Driverless Hype Collides with Merciless Reality • Hacker News Board

Re: Driverless Hype Collides with Merciless Reality

#201

Post by 01100011 » Thu, Sep 13, 2018, 9:03 PM UTC

pkulak wrote: post #170

Animats wrote: post #71

Driverless cars are coming along just fine. Waymo is making steady progress. The sensors are getting better. The problem comes from all those "fake it til you make it" startups, Uber and Tesla being the worst. Both have killed people. This is mostly about sensors and geometry. Machine learning has a role, but only in target identification. That's how Waymo does it. The fake it til you make it crowd had the fantasy th…

No one thinks that you just hook up ML to a camera and you're done. Literally everyone working on this only uses ML for identification. When that Model X crashed into a highway divider, it was because the lines on the road were very faint and there was a fork in the line that actually wasn't the _real_ lane line. The car followed that by mistake, but only because of the misidentification of the lane line. How are mor…

Well, my optical sensors(i.e. eyes) provide me with enough information to discriminate a cat from a bag, so I suppose with enough processing we should be able to do the same with a control system. Cars have an added advantage that their vision systems could work in lower light and non-optical wavelengths. I don't see any fundamental issues with getting self-driving cars working. It may not happen in 5 years, but it will happen.

Re: Driverless Hype Collides with Merciless Reality

#202

Post by Fricken » Thu, Sep 13, 2018, 9:07 PM UTC

jimmy1 wrote: post #152

Animats wrote: post #71

Driverless cars are coming along just fine. Waymo is making steady progress. The sensors are getting better. The problem comes from all those "fake it til you make it" startups, Uber and Tesla being the worst. Both have killed people. This is mostly about sensors and geometry. Machine learning has a role, but only in target identification. That's how Waymo does it. The fake it til you make it crowd had the fantasy th…

> Driverless cars are coming along just fine. I am curious why you believe this to be the case? Driverless cars have yet to even come close to being workable in adverse conditions, to my knowledge, please correct me if I am wrong. > This is mostly about sensors and geometry. And mostly about the limitations of sensors, and the limitation of available algorithms to overcome geometry. A algorithm a driverless car uses…

The infrastructure driverless vehicles rely on is their own internal maps, which have lanes and signage marked out. Being able to visually detect these things is important, but it is a measure for added redundancy.

Great strides have been made in using machine learning to filter out obscurants such as snow. Perception for autonomous vehicles is effectively a solved problem.

The biggest technical challenges are related to planning. So say you're approaching an intersection. There is a pedestrian about to cross, a cyclist in front of you and another vehicle waiting to turn left across your path. As a human, you understand that if you behave one way, it will cause the pedestrian, the cyclist and the other car to respond a certain way, but if you respond to the situation another way, it will cause all three to respond differently.

Our ability to game out scenarios like this is intuitive, but for AI to predict how it's behavior as an agent will influence the behaviour of other agents on the road is a daunting undertaking, particularly when taking into account the full scope of scenarios that need to be mastered before an autonomous can reliably safely navigate anywhere.

Re: Driverless Hype Collides with Merciless Reality

#203

Post by icebraining » Thu, Sep 13, 2018, 9:12 PM UTC

baybal2 wrote: post #100

emiliobumachar wrote: post #84

Earlier quoted context omitted.

We've had flying cars for decades. They're called helicopters. Unfortunately they're much too expensive for the masses. But I don't see why anyone would consider the concept stupid.

My point is that the popular image of flying cars with retractable wings and wheels turning into jet engines is ludicrous, and that yes, light personal airplanes and helicopters are there. And the same parallel for driverless car, except for even when well executed it is kinda of arguable usefulness/value.

arguable usefulness/value

Are regular cars useful? A self-driving car is just as useful for someone who can't drive: minors, many seniors, many people with disabilities - or people who just don't have a driving licenses, like me.

