Search results

This makes me want to build an autonomous mower that isn't terrible. Better wheels, better blades and a pathing algorithm that doesn't make it look like you gave a schoolbus of children 2 minutes of push time each before your lawn was finished.

When my grandkids try that out I'll have them tell your grandkids how it is.

I think this will be a pretty big consideration for a lot of people. I know when I went to school I had to drive through the city. Because of how poorly timed the lights were, if I went the speed limit it took me 35-40 minutes to get to school. If I sped some 10-15mph over and got lucky with the...

A bunch of people are doing vehicle-to-vehicle (v2v) and vehicle to infrastructure (v2i) stuff, but it's not widespread enough to create the kind of hive mind we're hoping for. It will be eventually. It already exists with a driver-in-the-loop system for me in that very frequently I'll get...

They should do what human drivers are told to do. Slam on the brakes. There's a few reasons for this: 1. Programming a planning system to successfully navigate high slip manoeuvres is bad for a variety of reasons. 2. Generally, slamming on the brakes is the best outcome for many of these...

pics or it didn't happen. I also use FLIR for work, a lot of fun to see things in a different spectrum. I've always wanted to get some sweet goggles like this: For you know, home invasion and keeping my family and this country safe. Definitely not for doing Michael Ironside impressions or...

I don't even know what your plan is, what you're trying to solve or how you think transponders and collision radar will solve it.

People are talking about getting banned from pretty much any third party access. It's very easy for them to tell when that happens. Less easy (impossible if done right) for them to tell when you're packet sniffing through a VPN, which can get you solid IV numbers.

Gavin has a dream team of wrecking ball pokemon. He's not butthurt because he has to beat the shit out of someone who may or may not be cheating.

What about all the dynamic obstacles that you can't put a transponder on?

Even worse!

Inconsequential point: I'd be really annoyed to be a truck driver who has to go from driving a truck cross-country 7.5 hours a day and load/unload 0.5hours a day to having to unload/load trucks 8 hours a day.

Tesla vehicles wouldn't operate on a race track without some specific modifications. Autonomous vehicle racing would be a lot of fun though. You could cut loose on a lot of the safety features and open up the design restrictions.

What?

Reminds me of what Artamis drove in Ready Player One.

My argument is this: 1. It's unlikely that Peterbilt, Daimler, Freightliner, Kenworth etc are 10 years away from selling inter-city autonomy commercial vehicles. Perfectly viable in 5 years is fiction. 2. Even if #1 was false, it's unlikely the first few versions over 5-10 years would cost less...

I got to see the delphi system recently. It's very cool. I'm really looking forward to seeing it and all the other systems integrated. I'm betting that the driving behavior of autonomous cars will be very noticeable. My guess is that it will be very cautious.

Cool paper. Mono and stereo cams play a big role in autonomy, but outdoor computer vision (CV) is hard. People don't realize how great our eyes are, with all their cones and rods. The human eye is really amazing and our cameras can beat it in some ways, but in handling variety they are no where...

Don't bet the house on that.

So, it's likely that Ford will be able to do XYZ by 2021 (by XYZ, I mean have a limited release of highly constrained, driverless taxis. Japan will do it in 2020 for their olympic games). But once that occurs, deep market penetration of autonomy at that level won't happen quickly. And further...

This is actually pretty disruptive.... ingenious.

This is a big problem in advanced research. You're trying to push grand ideas and get funding for them, but you don't have a real idea for what the timeline is going to be. Even research in private sector, bluechip companies like Ford this is important, because the executives and board control...

The Jugs have no need for a union, because they run this show now.

What advantages does that have over ?

By the time we get to the point where the traffic rules change such that no, you won't be allowed to break rank and meatbag drive, you'll probably want to use the autonomous system as it will be hurtling you across the interstate at 100+mph

Virtually every obstacle whether it's a garbage bag, stroller, box or turkey will be responded to in the same way: hard stop and wait for the obstacle to move away, or move around it if possible.. Google is doing some very amazing things with object analysis from lidar scans, but that's more for...

I don't have a projected timeline. Also the differences between level 2,3,4 and 5 are muddier than SAE will admit. It's arguable that until you can go from curb to curb, you're not at level 3. It's not an argument I'd care to have though, because it misses the core point, which is this: People...

Please contact your congressmen and tell them you care deeply about this issue and they should push for looser money in vehicular autonomy.

RIP pizza delivery. I actually think drones delivering pizza/jimmyjohns etc will become more prevalent than cars. And it's going to be awesome.

I agree, but standard units change. When driving outside of normal hours I have to realize I'm not driving with the regular pros of my commute but the shitter-class. People in their regular commute are actually very good drivers when it comes to normal speed.

Suggesting what is bullshit? The work the automotive industry is doing, including uber/google/tesla, volvo/bosch/ford/freightliner/gm/honda/fuckingeveryone, cmu/standford/mit is not bullshit at all. What is bullshit, or rather is too optimistic, is the idea that we're going to all be driven...

I'm not big into v2v stuff, but I went to a autonomous vehicles (AUVSI) conference recently and asked all the v2v guys what their security systems were. There's really no good answers yet. In the end I think the difficulty in doing so combined with the ease of creating havok with much simpler...

Aye, that's the basis of it and one of the big upsides of V2V comms. But it's going to be a long road to get there. And even when you're there, it's going to be a shitshow getting all the vehicles to abide by the same standard.

The camera operator will let you watch porn as long as you let the camera watch too.

What's funny about this is that until there is widely accepted V2V communication, the opposite will be true. You know how you drive on the freeway during rush hour and most people are driving 70mph+ like 2 car lengths away from the car in front them? An autonomous system will say fucking nope to...

Question is whether that's a good thing or not. Right now you have a X mph road that everyone travels X+10 on and gives police officers an easy reason to pull everyone over, but everyone is generally happy with the speed. If you systematically enforce it, you now have more traffic and people...