• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Ed Self-Driving Cars: Pros, Cons, and Predictions

Evaluate Self-Driving Cars on a scale of 1-5 (1 = Terrible, 3 = Meh, 5 = Great)

  • 1

    Votes: 10 6.6%

  • 2

    Votes: 11 7.2%

  • 3

    Votes: 24 15.8%

  • 4

    Votes: 28 18.4%

  • 5

    Votes: 79 52.0%
  • Total voters
    152
  • Poll closed .
I have a problem with only one company owning the technology. It won't be affordable.
 
If nobody can afford it, nobody is going to make any money. If it's genuinely desirable and priced affordably, they'd clean up. Look at what Ford did. Make something cheap enough to create a mass market, and you're sorted.
 
Rolfe said:
Well then you need to study the results of the trials that have already been done and see what holes you can pick in them. However, I'm not sure what hole you think you'll find when it's a simple comparison between FSD never being activated and FSD being activated sometimes.

It's pretty much inevitable that some crashes, even fatalities, will occur with FSD that would (at least probably) not have occurred with a human driver. However, if these are offset by avoiding even more crashes and fatalities that are caused by a human driver, you have a hard sell to oppose this.
Click to expand...
People who activate FSD are probably more likely to have simple routes to navigate. Those who have more complex driving tasks do not activate FSD as frequently. I.e., the two groups have, on average, a different complexity of driving task.

Does Tesla FSD have an algorithm to solve the trolley car problem? Are Musk's social and political views encoded in it? Or does FSD just give up when it can't cope and defer all responsibility for the accident that happens a few seconds later to the idiot behind the wheel?

Here's an idea for a couple of comparatively simple pieces of technology that would significantly reduce the number of accidents: a breathalyser that prevents the vehicle moving if drugs or alcohol are detected and a device that automatically slows the car down if the driver is too close to the vehicle in front or has exceeded the speed limit for more than ~10 seconds.
 
Ivor the Engineer said:
...snip...

Here's an idea for a couple of comparatively simple pieces of technology that would significantly reduce the number of accidents: a breathalyser that prevents the vehicle moving if drugs or alcohol are detected and a device that automatically slows the car down if the driver is too close to the vehicle in front or has exceeded the speed limit for more than ~10 seconds.
Click to expand...
Interlocks exist, and the second exists as well.
 
Elaedith said:
You never saw one of those high-speed traffic jams?
Click to expand...
I assume that what Ivor is referring to is that situation on British motorways where the road is packed with a lot of traffic that is all moving quickly. You know, that incredibly dangerous situation where if someone has an accident, those behind can't see it until it's too late to avoid joining it and you end up with a 48-car pileup.
 
acbytesla said:
I have a problem with only one company owning the technology. It won't be affordable.
Click to expand...
Owning what technology? FSD is just driver assistance with a fancy name, and all of the other companies have competing versions. Mercedes was the first to level 3 autonomy, with couple others having a few vehicles with that level.
 
acbytesla said:
Yes to all of that.

But is some service better than no service? Service from humans is expensive. In the last six years I've noticed that the bus stopping at the Park and Ride that runs only on weekdays and non-holidays has gone from 12 stops a day to 8 stops twice in the last 4 years.

Maybe deleting the driver could increase service to 14 times a day 7 days a week? I don't know which is more important. That is subjective.
Click to expand...
If only governments could come up with a way to reallocate resources from, say, people with $500 billion to those who need a regular bus service to participate in society.
 
arthwollipot said:
I assume that what Ivor is referring to is that situation on British motorways where the road is packed with a lot of traffic that is all moving quickly. You know, that incredibly dangerous situation where if someone has an accident, those behind can't see it until it's too late to avoid joining it and you end up with a 48-car pileup.
Click to expand...
What is even more surprising is that the UK has some of the safest roads in the world, even with the penny pinching and poorly implemented full-lane running motorways.
 
Dabop said:
www.facebook.com

52K views · 499 reactions | Hacker uncovers Tesla crash evidence in minutes | Jake Tapper

Hacker uncovers Tesla crash evidence in minutes
www.facebook.com www.facebook.com
About 6 min long- CNN story about the hacked Tesla crash data they actively tried to conceal..... took a hacker less than 5 mins for him to find the data Tesla said they couldn't find in over six months.....
Riiiiight......
View attachment 64481
Click to expand...
I didn't realise Tesla is still persisting with only using computer vision. That means if the computer doesn't understand or misinterprets what it "sees" then the best outcome is that it stops the vehicle. At a bare minimum you want another type of sensor to detect physical objects. I suspect this is coming from the self-described "nano-manager" at the top and not the actual engineers, who probably realise it's not a good approach but don't dare tell truth to power.
 
Ivor the Engineer said:
I didn't realise Tesla is still persisting with only using computer vision. That means if the computer doesn't understand or misinterprets what it "sees" then the best outcome is that it stops the vehicle. At a bare minimum you want another type of sensor to detect physical objects. I suspect this is coming from the self-described "nano-manager" at the top and not the actual engineers, who probably realise it's not a good approach but don't dare tell truth to power.
Click to expand...
Oh yeah, musk HATES lidar with a vengeance- he's 'vision is perfect and cameras are all thats needed' and once he pronounces the 'truth according to him' thats the end of it- mere engineers be damned, he's THE engineer of engineers (in his own tiny mind)...
 
Dabop said:
Oh yeah, musk HATES lidar with a vengeance- he's 'vision is perfect and cameras are all thats needed' and once he pronounces the 'truth according to him' thats the end of it- mere engineers be damned, he's THE engineer of engineers (in his own tiny mind)...
Click to expand...
Yeah, but those Lidar devices on the roof would spoil the sleek lines of a Tesla, so safety be damned!
 
Dabop said:
Oh yeah, musk HATES lidar with a vengeance- he's 'vision is perfect and cameras are all thats needed' and once he pronounces the 'truth according to him' thats the end of it- mere engineers be damned, he's THE engineer of engineers (in his own tiny mind)...
Click to expand...
Yeah, I don't get this at all. The fact that LIDAR is expensive is because it hasn't been mass produced. There is no reason it's cost couldn't be dramatically made cheaper to manufacture.
 
I don't know who's right in the "LIDAR or no LIDAR" argument, but accidents are going to happen with both human and autopilot driving and anyone who claims that even today's FSD is "perfectly safe" is delusional. Never mind 2019 technology.

The alarming thing about all that isn't that the accident happened, but Teslas's attempt to cover it up. And that any driver in 2019 was so blasé as to think he could scrabble for his phone on the floor while his car was moving.
 
Last edited:
The issue with adding sensors is it generally hits the Mean Time Between Failure (MTBF). To get that back up requires improving the MTBF of other parts of the system, which drives up the cost more than just the price of the additional sensors.
 
Ivor the Engineer said:
The issue with adding sensors is it generally hits the Mean Time Between Failure (MTBF). To get that back up requires improving the MTBF of other parts of the system, which drives up the cost more than just the price of the additional sensors.
Click to expand...
This isn't clear to me. If LIDAR and regular cameras are needed, they are needed.
 
You must log in or register to reply here.

Back
Top Bottom