[insanecoder]

I wont be buying one of these

[clbmw]

All these discussions about Tesla and the first death (RIP) make me wonder why they call it Autopilot in the first place, when they and everyone else want Tesla owners to just use it to assist regular driving i.e keep alert and hands on the wheel. Like calling a revolver, that can be loaded with blanks, a starting pistol. Unless it is exactly what it is don't call it that.

Ergo call it what it is: Pilot-Assist not Autopilot

[ec_E92]

No. No. No.

[ORIGIN M.]

To late.

All worth while tech, arrives before 2020.

[ORIGIN M.]

Originally Posted by insanecoder

I wont be buying one of these

As long as you are driving you are fine.

I'd rather have the car driving, then some moron sending out a Facebook posting.

[-c-]

Originally Posted by paul223

Only way to progress further is bring the new product to market. A company can not study or test product 100% under controlled environment. When new tech emerges, buyers become the test subjects. In car automation, consumers are going to die because they chose to be the first. Somebody has to be first to take lose or be a winner.

To a point, auto industry is a tad different than your common cell phone or teck toy, Things are pushed out sometimes before their ready because of dates the product board makes. It's not as cut and dry as you think.
I only say this because it happens all the time in my industry and it bothers me, they could wait and make it that much better but choose to release something sometimes even with known glitches.

We do testing on real roads with real traffic not a closed course. But still things happen you can't always plan for.

You see we drive this technology, we think it up and sell it to our company to have marketable competitiveness against our competition so when these things happen I feel somewhat responsible for this stuff because we're the people who dreamt it up in the first place. Or the suppliers come and try to sell it to us promising the world.

We follow the design from cradle to grave and push for the latest in wiz bang stuff and love it when we sell it to the product board but then implementing it and making it safe and all that is another story.

Technology in the car should help you and keep you safe not kill you and no people should not die to further make a product better. In the end fault lies with the driver, he should have been paying attention.

[bmc335]

I'm all for a fully autonomous vehicle for the daily commute - or sending it to pick someone/something up for you.

But in reality, I don't think getting this type of technology to the point of say - the same level of risk you take when flying a commercial airline or better (because that's what people will expect even though they assume far more risk than that when the get behind the wheel today)- unless there are some major changes to our roadways and all other vehicles on those roadways so that these 'autonomous' vehicles can interact and take input in better ways. We simply just can't code for all scenario's, we would have to limit the scope by controlling the whole thing. I think at that point it might have a fighting chance. Otherwise, people will die while we try to make it as low risk as possible which will in turn scare people away from it...

But I've never looked deeply at how they make these work today and what the success/failure rate is - maybe its better than I think it is...

[Efthreeoh]

Originally Posted by -c-

To a point, auto industry is a tad different than your common cell phone or teck toy, Things are pushed out sometimes before their ready because of dates the product board makes. It's not as cut and dry as you think.
I only say this because it happens all the time in my industry and it bothers me, they could wait and make it that much better but choose to release something sometimes even with known glitches.

We do testing on real roads with real traffic not a closed course. But still things happen you can't always plan for.

You see we drive this technology, we think it up and sell it to our company to have marketable competitiveness against our competition so when these things happen I feel somewhat responsible for this stuff because we're the people who dreamt it up in the first place. Or the suppliers come and try to sell it to us promising the world.

We follow the design from cradle to grave and push for the latest in wiz bang stuff and love it when we sell it to the product board but then implementing it and making it safe and all that is another story.

Technology in the car should help you and keep you safe not kill you and no people should not die to further make a product better. In the end fault lies with the driver, he should have been paying attention.

Auto pilot .... "auto" - ergo the car automatically drives itself... meaning the driver doesn't have to pay attention because the car is driving itself; that's the whole point of autonomous driving isn't it?

So why would anyone on God's green earth put out a system that is called auto pilot and caveat its use by saying the driver must have his hands on the wheel an be fully alert at all times, because duh, that is what DRIVING is; paying attention, with both hands on the wheel, and a foot on the accelerator (or brake). If you have something better to do than driving to get some place, then do it; if not, then PARTICIPATE.



This shit is just funny.

[Efthreeoh]

Originally Posted by bmc335

I'm all for a fully autonomous vehicle for the daily commute - or sending it to pick someone/something up for you.

But in reality, I don't think getting this type of technology to the point of say - the same level of risk you take when flying a commercial airline or better (because that's what people will expect even though they assume far more risk than that when the get behind the wheel today)- unless there are some major changes to our roadways and all other vehicles on those roadways so that these 'autonomous' vehicles can interact and take input in better ways. We simply just can't code for all scenario's, we would have to limit the scope by controlling the whole thing. I think at that point it might have a fighting chance. Otherwise, people will die while we try to make it as low risk as possible which will in turn scare people away from it...

But I've never looked deeply at how they make these work today and what the success/failure rate is - maybe its better than I think it is...

