5 ms·

The NTSB report confirms that. The crash attenuator—an accordion-like barrier that's supposed to cushion a vehicle when it crashes into the lane separator—

The NTSB report confirms that. The crash attenuator—an accordion-like barrier that's supposed to cushion a vehicle when it crashes into the lane separator—had been damaged the previous week when a Toyota Prius crashed at the same location. The resulting damage made the attenuator ineffective and likely contributed to Huang's death. kinda sounds like maybe that part of the road isn't well designed or marked too.

twblalock 8y ago
Any kind of self-driving system that is going to be usable and safe in the real world needs to be able to deal with the kind of roads we have in the real world. Maybe one day, when the majority of cars are self-driving, road design will change. To get there, self-driving cars will need to prove themselves on today's roads so people will buy them.
- stephengillie 8y ago
  Can all driving situations on all roadways today be computed in Polynomial time? Or are some NP? Some intersections are known to be complex for humans to navigate.
finnthehuman 8y ago
A crash attenuator is a “nice to have” kind of feature. The minimum drivable roadway conditions can be, and often are, worse. We have higher standards for human drivers and it wouldn’t be news except for the autopilot. Unless you want to ramp up caltrans to the point where it can fix non-essential parts of any and every roadway within hours of degredeation, it’s entirely impractical to expect them to be in perfect shape.
- jonhendry18 8y ago
  The simplest form of attenuator is just a group of plastic barrels filled with sand or water, which ought to be pretty quick and easy to replace. If California uses something fancier, then sand or water barrels would be an easy stopgap measure.
stetrain 8y ago
It isn't: http://www.dailymail.co.uk/sciencetech/article-5582461/Tesla-autopilot-nearly-crashes-exact-spot-Apple-engineer.html http://www.dailymail.co.uk/sciencetech/article-5582461/Tesla... I do think that an "Autopilot" / "ProPilot" / "SuperCruise" type system that does adaptive cruise + lane following + emergency braking should do a better job at detecting and reacting to that kind of barrier, but there is also evidence that this is a problematic location that has seen crashes by plenty of non-computer-assisted vehicles. Edit: Google street view of the location: https://www.google.com/maps/@37.410912,-122.0757037,3a,75y,205.08h,52.79t/data=!3m6!1e1!3m4!1sjXdU2LMnTz4QQshWHKU0Tg!2e0!7i16384!8i8192 https://www.google.com/maps/@37.410912,-122.0757037,3a,75y,2...
erobbins 8y ago
The previous crash was a drunk driver, at night. Tesla has hit a milestone, they are almost as good as impaired drivers!
kchhina 8y ago
It should be fairly easy to detect a chevron sign with yellow and black stripes [0] and take evasive measures to slow down, brake heavily or change lanes. From my understanding of CNN neural networks it's a fairly simple classification problem (that A. Karpathy researched, and taught at Stanford; and he's now leading the AI/vision team at Tesla). Looking at the video of the guy that tried that same route again, it looks like the sun was straight ahead (see starting 37th second of video [1]), and I am wondering if the camera had an issue identifying the sign due to the glare from the sun. Do we know if this problem has been fixed now? Can the chevron be identified at the same location by other Tesla drivers today, months after the guy that tested it and posted the video of his car swerving at the same place [1]? Following lane markings are never going to be a 100% reliable method for navigating roads. Even humans miss merging lanes, or have issues at complex surface street intersections where there are multiple overlapping lanes. But humans take evasive action.. they slow down, re-access, look at signs and probably get honked at. Following other cars is never going to be a reliable method for navigating roads. I don't know what the guy in front of me is going to do! I keep my safe distance and take evasive action if the guy in front of me does something stupid. I feel signs are most reliable in that sense. If there is a missing sign, a human is more likely bound to make a mistake as well. Maybe I am biased, because as someone who sails, signs (buoys, lights, colors, etc.) are what guide me in my navigation as there are no lane line markings on the water (not a perfect analogy because the speeds/traffic patterns are very different, but the point I am trying to make is about humans identifying the signs with muscle memory). A vision based neural network should be able to do the same. [0] https://english.stackexchange.com/questions/249780/what-do-you-call-this-segment-of-road-between-the-lane-and-an-exit https://english.stackexchange.com/questions/249780/what-do-y... [1] https://youtu.be/VVJSjeHDvfY?t=37 https://youtu.be/VVJSjeHDvfY?t=37

twblalock 8y ago

Any kind of self-driving system that is going to be usable and safe in the real world needs to be able to deal with the kind of roads we have in the real world. Maybe one day, when the majority of cars are self-driving, road design will change. To get there, self-driving cars will need to prove themselves on today's roads so people will buy them.

