5 ms·

Google's self-driving Lexus cuts off self-driving Audi

11 points by MattGrommes 11y ago

seren 11y ago
Which raises the question why isn't there a standard protocol so that two near-autonomous car could broadcast and/or negotiate their next move ?
- Someone1234 11y ago
  Because autonomous cars drive in a world that is 99.99% human drivers, so for automation to be considered viable they have to cope with that not the 00.01% of autonomous-autonomous scenarios. There is also legitimate security concerns, bad data from bad actors, interference, weather issues, and so on. It is just better to depend on reliable data from sensors and design systems around them, than assuming perfect data from other cars (which might be flawed).
  - seren 11y ago
    That makes a lot of sense. However at some point, I expect some kind of approval body that would vet an autonomous car comm, similar to GSM for example. This is also a situation where you don't want bad actors emitting in the wrong frequency band, at the wrong time, with the wrong power, etc..
  - codezero 11y ago
    Ah yes, very good point, I imagine a trolling-car-communication hack would come very soon after a car-to-car link is created. :)
- codezero 11y ago
  I expect this will be inevitable, but since the density of self-driving cars is so low, developing such a protocol is premature. Also, the over all design should exist in such a way that it doesn't rely on such car to car communication. For example, if a person is standing in the road, or a deer jumps into the lane, the car should be able to react accordingly. Having a car-to-car protocol would be helpful for the average case, but since there are so many tricky edge cases, I suspect full autonomy being developed to its maximum potential may make it unnecessary.
chrisper 11y ago
Since these Google cars drive around here in the Bay Area, I guess they applied the local driving culture (aggressive) to the self driving cars.
SlashmanX 11y ago
> "The story was taken completely out of context when describing a type of complex driving scenario that can occur in the real world. Our expert provided an example of a lane change scenario that our car recently experienced which, coincidentally, was with one of the Google cars also on the road at that time. It wasn’t a 'near miss' as described in the Reuters story". > Instead, she explained how this was a normal scenario, and the Delphi car performed admirably. > "Our car did exactly what it was supposed to," she wrote. "Our car saw the Google car move into the same lane as our car was planning to move into, but upon detecting that the lane was no longer open it decided to terminate the move and wait until it was clear again." Source: http://arstechnica.com/cars/2015/06/no-2-self-driving-cars-didnt-have-a-close-call-on-silicon-valley-streets/ http://arstechnica.com/cars/2015/06/no-2-self-driving-cars-d...
- forthefuture 11y ago
  I read the original title and immediately thought that a more accurate title would have been, "Two self driving cars don't hit eachother".
  - SlashmanX 11y ago
    They just updated the article it seems with a quote from a Google spokeswoman > Courtney Hohne, a Google spokeswoman, e-mailed Ars: "The headline here is that two self-driving cars did what they were supposed to do in an ordinary everyday driving scenario."
Someone1234 11y ago
Should likely just link to the Reuters article instead: http://www.reuters.com/article/2015/06/26/us-autos-selfdriving-nearmiss-idUSKBN0P601T20150626 http://www.reuters.com/article/2015/06/26/us-autos-selfdrivi... This article is just a snippet of Reuters with nothing added and some stuff removed.
- Patrick_Devine 11y ago
  If you follow the original link, it turns out Reuters fabricated the whole thing and still hasn't updated their article to retract it. Here's the update: > A Delphi spokesperson emailed Fortune with the following statement: “During a demo drive of our automated vehicle, our expert used the interaction with the Google car as an example of the types of scenarios that the car can encounter in real-world driving. It was an anecdote of an interaction, not a ‘near miss’. Reuters completely misrepresented the facts.”
ams6110 11y ago
The near collision raises important questions about self-driving cars, including who’s responsible when two are involved in an accident. Simple. The same sort of accident investigation that we do now would take place, and the owner of the car that caused the accident would be liable.

11 points by MattGrommes 11y ago

seren 11y ago

Which raises the question why isn't there a standard protocol so that two near-autonomous car could broadcast and/or negotiate their next move ?

Someone1234 11y ago
Because autonomous cars drive in a world that is 99.99% human drivers, so for automation to be considered viable they have to cope with that not the 00.01% of autonomous-autonomous scenarios. There is also legitimate security concerns, bad data from bad actors, interference, weather issues, and so on. It is just better to depend on reliable data from sensors and design systems around them, than assuming perfect data from other cars (which might be flawed).
- seren 11y ago
  That makes a lot of sense. However at some point, I expect some kind of approval body that would vet an autonomous car comm, similar to GSM for example. This is also a situation where you don't want bad actors emitting in the wrong frequency band, at the wrong time, with the wrong power, etc..
- codezero 11y ago
  Ah yes, very good point, I imagine a trolling-car-communication hack would come very soon after a car-to-car link is created. :)
codezero 11y ago
I expect this will be inevitable, but since the density of self-driving cars is so low, developing such a protocol is premature. Also, the over all design should exist in such a way that it doesn't rely on such car to car communication. For example, if a person is standing in the road, or a deer jumps into the lane, the car should be able to react accordingly. Having a car-to-car protocol would be helpful for the average case, but since there are so many tricky edge cases, I suspect full autonomy being developed to its maximum potential may make it unnecessary.

chrisper 11y ago

Since these Google cars drive around here in the Bay Area, I guess they applied the local driving culture (aggressive) to the self driving cars.

SlashmanX 11y ago

> "The story was taken completely out of context when describing a type of complex driving scenario that can occur in the real world. Our expert provided an example of a lane change scenario that our car recently experienced which, coincidentally, was with one of the Google cars also on the road at that time. It wasn’t a 'near miss' as described in the Reuters story". > Instead, she explained how this was a normal scenario, and the Delphi car performed admirably. > "Our car did exactly what it was supposed to," she wrote. "Our car saw the Google car move into the same lane as our car was planning to move into, but upon detecting that the lane was no longer open it decided to terminate the move and wait until it was clear again." Source: http://arstechnica.com/cars/2015/06/no-2-self-driving-cars-didnt-have-a-close-call-on-silicon-valley-streets/ http://arstechnica.com/cars/2015/06/no-2-self-driving-cars-d...

forthefuture 11y ago
I read the original title and immediately thought that a more accurate title would have been, "Two self driving cars don't hit eachother".
- SlashmanX 11y ago
  They just updated the article it seems with a quote from a Google spokeswoman > Courtney Hohne, a Google spokeswoman, e-mailed Ars: "The headline here is that two self-driving cars did what they were supposed to do in an ordinary everyday driving scenario."

Someone1234 11y ago

Should likely just link to the Reuters article instead: http://www.reuters.com/article/2015/06/26/us-autos-selfdriving-nearmiss-idUSKBN0P601T20150626 http://www.reuters.com/article/2015/06/26/us-autos-selfdrivi... This article is just a snippet of Reuters with nothing added and some stuff removed.

Patrick_Devine 11y ago
If you follow the original link, it turns out Reuters fabricated the whole thing and still hasn't updated their article to retract it. Here's the update: > A Delphi spokesperson emailed Fortune with the following statement: “During a demo drive of our automated vehicle, our expert used the interaction with the Google car as an example of the types of scenarios that the car can encounter in real-world driving. It was an anecdote of an interaction, not a ‘near miss’. Reuters completely misrepresented the facts.”

ams6110 11y ago

The near collision raises important questions about self-driving cars, including who’s responsible when two are involved in an accident. Simple. The same sort of accident investigation that we do now would take place, and the owner of the car that caused the accident would be liable.