It's not.

The term "auto pilot" comes from the aviation industry. Auto pilot works for commercial aircraft because commercial air transportation happens at very high altitudes with a low-dynamic traffic situation - i.e. little air traffic, 5-mile separation, and altitude separation. And the main point... air traffic is managed by humans and automation systems, so placement of each aircraft is monitored and controlled by a third (independent) party.

In the case of the Tesla incident, neither vehicle's placement was monitored nor controlled by a third, independent party. This is why autonomous automobile operation will never work without a vehicle traffic management system. I doubt such a system could be economically devised for ground automobile vehicular traffic. Even if it were able to happen then there is the issue of privacy rights...

[-c-]

Originally Posted by Efthreeoh

Auto pilot .... "auto" - ergo the car automatically drives itself... meaning the driver doesn't have to pay attention because the car is driving itself; that's the whole point of autonomous driving isn't it?

So why would anyone on God's green earth put out a system that is called auto pilot and caveat its use by saying the driver must have his hands on the wheel an be fully alert at all times, because duh, that is what DRIVING is; paying attention, with both hands on the wheel, and a foot on the accelerator (or brake). If you have something better to do than driving to get some place, then do it; if not, then PARTICIPATE.



This shit is just funny.

Oh I agree! Funny thing is even though tesla calls it auto pilot they say driver must keep hands in wheel and stay alert at all times in the instructions.


You reminded me of a lawsuit a while ago when winabego not sure of the spelling had nowhere in their user manual that you had to stay in your seat at all times when cruise control was engaged, so the driver got up to make himself a cup of coffee and the motor home drove off the road injuring him. Well he won and now in the user manuals it said driver must be seated at all times.

Not sure if this was a fact or fiction but when I think of it I can't help but laugh.

[Bbb34]

Originally Posted by -c-

....even though tesla calls it auto pilot they say driver must keep hands in wheel and stay alert at all times in the instructions......

I personally find this amusing.
So, you have to be alert, or, half alert, or..... how does this help you then?

Imagine 7 hour trip, or even commute to work, where you're not actually driving, but can't quite do other things. Instead you must be alert and keep your hands on the wheel. Doesn't sound like fun, or relaxing experience to me. This is worse than actually having to focus on driving, be aware of your surroundings and anticipate whatever is coming your way. And once this tech is ready and truly "fully autonomous", imagine pricing structure on "vehicle software packages" that are "safer, newer, have more up to date maps, security features"... etc.... can you spell "software license fee revenues" on the rise, for all vehicle manufacturers.

I get that this is cool tech and likely way things are going to be, but in my view, it'll take a while. It'll be nearly impossible to get the safety part "just right" for all scenarios (weather/other participants in traffic/pedestrians/bikes/road imperfections/software and hardware failures...etc..). Unless governments pass the laws where all cars on the road have to be fully autonomous it'll be incredibly hard to get this to work perfectly. Not to mention sorting the laws and insurance liabilities in all this. Passing laws like these is not exactly easy thing to do in democratic societies.

That said, I truly enjoy driving and will do so myself as long as I can, so I might be slightly biased on this particular topic.

[Viffermike]

Originally Posted by Efthreeoh

It's not.

The term "auto pilot" comes from the aviation industry. Auto pilot works for commercial aircraft because commercial air transportation happens at very high altitudes with a low-dynamic traffic situation - i.e. little air traffic, 5-mile separation, and altitude separation. And the main point... air traffic is managed by humans and automation systems, so placement of each aircraft is monitored and controlled by a third (independent) party.

In the case of the Tesla incident, neither vehicle's placement was monitored nor controlled by a third, independent party. This is why autonomous automobile operation will never work without a vehicle traffic management system. I doubt such a system could be economically devised for ground automobile vehicular traffic. Even if it were able to happen then there is the issue of privacy rights...

EXACTLY. And that's just one point of several against autonomous tech in U.S. private vehicles being a widespread reality within the next 20 years, much less five. Also consider:

1). The military still pilot drones; virtually none of them are fully automatic. Why? Because of governmental culpability for collateral damage if something goes wrong.
2). The aforementioned autopilot function in commercial aircraft -- already the most complex and systems-redundant vehicle on the planet short of a spacecraft. Yet pilots are still required to take off and land an airplane. Why? The same reasons as 1)., but the culpability falls on airlines and manufacturers instead of governments in most cases.
3). Insurance. Autonomous vehicles would shift the burden of insurance culpability -- for both property damage and bodily harm -- from operators to manufacturers. Tesla's Fatality #1, and the reaction to it, is direct evidence of this effect; most are blaming Tesla, not the operator, for the death. This is a paradigm shift in the industry that will take years upon years of political and institutional wrangling to achieve -- and for this reason, part of the purchase price of an autonomous vehicle could very well be an insurance 'premium' that will delay the technology being affordable for the masses. One thing can mitigate this: the government becoming involved, either by an aforementioned autonomous control grid or by subsidizing the insurance requirements -- and we all know how well the latter scenario flies in our country.

