Tesla Driver Killed in Crash While Using Car's 'Autopilot'
 

Didn't think that would take long to surface on here.....

Sad for the driver, but saddening for the motor industry as it could be a huge and unfortunate set back in the world of autopilot driving, whether the Tesla ends up to blame or not....

Following with interest....

Even though this sounds more likely to be a driver error, rather than the "auto-pilot" feature error, it's likely to have consequences especially as it's happened in America.

From what I've read so far, speed is likely to be a contributing factor as the other driver didn't see him before turning. The Tesla driver should have been watching where he was going too.

I do wonder also, and it may not have made much difference, but why did the trailer not have guard rails to help prevent cars from going underneath?

My thoughts exactly.

While human-fault accidents like this happen every day, rightly or wrongly, there is zero public tolerance when technology is to blame. It's certainly a set-back.

Clearly there's still some way to go to perfect these systems. In this particular case, it seems to me that the system shouldn't have been relying on cameras alone. Surely distance/proximity sensors should help distinguish between truck and sky!?

But in any case, the driver should not have been relying entirely on the system; these things aren't foolproof, yet.

There are no U.S. national regulations for trailers having sideguards believe it or not, just rear rails. It's only in the last couple of years that city-owned vehicles have been fitted with them. Boston, MA was the first, then NYC.

If U.S. trailers have them already fitted, the chances are it's in the interests of pedestrian and cyclist safety in the cities. The rails would have to be upgraded and redesigned quite substantially to be suitable for reducing fatalities like this.

I read another article about this earlier today, and since tintin is the only one of us with a Tesla, it would be interesting to get his opinion on the Autopilot feature and how it works.

The article I read can be found here:

https://www.teslamotors.com/blog/tra...ss?redirect=no

In it, it states that Autopilot will check to ensure the drivers hands are on the wheel, and it will give visual and audible warnings, and will eventually slow the car down if the driver does not respond.

I have no idea what these look or sound like, but wouldn't you get annoyed with these alerts and just put your hands on the wheel? I was a passenger in a car where the driver didn't put his seat belt for about half a mile, and I found the warning noise annoying.

Since Sean asked, in my view Autopilot is brilliant - I couldn't drive 200 motorway miles at night when very tired in any other car with anything like such confidence, and even in more "normal" circumstances it's a massively effective driver aid. But it's not fully autonomous, and isn't meant to be.

The really sad fact about this (apart from the loss to his family and friends) is that he was such a strong Tesla advocate, and will be missed by the community: I just hope it wasn't caused by him pushing the boundaries of the system too far.

One of the Tech e-zines that I subscribe to came up with this:-

http://www.techrepublic.com/article/...&bhid=86036180


Looks like the usual media sensationalism :p

Good article! +++

Quote:

Originally Posted by tc4332 (Post 114466) Looks like the usual media sensationalism :p

Yep. Unfortunately the masses prefer to get their education from the gutter press.

I have to disagree with the article.

"Tesla itself have been anticipating the first fatal accident"

Not anticipating it very well, since they could have reduced the Autopilot feature to make it much less likely for this kind of accident to happen. For example Mercedes Benz anticipated it much more strongly by their system requiring a hand on the wheel at all times.

And this is just my opinion, they should have known any image processing program (robot) is guaranteed not to be able to cope with the complexity of the real world therefore should not have released the system at all.

I used to work in the robotics industry 16 years ago, although only very briefly (8 months). When I was there one of the projects was to replace human mushroom pickers with image processing gantry robots. The main challenge of course as with all robotics was to program the robot to interact with the real world, in this case via image processing. The programmers showed us the problems they were facing, the huge difficulty trying to program a robot to recognise ripe mushrooms, to understand 3D that some mushrooms are below others and not to try to pick shorter mushrooms first, etc. So many factors make it is impossible to code for every eventuality, slight changes in lighting conditions, change in hue due to mushroom growing at angles, etc.
If I remember correctly they gave up in the end, mushrooms are still picked by hand.

What caused auto manufacturers to think they can code image recognition for the even more varied and complex world of roads and traffic?
I say it is impossible and should not even be attempted. There is a good reason why rails exist, and even then accidents happen.

At the end of the day, there is no beating the human brain, its ready for anything.

http://neurosciencenews.com/neurosci...primates-4439/

Except, perhaps, participating in a referendum. :)

Or knowing when to call one. :rolleyes:

Or what to do after one. :o

Maybe we should all stick to picking mushrooms :ROFL:

Certainly simple rule-based algorithms are no match for the complexity and plasticity of the human brain (nor most animal brains for that matter).

However, artificial neural networks that are able to model the brain's pattern recognition and learning abilities do have the potential to match and even beat the human brain, at anything; the human brain is just a biological computer after all.

Given enough complexity/power there's no reason a human brain couldn't be entirely artificially modelled. Artificial neural networks as powerful as that are still a long way off of course, but they have been used successfully to a lesser extend in many systems, though I suspect that most self-driving systems presently use rule-based algorithms for the most part.

Computers don't get tired/lose concentration/behave irrationally - those human characteristics alone will outweigh the occasional benefits of superior intelligent reactions when it comes to overall accident statistics, which is why this is only going to head in one direction in the long run.

Nicely put Mark.
"Auto Pilot" is a tool to be used, not abused.
When clouting a chisel with a hammer you do not let go of the chisel, or the hammer either.
Hope that makes sense!!! I'd better go to bed, too much sun this evening.

I have wired up the second set of my foggers in the water effect and the neighbour came knocking on the door bell to warn me that my water effect was on fire. :ROFL:

A good article on this (from Vanity Fair FFS!..)
 

Hmmm, slightly confused neighbour - knocking on the door bell! :)

Just testing. I usually say "Ringing the door knocker".
Proves that someone is paying attention +++

A second crash is also being investigated, suspected Autopilot again.

http://www.bbc.co.uk/news/technology-36736103

Not sure that one has been mentioned yet.

This is worth a read, in case you think these all of these press articles are just simple, straightforward investigative journalism - http://electrek.co/2016/03/07/koch-b...hicles-begins/

Here is another article, this one from a technical ezine, with the flip side to this.

http://www.techrepublic.com/article/...&bhid=86036180