stephengillie 8y ago
Can all driving situations on all roadways today be computed in Polynomial time? Or are some NP? Some intersections are known to be complex for humans to navigate.

finnthehuman 8y ago

A crash attenuator is a “nice to have” kind of feature. The minimum drivable roadway conditions can be, and often are, worse. We have higher standards for human drivers and it wouldn’t be news except for the autopilot. Unless you want to ramp up caltrans to the point where it can fix non-essential parts of any and every roadway within hours of degredeation, it’s entirely impractical to expect them to be in perfect shape.

jonhendry18 8y ago
The simplest form of attenuator is just a group of plastic barrels filled with sand or water, which ought to be pretty quick and easy to replace. If California uses something fancier, then sand or water barrels would be an easy stopgap measure.

stetrain 8y ago

It isn't: http://www.dailymail.co.uk/sciencetech/article-5582461/Tesla-autopilot-nearly-crashes-exact-spot-Apple-engineer.html http://www.dailymail.co.uk/sciencetech/article-5582461/Tesla... I do think that an "Autopilot" / "ProPilot" / "SuperCruise" type system that does adaptive cruise + lane following + emergency braking should do a better job at detecting and reacting to that kind of barrier, but there is also evidence that this is a problematic location that has seen crashes by plenty of non-computer-assisted vehicles. Edit: Google street view of the location: https://www.google.com/maps/@37.410912,-122.0757037,3a,75y,205.08h,52.79t/data=!3m6!1e1!3m4!1sjXdU2LMnTz4QQshWHKU0Tg!2e0!7i16384!8i8192 https://www.google.com/maps/@37.410912,-122.0757037,3a,75y,2...

erobbins 8y ago

The previous crash was a drunk driver, at night. Tesla has hit a milestone, they are almost as good as impaired drivers!

kchhina 8y ago

It should be fairly easy to detect a chevron sign with yellow and black stripes [0] and take evasive measures to slow down, brake heavily or change lanes. From my understanding of CNN neural networks it's a fairly simple classification problem (that A. Karpathy researched, and taught at Stanford; and he's now leading the AI/vision team at Tesla). Looking at the video of the guy that tried that same route again, it looks like the sun was straight ahead (see starting 37th second of video [1]), and I am wondering if the camera had an issue identifying the sign due to the glare from the sun. Do we know if this problem has been fixed now? Can the chevron be identified at the same location by other Tesla drivers today, months after the guy that tested it and posted the video of his car swerving at the same place [1]? Following lane markings are never going to be a 100% reliable method for navigating roads. Even humans miss merging lanes, or have issues at complex surface street intersections where there are multiple overlapping lanes. But humans take evasive action.. they slow down, re-access, look at signs and probably get honked at. Following other cars is never going to be a reliable method for navigating roads. I don't know what the guy in front of me is going to do! I keep my safe distance and take evasive action if the guy in front of me does something stupid. I feel signs are most reliable in that sense. If there is a missing sign, a human is more likely bound to make a mistake as well. Maybe I am biased, because as someone who sails, signs (buoys, lights, colors, etc.) are what guide me in my navigation as there are no lane line markings on the water (not a perfect analogy because the speeds/traffic patterns are very different, but the point I am trying to make is about humans identifying the signs with muscle memory). A vision based neural network should be able to do the same. [0] https://english.stackexchange.com/questions/249780/what-do-you-call-this-segment-of-road-between-the-lane-and-an-exit https://english.stackexchange.com/questions/249780/what-do-y... [1] https://youtu.be/VVJSjeHDvfY?t=37 https://youtu.be/VVJSjeHDvfY?t=37