There are other reasons, but the above is enough. Self-driving cars will not happen in the U.S. for a long, long time.

[raysspl]

Well, at least this will be good for the elderly grandparents...

[afadeev]

Originally Posted by Bbb34

Imagine 7 hour trip, or even commute to work, where you're not actually driving, but can't quite do other things. Instead you must be alert and keep your hands on the wheel. Doesn't sound like fun, or relaxing experience to me. This is worse than actually having to focus on driving, be aware of your surroundings and anticipate whatever is coming your way.

Except that most people don't do the above. Or at least don't do it often enough.
Frankly, neither do I.

I had two accidents in my life, both due to distractions in heavy bumper-to-bumper traffic.

One time a dude was busy doing something other than paying attention, and did not stop when the rest of the traffic stopped in front of him. So he plowed into my trunk, and pushed me into two cars ahead of me. I think Tesla's autopilot would have avoided this one just fine (if he had it). But this happened long before the Tesla or Mobileye (eyes and brains behind much of the autopilot) came into being.

The second time I screwed up. Changed lanes when someone had zoomed in to fill in a spot that was empty just before I got distracted. Not sure if Tesla's autopilot would have saved me there, as I don't think it would have initiated lane change in that situation. Or may be it would have, since it would have forced me to sit patiently in the slow lane.

Long story short - there are situation where the current state of the art auto-pilot systems are both good enough, and useful. Repeated stop-go 1st gear traffic is probably one of those. I know I get bored and frustrated with that driving (try crossing Manhattan island twice a day!). I test drove Model S in just that environment, and loved it!

Hands off the wheel between stop lights loved it!

However, driving at highway speeds in low-contrast visual environments (Mobileye relies on visual recognition) - night, dask/dawn, rain, snow, into sunlight - those are probably still ill advised behaviors. I would probably add relying on auto-pilot to save your skin at highway speeds in general.

Driving while watching DVDs will, evidently, kill you, with or without an autopilot.
Shocking, I know.

I half expect to hear that the dude's family went out and filed lawsuit against Warner Bros (Harry Potter's studio).

a

[Bbb34]

Originally Posted by afadeev

Long story short - there are situation where the current state of the art auto-pilot systems are both good enough, and useful. Repeated stop-go 1st gear traffic is probably one of those.

Agree with pretty much everything you said in your post, just want to clarify my point.

I was thinking more along the lines that, sure I can understand the value of fully autonomous driving vehicle. It's not for me, I just love driving, no matter what (maybe I change as I age) but I get why some people are looking forward to it.

What I don't get is "semi" autonomous driving. You're neither "here nor there" you're not driving, but can't really do anything else and have to alert at all times. I get that this is where technology is right now and it might be perfect one day, but the state we are at now to me just doesn't seem beneficial to the owners.

[afadeev]

Originally Posted by Bbb34

What I don't get is "semi" autonomous driving. You're neither "here nor there" you're not driving, but can't really do anything else and have to alert at all times. I get that this is where technology is right now and it might be perfect one day, but the state we are at now to me just doesn't seem beneficial to the owners.

"Semi" part of "semi"-autonomous moniker is a fig leaf.

All it says is that auto-pilot "semi"-works under some circumstances.
When you recognizes the circumstances that allow it to work well, it's great, and is materially beneficial to the owners. You just need to recognize when the circumstances change and you need to disengage the semi-autonomous autopilot. And keep one hand on the wheel to provide token confirmation that you are doing the above.
Nothing more, nothing else.

The state of the art is pretty much lane departure control plus active cruise control (speed up) with active collision avoidance (slow down or stop). The above is what you can accomplish today with forward radars and forward facing optical sensor(-s). Add GPS to help figure out which way the road is turning ahead.

360-degree ultrasonic sensors allow you to detect short range obstacles (or lack thereof = parking spot availability to auto-park), and avoid collisions with cars that drift into the ~16" radius around the car. That's it.

Active maneuvering (car initiated lane changes to follow GPS directions, for example) will be the next steps towards near-autonomous driving. But that will require much more processing power, multi-mode radars for varying ranges, stereo infra-red cameras, etc, etc. Google is a big believer in laser range finders. It's more precise than optical cameras, but much much more expensive:
http://www.techinsider.io/difference...s-cars-2015-12

That's the difference between tracking straight and reacting to close-proximity incursions (with high, but not 100% accuracy), and proactive maneuvering! To me, the latter is the auto-pilot.

Anything short of that is "semi"-something.
Nevertheless, what Tesla does today is already good enough for making progress through the traffic jams in near perfect safety. Open road speeding in conditions that compromise effectiveness of its camera sensors - not so much.
Not a bad start at at all, but certainly, far from perfect.

a

[dzvero]

welcome to the Matrix !