Jump to content

Poll & Prediction: Autonomous Car Equipment at 5k by 2019


Mallette

Autonomous Vehicles: Good or Bad  

49 members have voted

  1. 1. Are autonomous vehicles a good witch, or a bad witch?

    • Good
      20
    • Bad
      28


Recommended Posts

3 hours ago, DrWho said:

Haha, it'd also be a four million dollar car if they did one.

 

Where did all those guys end up? I've only ever heard great things about their engineering teams.

 

They were bought and sold about 6 times and as far as I know not much left.

JJK

Link to comment
Share on other sites

3 hours ago, Mallette said:

Odd, why would it work so well on aircraft and so many other systems and not here.  But, no need to respond since your denial flies in the face of already established fact.  If you choose to disbelieve feel free.  Millions don't believe the climate is warming, but it doesn't change the observable fact that it is.  BTW, let's don't get off on that as it seems to lead to locked threads.

Automation in flight is a much much simpler task. Heck, I'm confident I could pull it off on my own. Much of the automation in factories and the like (including flight) is implemented with infrastructure outside of the "vehicle"....which is what I've argued for in regards to vehicles too. That's the safest way to do it.

Link to comment
Share on other sites

2 minutes ago, DrWho said:

Automation in flight is a much much simpler task.

Okies, Mike.  Given I've a rather different viewpoint on that but since it's already operational in both cases I see no point.  I CAN see that when folks are ready, and realize the benefits, of autopiloted planes we'll see that 2 minute distance shrink between takeoffs and landings at major airports by a factor of 5.  Save a LOAD of money on airports as well as keep things on time far better.  What about trains?  What the hell was that dead head engineer doing missing the speed limits on that inaugural run of a high speed route?  I'd hope you'd agree that was totally insane.  No matter what we think about AVs and airplanes, I'd hope we can agree that engineers on trains are downright major risks easily avoided.

 

Dave

Link to comment
Share on other sites

I've argued for years that trains should be automated. That's a closed system and very easy to automate. I actually think it's a crime that those systems aren't automated already, but the main reason it hasn't is money - they still need someone on the train to get out and manually flip switches way out in the middle of nowhere. Planes are effectively a closed system - the weather stuff poses some challenges, but we measure a lot of that outside of the aircraft. The physics to keep a plane in flight are more complicated, but the control systems for automation are straightforward.

 

I just don't don't have a context for how one could be fact driven if they think the automobile is easier to automate than an airplane. That's like claiming Bose is better than Klipsch....or the khorn is better than the Jubilee, hah! :P

Link to comment
Share on other sites

49 minutes ago, Mallette said:

I CAN see that when folks are ready, and realize the benefits, of autopiloted planes we'll see that 2 minute distance shrink between takeoffs and landings at major airports by a factor of

I wanted to comment on this separately.....

 

I'm fairly certain Wake Turbulence is the driving factor for aircraft separation. Turns out it's heavily studied by the FAA (I tried to look up a quick number, but it turns out to be incredibly complicated). Obviously there's a lot of pressure on the FAA to minimize safe distances for monetary reasons. The reports I've been reading seem to have very little to do with pilot skill, and more to do with the physics of turbulence and radar accuracy.

Link to comment
Share on other sites

4 minutes ago, vasubandu said:

No doubt.  They do have train tracks that will slow trains going to fast, but implementation is slow.  As you say, it is a closed system, and they can control more than just the train

Glad we can ALL agree on this.  Lionel has had sophisticated automated train layouts for years that run flawlessly.  Further, I am in rare violent agreement with Mike that failure to automate trains is downright criminal and a good lawyer could probably force the issue after the Washington derailment that was clearly an error no computer would make.  Heck, doesn't even need sensors to be programmed to know the limits.  Just code.  In this case, even I code write the code.

Dave

Link to comment
Share on other sites

1 hour ago, Mallette said:

Glad we can ALL agree on this.  Lionel has had sophisticated automated train layouts for years that run flawlessly.  Further, I am in rare violent agreement with Mike that failure to automate trains is downright criminal and a good lawyer could probably force the issue after the Washington derailment that was clearly an error no computer would make.  Heck, doesn't even need sensors to be programmed to know the limits.  Just code.  In this case, even I code write the code.

Dave

 

I had a model train layout and if it was automated I wouldn't have had any fun crashing stuff.

JJK

Link to comment
Share on other sites

2 minutes ago, vasubandu said:

They are, they are.

Good on them.  This is an issue that needs to be addressed.  Trains should be safer than planes.  That accident was negligence, and the worst culprit isn't the engineer who's confessed to missing the signals, it's the management that put him there in the first place.  

 

Dave

Link to comment
Share on other sites

Seems some expect perfection of something in the first editions.  I am a realist.  It's going to take millions of road miles to reach 99.999 perfection.  But, as even the Wired article says, this technology is already saving lives and will continue to do so.

