Live data from Hacker News

BMW shows off 'drifting' self-drive cars

bbc.co.uk

61–70 of 144 posts

Re: BMW shows off 'drifting' self-drive cars

#61
post #29
post #16

This is bad news (despite being a BMW owner) In future driving will no longer be fun, if you want to drive "non-automated" car or disable "auto-pilot" your insurance company will automatically jack up your premium, since they would consider self driving cars "safer"

I can't see any reason why we can't have both; you control the car until it detects a dangerous situation and then takes over. Sounds ideal really, you could go faster than you can now, safe in the knowledge that if things go bad the computer will kick in and take over.

I'd think that very often by the time things have gone wrong, it's too late to prevent the crash even with perfect intelligence. Unless perhaps you have the system take back over so often that it's more frustrating to manually drive than to just leave it on auto-pilot anyway.

Re: BMW shows off 'drifting' self-drive cars

#62
post #54

Anybody know if self-driving cars can distinguish between safe and unsafe object; say a big stuffed animal, and a rock in the middle of the road. Would they make big maneuvers to avoid both?

Human beings aren't especially good at that either...

Tell me about it. I am constantly surprised by how close I need to get to the target in low light conditions to realize it is really not what I earlier recognized it to be.

It usually makes me wish I could enable some sort of Splinter Cell heat sensor.

Re: BMW shows off 'drifting' self-drive cars

#63
post #24
post #16

This is bad news (despite being a BMW owner) In future driving will no longer be fun, if you want to drive "non-automated" car or disable "auto-pilot" your insurance company will automatically jack up your premium, since they would consider self driving cars "safer"

If you want to have fun, there will no doubt be private tracks where you can go and drive in a fun way. As someone who just gave up commuting about 1 traffic-laden hour each way, driving stopped being fun long, long ago. Too many of my commutes were made an unnecessary 50% or more longer because of drivers who were probably having a lot of fun until they caused an accident and backed up the road for everyone else.

At the same time, the entire continent is finally pushing back on that 1-hour commute. People are returning to cities, transit and infill and dense construction are becoming political issues, etc. Part of the impetus for this re-urbanization is the misery of commute.

Beyond commuting woes, density is also more efficient for providing services and for environmental concerns.

So I'm slightly worried that a move towards self-driving cars will allow us to backslide on this progress. Obviously I want the technology - it will be safer, more efficient, and healthier - but I'm worried about how culture could shift around it.

Re: BMW shows off 'drifting' self-drive cars

#64
post #24

Earlier quoted context omitted.

If you want to have fun, there will no doubt be private tracks where you can go and drive in a fun way. As someone who just gave up commuting about 1 traffic-laden hour each way, driving stopped being fun long, long ago. Too many of my commutes were made an unnecessary 50% or more longer because of drivers who were probably having a lot of fun until they caused an accident and backed up the road for everyone else.

I autocross and take my car to a private track for time trials, I'm starting to face the fact that this is not the best plan because there is now a karting track (0-120 in 6 seconds) that is closer to me. My car, that I justified it's sportiness because of this hobby, is more expensive at the track, burns more, fuel, has more maintenance issues, and could just never perform the 3gs that a kart can.

Embrace the Kart. My dad used to do that as a hobby and they're impressive machines. Just remember that you've got no crumple zones and your legs are right at the front.

Re: BMW shows off 'drifting' self-drive cars

#65

Earlier quoted context omitted.

A self-driving car system might be statistically safer than a human drivers in overall conditions, and still be prone to failure in an exceptional situation a particular driver wouldn't have been. If you have the potential to be that driver in that situation, I could understand feeling the scare quotes are warranted. Which percentile of the population does an automated system need to outperform to be safer rather tha…

I am (in my own estimation) a very good driver. I drive a manual transmission car, and enjoy using the flexibility that provides in my daily driving. I drive emergency vehicles of various sizes, in a lot of less than ideal conditions (I live in upstate New York, so weather is a big factor in driving conditions). I have no problem believing that a computer will be able to do an objectively better job of driving than I…

Further, once all the cars on the road are automated, they can alert one another to their own changes in routing, mechanical failures, etc, allowing everyone else the chance to reroute and avoid as needed. We could eliminate traffic lights and stop signs and we'll not only get to our destinations safer, but also sooner.

Re: BMW shows off 'drifting' self-drive cars

#66
post #65

Earlier quoted context omitted.

I am (in my own estimation) a very good driver. I drive a manual transmission car, and enjoy using the flexibility that provides in my daily driving. I drive emergency vehicles of various sizes, in a lot of less than ideal conditions (I live in upstate New York, so weather is a big factor in driving conditions). I have no problem believing that a computer will be able to do an objectively better job of driving than I…

Further, once all the cars on the road are automated, they can alert one another to their own changes in routing, mechanical failures, etc, allowing everyone else the chance to reroute and avoid as needed. We could eliminate traffic lights and stop signs and we'll not only get to our destinations safer, but also sooner.

Not only that, but emergency vehicle will be able preempt roads directly, rather than relying on the 18 year old kid with headphones on to a) recognize the lights and sirens, b) care, and c) do something not stupid (like pull into the way of the fire truck...)

Re: BMW shows off 'drifting' self-drive cars

#67
post #16

This is bad news (despite being a BMW owner) In future driving will no longer be fun, if you want to drive "non-automated" car or disable "auto-pilot" your insurance company will automatically jack up your premium, since they would consider self driving cars "safer"

Considering insurance companies already insure airplanes, and you can turn the autopilot off without losing insurance, I somehow doubt this future. Airplanes are both more expensive and cause more damage when you crash them, and yet, you can rent them and insure them all the same.

Re: BMW shows off 'drifting' self-drive cars

#68
post #35
post #24

Earlier quoted context omitted.

If you want to have fun, there will no doubt be private tracks where you can go and drive in a fun way. As someone who just gave up commuting about 1 traffic-laden hour each way, driving stopped being fun long, long ago. Too many of my commutes were made an unnecessary 50% or more longer because of drivers who were probably having a lot of fun until they caused an accident and backed up the road for everyone else.

As someone with a similar commute, I lament I can only upvote this once. Every time I-95 severely backs up because some asshole managed to crash on a day with perfect weather and visibility, I think to myself: "I hope it was worth it."

As bad as being stuck in traffic is, usually you're having a better day than whoever was involved in the wreck, most of whom probably didn't do it on purpose.

Re: BMW shows off 'drifting' self-drive cars

#69
"One 2013 study by the Eno Center for Transportation suggested that if 10% of cars on US roads were autonomous this could reduce fatalities by about 1,000."

This doesn't quite seem right. How does a 10% shift towards autonomous vehicles only equate to a 2% decrease in fatalities? Why wouldn't this scale linearly?

Re: BMW shows off 'drifting' self-drive cars

#70

"One 2013 study by the Eno Center for Transportation suggested that if 10% of cars on US roads were autonomous this could reduce fatalities by about 1,000." This doesn't quite seem right. How does a 10% shift towards autonomous vehicles only equate to a 2% decrease in fatalities? Why wouldn't this scale linearly?

Who says it doesn't scale linearly? That certainly can't be concluded from the sentence you cited. Are you assuming that self driving cars result in 0 fatalities?
Post reply on HN