4 ms·
>> “It’s really, really hard,” Krafcik said during a live-streamed tech conference. “You don’t know what you don’t know until you’re actually in there and tryin
by YeGoblynQueenne 8y ago
>> “It’s really, really hard,” Krafcik said during a live-streamed tech conference. “You don’t know what you don’t know until you’re actually in there and trying to do things.”
Now, where was that Rodney Brooks interview? Ah, here it is:
https://cdn.arstechnica.net/wp-content/uploads/2018/06/2nd-Part-RB-for-Ars-Transcript-2.pdf https://cdn.arstechnica.net/wp-content/uploads/2018/06/2nd-P...
>> Rob Reid: (...) Why do you think these people and others are overestimating the rate of development in this field?
>> Rodney Brooks: I think they're making a bunch of mistakes. I asked them when did the first car drive down a freeway for 10 miles at 50 miles an hour. They know that the Google cars did that in 2004 or 2005. It was actually done in Germany in 1987.
>> Rob Reid: Wow.
>> Rodney Brooks: When are we going to get the first car, hands off the steering wheel, feet off the pedals, drive coast to coast in the US? Yeah, well, it actually happened in 1995 with the Navlab Project from Carnegie Mellon University. My point is, everyone thinks, oh this is just a [inaudible 00:44:50], this is going to happen quickly. It's been around a long time to get to where we are. I have now demonstrated to them that their scale is wrong, their start point is wrong. It has taken a lot longer to get to where we are now.
So, to anyone who had any understanding of the technology it must have been obvious for a long, long time that "it's really, really hard" and that it wouldn't just take a few years after the time that Google announced the Google car. I struggle to believe that Krafcik is not one of those people, or that major automotive company CEOs are not.
Basically, those people have been bullshitting everyone and they're still bullshitting everyone, and they were probably not even interested in developing self-driving cars in the first place, it was all just some stupid marketing game supported by a very excitable tech press and a gullible public.
- save_ferris 8y agoYour point reminds me a bit of Theranos, and how they were able to coast on a completely fraudulent but awe-inspiring idea. It might not be related, but whenever I read stories this, I'm reminded of the ways that tech is marketed in the world. It's impossible for the public to understand whether or not a certain technology is feasible, practical, or even possible, but marketing teams get an unbelievable amount of leeway when marketing products and services. And they get to do it without really telling you what data is collected and how it's used, etc., etc. The power asymmetry between what tech companies are allowed to say versus the limited technical understanding of the public is brutal, and not getting any better.
- DEADBEEFC0FFEE 8y agoHaving bold dreams are important too. Trying hard and failing is a perfectly useful data point. Pretty sure there would have been many naysayers 20 years ago, when presented with what is today's mobile technology.
- pm90 8y agoI don't think they're the same thing though. Touchscreen phones with high speed internet was an unsurprising goal, even if the pace of smartphone adoption has been fantastic. Driverless cars... unless there have been some fundamental breakthroughs in AI tech that allows for such a fully unsupervised system, its hard to see how this can be a possibility. Now, reducing the complexity far more... assisted driving tech is already pretty mainstream. Autonomous driving for predictable, long haul routes (inter-city trucks on US highways) is an exciting market too.
- hndamien 8y agoHow about reusable rockets?
- sverige 8y agoEveryone is wrong about long haul trucks, frankly. It will be easier to get fully autonomous cars than trucks. They're much longer and wider. They're at least twenty times heavier. They require much more room to maneuver, to start, and to stop. If they're involved in a crash, they cause a lot more damage. Not to mention that there's nothing really predictable about a long route. Traffic, weather, construction, and every other variable is more likely to change over a longer route as well. It would be far better to focus on making autonomous cars first.
- Dylan16807 8y agoIt's not that trucks will have it first, per se. It's that trucks spend a much larger fraction of their time on freeways, so they can get a much bigger benefit from a system that only works there. Specifically on freeways, the difference in difficulty is pretty minor. The lanes are wide and you don't need to accelerate very rapidly. > It would be far better to focus on making autonomous cars first. Mu. The same system's going to be on both types of vehicle.
- chroma 8y agoRodney Brooks is misleading. He says things like, "It was actually done in Germany in 1987." but neglects to mention that the road had no other cars on it.[1] He also claims that the first autonomous coast to coast drive happened in 1995, but actually the project was autonomously controlling steering, not gas or brakes, and 150 miles of the trip were driven by humans.[2] Modern autonomous vehicles are much more impressive. They were successfully navigating urban environments 12 years ago.[3] Waymo's cars have driven over 10 million miles and disengagement rate is once every 11,000 miles.[4] We're at the point where no major breakthroughs are needed, just incremental improvements. That's what's different from earlier eras of autonomous vehicles. 1. https://en.wikipedia.org/wiki/History_of_self-driving_cars#1980s https://en.wikipedia.org/wiki/History_of_self-driving_cars#1... has this quote: > In the 1980s, a vision-guided Mercedes-Benz robotic van, designed by Ernst Dickmanns and his team at the Bundeswehr University Munich in Munich, Germany, achieved a speed of 39 miles per hour (63 km/h) on streets without traffic. 2. http://www.cs.cmu.edu/afs/cs/usr/tjochem/www/nhaa/nhaa_home_page.html http://www.cs.cmu.edu/afs/cs/usr/tjochem/www/nhaa/nhaa_home_... 3. https://en.wikipedia.org/wiki/DARPA_Grand_Challenge_(2007) https://en.wikipedia.org/wiki/DARPA_Grand_Challenge_(2007) 4. https://medium.com/waymo/an-update-on-waymo-disengagements-in-california-d671fd31c3e2 https://medium.com/waymo/an-update-on-waymo-disengagements-i...
- YeGoblynQueenne 8y agoYes, of course there has been progress since the '80s and '90s. Things would be really bad otherwise. >> Modern autonomous vehicles are much more impressive. It really depends on what you consider impressive. The DARPA Grand Challenge involved one road loop with stunt drivers instead of real traffic. These are still strictly controlled, laboratory conditions that tell us nothing about the ability of robot cars to operate in the real world. Waymo's disengagement rates don't really say anything, either. Perhaps Waymo is now driving its cars in easier conditions after noticing that they tended to disengage too often. What we know for sure is that Waymo doesn't have autonomous cars -as noone else does. If they did, they'd be out on the streets without safety drivers and counting autonomous miles, not miles without disengagement. According to the post you link, reporting rate of disengagement is required, but the fact that Waymo chooses to advertise theirs as a measure of improvement of their cars tells me that they have no real results to show and instead choose to tout a meaningless proxy just to make people believe that they are further ahead on the road to autonomy than they really are.