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

On ‎5‎/‎26‎/‎2018 at 2:56 PM, Mallette said:

Story of the airlines illustrates your point. It takes millions and millions of miles to prove airline travel is safe

 

I just checked the calendar and I see we are now in the 21st century. We did not have the computing power then that we have now when airlines were a new thing. Today we can debug software that controls unmanned systems via simulation, a task that was impossible then. It makes more sense to test and validate different aspects of autonomous control on virtual test tracks that can generate different scenarios and faults over 50 million or more miles and determine if a particular system is safe or not. 

Link to comment
Share on other sites

On 5/30/2018 at 9:45 PM, Jeff Matthews said:

I disagree, Mike.  I see no evidence of reckless disregard of known risks.  Taking a known risk is not always reckless.  That's why I drive every day without being reckless.

 

There is a known risk that you could stroke-out while driving down a busy road and kill a bunch of people... yet, you still get behind the wheel.  That's not immoral according to my understanding of commonly-accepted values.

If you think this angle affects the point I'm trying to make, then I'd suggest I may not be communicating clearly, or maybe you're not understanding what I'm trying to get at.

Link to comment
Share on other sites

5 hours ago, Don Richard said:

Today we can debug software that controls unmanned systems via simulation, a task that was impossible then.

Given my three decades in programming design, I can say with certainly this is still impossible. Humans are a chaos model, like the weather. Supercomputers spend days grinding on a single scenario of chaos and still, at best, approximate an answer. The only people who can test these softwares are human. Millions of miles of data of subsystems are accruing constantly as so many autos have adaptive cruise, braking, lane control, etc. Progress is rapid, but it's utter fantasy to expect to go from 50k deaths a year to zero without going through the space in between. I've no idea why this is so hard to comprehend.

 

Dave 

Link to comment
Share on other sites

35 minutes ago, Mallette said:

Given my three decades in programming design, I can say with certainly this is still impossible.

 

Well, given my 4+ decades in automation design and testing  I can state without hesitation that this is actually how this has been, and is being done. Computer based sims have been a game changer in the automation industry, particularly with regard to products for sale to the general public. You seem to have no idea whatsoever on what it takes to idiot proof something like a Level 4 autonomous vehicle. Note that I am not referring to debugging software like Autocad, but rather debugging the entire system - sensors, AI, control software, scenarios, and final control elements - that it takes to create an self -driving road vehicle.

 

Use your credentials to get a free subscription to this engineering publication, and get a copy of the Nov/Dec 2015 issue which has quite a bit of information on the state of the art regarding autonomous vehicle design and challenges:

 

http://insideunmannedsystems.com/subscribe/

Link to comment
Share on other sites

  • 7 months later...

I have faith in these systems...BUT they are NOT ready for that yet and only a complete idiot would ignore the manufacturers warnings to that effect. Clearly, his survival suggest it's very close but not up to him to decide. On a closed course where only his life is at risk, who cares. On a public highway? WRONG!

 

I continue to be amazed and thrilled at the rapid development and falling costs of these systems. Readily available well within average means now and if I had children I would no more drive on a freeway without at least automatic collision avoidance that let them go without seatbelts. 

Dave

  • Like 1
Link to comment
Share on other sites

I like the Tesla S (not enough range on a single charge, and still not enough readily accessible charge stations) , and I especially love the idea of the sub 10'sec quartermile Roadster...... but if the company can't turn a profit, paying 6 figures for Elon's cars is like playing russian roulette.

 

 

My daughters' Outback has accident avoidance, which almost caused me to wreck the other day while trying to get out of trouble.

 

Link to comment
Share on other sites

2 hours ago, Gilbert said:

(not enough range on a single charge, and still not enough readily accessible charge stations)

Not sure about your neighborhood, but it would be pretty hard to be 300 miles from a charger anywhere in the country, and nowhere on an interstate. As to range, the S3 optional large capacity does 300 miles...close to many cars. 

 

On profitability, than massive network has been put in place in just about 4 years. Amazon was unprofitable for a decade, may still be. Point is a good business model will make it and the investors have hardly deserted Tesla. 

 

Dave

today-northamerica.jpg

Link to comment
Share on other sites

Actually, I've always felt it a good idea to own one electric and one gas. Clearly electric is superior in all ways, but until charging is faster and ubiquitous, it will be an issue unrelated to AV issues. Certainly the 20 minutes it takes a supercharger is an improvement and hardly a major inconvenience given they are usually at major, full service stops, but until issues such as DR mentions are addressed there will be shortcomings. Granted, battery tech has advanced much faster than I anticipated but still has a long way to go.

 

Didn't really look at DR's link, but assume there is something to it. Doesn't apply to me as I avoid cold climes entirely.

 

Dave

Link to comment
Share on other sites

As I'd not heard about the range drop, went ahead and followed the link. Seems it's a sample set of one, in extreme conditions most do not face, and might well have been from other issues. 

"According to YouTuber Moshe (The Electric Israeli), his Tesla Model 3 range was down some 42 percent during a recent holiday trip to New York City. Temperatures in the area plummeted to about 17 Farenheit (-8 Celsius), and the wind chill made it feel much colder."

 

Doesn't really meet the strictures of science, though it certainly passes the intuition test.

Dave

Link to comment
Share on other sites

  • 4 weeks later...

Okies, naysayers, here's something maybe you can find value in, a win-win situation. You save money as a taxpayer, and those of us  who want AV technology get it where we need it most. 

 

I'd suggest reading this brief Motor Trend article first. MT isn't exactly a mag devoted to fantasy.

 

So, point is a Californian could save over 70 billion in tax money plus, use it or not, see a decrease in density in the "do it yourself" lanes and AV lovers can zip along in their deathtraps at 120mph unimpeded. Of course, for the AV lover, there's also the fact that this would speed up the day of near perfect, airline like, safety given the millions of miles of data that would be available. Here in Texas, we could link all the major cities as well NS/EW effectively halving travel times for AV users and improving the situation for doubters. Longest would be El Paso to Texarkana, bit over 900 miles, at a worst case 8 billion or so at current costs. Even that could be reduced dramatically by, rather than building new ones simply adding an AV lane much like the walled HOV lanes that exist in the urban areas to existing Interstates.   

 

I love high speed rail, but it's too late. If built 20 years ago they'd have paid for themselves. However, AV technology is now too close for them to have a chance at profitability given the cost efficiencies of the above concept. 

 

An even less expensive solution that could greatly improve the current AV programming would be development of RF equipped reflectors of the type already found in Interstates that light up the center line and lane edges. This would eliminate the current problems AV sensors have with snow, heavy rain, and the like in knowing the precise heading of the vehicle. Even the most hardened Luddite would possibly be willing to pay a little extra for a warning of lane departure. 

 

There are immense benefits in these concepts for all of us, regardless of our position on technology. Plus, everything needed to implement is already available. Certainly Cadillac and Tesla technology is pretty much road ready for these "guideways" even with current programming. 

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...