring the car and only half-driving is a lot less intense than full-driving.

Should I go out to get one to try for myself?

I tried to make the connection to (regular) cruise control - and I do find that more relaxing. So why? OK, for one, I have confidence in it (it's not complex, I have lots of experience with it), so I can rely on it to keep me from inadvertently speeding, if my attention is on other things, sometimes I may not realize I've sped up by another 10 mph. It allows me to move my feet and leg around a little, which keeps me more relaxed, less fatigued. And defeating it is simple, at the most basic level, hitting the brake like I would do in a 'exciting' situation is the exact same thing I would do anyhow. No delay, no change in thinking. Seems different to me.

-ERD50

I have no problem with cruise control and use it all the time, particularly on my RV which is a beast to keep in the lane. Driving the RV is a full-time job without watching the speedometer all the time. I am not kidding.

I have confidence in the cruise control. Here's why.

1) Cruise control is simple, very little can go wrong.

2) The failure modes are benign. If it is stuck, it cannot maintain speed when I go up hill or down hill. No catastrophe here.

3) Can it go berserk? If it suddenly releases, well it does that when it cannot maintain speed up a steep hill, and that is no big deal (although by the time I stomp on the gas pedal, I have lost a couple more mph in speed, and have a hard time getting it back)

Can it suddenly stomp on the gas? Have never seen it, and if it does, gee, I stomp on my gas pedal all the time to get this RV going when I tow my car.

In cars, the cruise control actuator often has a limited authority, meaning it can only pull the cable so far. And then, it has a slow rate of movement by design.

In short, cruise controls are quite benign. A full authority speed control like in the Tesla is designed to be fast and has full control of a powerful motor, so of course will be more critical.

 

[Chuckanut]

According to an 'expert' in an article from Consumer Reports, getting to the point where self driving cards can handle 90-95% of road conditions and driving problems is just about done, It's that last 5-10% of oddball things that happen while driving that are the big problems.

Also, apparently they are working on networking the cars so if a car on a road a mile ahead of you spots a big pothole, you car can be instantly updated and avoid the pothole.

 

[Scout]

I love fast cars and I love to drive, but.....
I work in an ER and it is rare that I take care of a patient from a crash not caused by driver negligence at least per the accompanying cops (speeding, drinking, tailgating, texting, not minding road conditions, etc). Imagine if everyone drove sober and obeyed all laws. The roads would be boring for sure, but our loved ones would be safer I guess.

 

[NW-Bound]

Self-driving cars that are expensive will be useful to people who need them and can afford them. But to reduce accidents, you have to deny the masses the steering wheel. Then, it has to be inexpensive so that it becomes mandated, else the poor have no transportation. Reducing the cost is not easy to do. I do not really see how.

I follow traffic laws, and do not understand why people oppose speed cameras, and red-light cameras. I love to see a lot of them. Now these are not at all expensive.

 

[NW-Bound]

Speaking of speed cameras, drivers learn where they are, and only slow down when passing them. Hence, you only catch people who are not familiar with the road, and many of them are not as bad drivers as the ones you do not catch.

I read that in Europe, there are cameras that capture your license plate at two different points, say 5 or 10 miles apart, and compare the times you pass them. If your computed average speed is higher than the speed limit, bingo!

So, if you speed up at any time between the two points, you have to slow way down in order to bring your average speed down. Now that sucks for speeders.

 

[meierlde]

I love fast cars and I love to drive, but.....
I work in an ER and it is rare that I take care of a patient from a crash not caused by driver negligence at least per the accompanying cops (speeding, drinking, tailgating, texting, not minding road conditions, etc). Imagine if everyone drove sober and obeyed all laws. The roads would be boring for sure, but our loved ones would be safer I guess.

Just out of curiosity how many do you see where the occupant was belted in and in a modern vehicle with lots of airbags? Of course pedestrians, bicyclists and motocyclists loose in a collison with a car. Just like cars loose in a collision with a train.

 

[dixonge]

In short, cruise controls are quite benign. A full authority speed control like in the Tesla is designed to be fast and has full control of a powerful motor, so of course will be more critical.

And yet your cruise control will allow your RV to plow into any vehicle in the way without your intervention, either by changing lanes or braking. Modern systems automatically slow down or stop to avoid collisions. That seems like a good result.

 

[NW-Bound]

And yet your cruise control will allow your RV to plow into any vehicle in the way without your intervention, either by changing lanes or braking. Modern systems automatically slow down or stop to avoid collisions. That seems like a good result.

Throughout these kinds of threads, I have repeatedly said that I welcome an autopilot for my RV, as driving it is a real chore.

But it has to be proven safe, and is something that I can afford. I do not like to drive.

Driving the RV, I cannot take my eyes off the road at all. And I mean at all. The handling is bad, particularly when I tow a car. It is such that when I disconnect and drive the car, it feels SO DAMN GOOD.

Another funny thing is I go only 60mph with the RV, and am so used to that speed. When I switch to the car, I keep that speed and it takes me a while to get back to driving at 70.