Dave

Link to comment
Share on other sites

  • Moderators
On 1/26/2018 at 3:28 PM, DrWho said:

Automation in flight is a much much simpler task. Heck, I'm confident I could pull it off on my own. Much of the automation in factories and the like (including flight) is implemented with infrastructure outside of the "vehicle"....which is what I've argued for in regards to vehicles too. That's the safest way to do it.

No no no no no no no.

 

Flight is much, much more difficult.  The only thing that has been automated is engine and fuel management which takes a crew from 3 to 2.

 

There is no fail safe option like you have in a car (press red button and it safely pulls over to side of road).  

 

The autoland function isn't even automated, it has to be carefully set up by pilot and, as you point out, requires a sophisticated ILS system be installed at that particular airport.

 

A vehicle only has 3 main decision inputs power/acceleration, braking and steering. 

 

An airplane has roll, pitch, yaw, lift (angle of attack and flaps), drag and power (X2 engines).

 

The most sophisticated flight automation engineers in the world are at JPL, and they don't do any flights fully automated.   Lots and Lots of people doing lots and lots of things.

Link to comment
Share on other sites

2 hours ago, DizRotus said:

Teslas are not autonomous.

 

Explain "Autopilot".

 

If this is a semi-autonomous feature it certainly doesn't work very well. Full automation needs more than sensor inputs for safe vehicle control, and this level of fuzzy logic has not been completely developed yet, and likely will not be ready for another 20 years.

 

Meanwhile, those who wish to join in the experiment are free to do so. "Be the first on your block to be killed in a self-driving car". :emotion-45:

Link to comment
Share on other sites

  • Moderators
On 1/26/2018 at 10:14 PM, Mallette said:

Good on them.  This is an issue that needs to be addressed.  Trains should be safer than planes.  That accident was negligence, and the worst culprit isn't the engineer who's confessed to missing the signals, it's the management that put him there in the first place.  

 

Dave

There is no money for it.  The collision avoidance system for passenger trains was mandated by Congress but it keeps getting extended because there is no money for it.  

 

Safety has always come from two primary sources, tort litigation and legislation.  It rarely, if ever, comes from a company saying "I'm going to make a safer mousetrap even though it will cost more and we will sell less."

 

Profit motive and safety are diametrically opposed, that's just a fact of economics. That alphabet soup, FAA, EPA, OSHA, NIHTSA, etc. were all created to regulate various private sectors because a free market is incapable of regulating itself. 

 

The only reason any A/V manufacturer will produce a safe car is because they have to comply with regulations and because of product liability exposure.

 

 

 

 

Link to comment
Share on other sites

39 minutes ago, dwilawyer said:

The collision avoidance system for passenger trains was mandated by Congress but it keeps getting extended because there is no money for it.  

That's BS...there is ALWAYS money for such things.  And no one is allowed to bring up politics in my thread and when I say just what I did it's because ALL parties spend what they want regardless of the deficit...so this could be paid for easily.  I think the railroads would pay for this if someone sued them.  The likely reason they are not doing so is union pressure.  Remember, we had firemen on diesel engines for decades before the railroads could eliminate them.

 

Dave

Link to comment
Share on other sites

  • Moderators
3 hours ago, Mallette said:

That's BS...there is ALWAYS money for such things.  And no one is allowed to bring up politics in my thread and when I say just what I did it's because ALL parties spend what they want regardless of the deficit...so this could be paid for easily.  I think the railroads would pay for this if someone sued them.  The likely reason they are not doing so is union pressure.  Remember, we had firemen on diesel engines for decades before the railroads could eliminate them.

 

Dave

It has nothing to do with unions.

 

I posted on the Railway Safety Act of 2008 in this thread about 20 or 30 pages ago.  It required full compliance by 2015.  That was extended until the end of this year, and it will have to be extended again.

 

It is an unfunded mandate that carries with it the problems with all unfunded mandates.

 

There are major technological problems and compatibility problems, and Congress isn't giving enough money to the FRA to develop a uniform system where trains from different systems can talk to each other and how.

 

Some of the proposed systems EXCEED the cost savings from the avoidance of accidents.

Link to comment
Share on other sites

Not buying it.  I designed a railroad device to run ahead of trains and feedback information to controls 10 years ago or whenever that AMTRAK train went into the river in Alabama or near it.  It was certainly conceptional, but it was based on off the shelf tech and I know it was buildable.  It was to run far enough ahead of the train to allow the train a full stop before reaching it.  It would have closed gates, and sat in the intersection long enough to know it was clear.  It would sense bad track, bridges out, etc.  It would be cheaper than engineers, preventable property damage, and injury and death to humans.  It is a crime that trains are not safer than aircraft, and, while safe, they are nowhere close.  It may not be unions, but it is not tech or money.  If the could do no better than the Lionel automation...and they could do MUCH better than that...it would be an improvement.  

 

Dave

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...