Re: Driverless Hype Collides with Merciless Reality

#204

Post by linuxftw » Thu, Sep 13, 2018, 9:13 PM UTC

jimmy1 wrote: post #152

Animats wrote: post #71

Driverless cars are coming along just fine. Waymo is making steady progress. The sensors are getting better. The problem comes from all those "fake it til you make it" startups, Uber and Tesla being the worst. Both have killed people. This is mostly about sensors and geometry. Machine learning has a role, but only in target identification. That's how Waymo does it. The fake it til you make it crowd had the fantasy th…

> Driverless cars are coming along just fine. I am curious why you believe this to be the case? Driverless cars have yet to even come close to being workable in adverse conditions, to my knowledge, please correct me if I am wrong. > This is mostly about sensors and geometry. And mostly about the limitations of sensors, and the limitation of available algorithms to overcome geometry. A algorithm a driverless car uses…

I think interesting edge cases include: 'empty cardboard box or log?' 'Deer just crossed road, is there another one close behind?' 'Driving temporarily on the wrong side of the street due to construction or a downed tree limb' 'Police directing traffic' 'Driving on wrong side of road to avoid collision' 'One lane bridges' 'Busy parking garage with blind corners' 'City Bus driver forces themselves recklessly into your lane' 'Other driver in round about about to cut you off' 'Pulling out to cross a busy 4 line highway with no median or traffic signal'

Not to mention, any snow on the road is going to make it supremely hard for a computer to determine where their lane is, or even the road is.

Re: Driverless Hype Collides with Merciless Reality

#205

Post by Zigurd » Thu, Sep 13, 2018, 9:14 PM UTC

boxcardavin wrote: post #115

Zigurd wrote: post #40

Earlier quoted context omitted.

"Driverless car" is probably not the outcome for autonomous vehicles. At one end of the scale, cities will use road space vastly more efficiently with optimized vehicles for groups of riders, and at the other end of the scale, short-haul air travel will be challenged by door-to-door 150mph+ highway vehicles. None of these will look much like a personal car, and only the wealthy will consider owning one that will beco…

Self-delivering bicycles my man, too light to kill people in impacts, slow when autonomous.

When cars don't crash they won't have to be crashworthy. They could be much lighter.

Re: Driverless Hype Collides with Merciless Reality

#206

Post by beat » Thu, Sep 13, 2018, 9:15 PM UTC

dragontamer wrote: post #194

beat wrote: post #182

Earlier quoted context omitted.

The 100% attention thing I why I have tremendous faith that computers will be out-driving humans in short order. Even really good human drivers have a limited perspective and focus. And even the best human drivers are subject to emotional state.

Yeah, humans suck at attention. But until we can program computers to tell me what is a foreground moving object, what is a shadow, and what is the background... you literally can't solve the "Should the car apply the brakes right now" problem. https://arxiv.org/pdf/1801.02225.pdf Just go through this recent paper I found on DDG.gg and see the amount of effort it takes to parse foreground / background data that a sel…

That's why I think the future lies with a mix of (possibly layered) sensing and decision-making technologies. LIDAR + cameras + downloaded maps + machine learning + whatever.

The "until we can program computers to..." is a when question, not an if question. There's nothing about driving, in any situation, that doesn't fall in the face of "assuming infinite computing power and infinitely good sensors". Driving isn't a creative act, it's a responsive one.

Re: Driverless Hype Collides with Merciless Reality

#207

Post by naravara » Thu, Sep 13, 2018, 9:15 PM UTC

Animats wrote: post #71

Driverless cars are coming along just fine. Waymo is making steady progress. The sensors are getting better. The problem comes from all those "fake it til you make it" startups, Uber and Tesla being the worst. Both have killed people. This is mostly about sensors and geometry. Machine learning has a role, but only in target identification. That's how Waymo does it. The fake it til you make it crowd had the fantasy th…

>You can maybe identify "traffic light", "car", "pedestrian", and "deer" with machine learning. That's just used to guess what they're going to do next. It's not used to decide if they're an obstacle. Obstacle detection and avoidance is all sensors and geometry.

Honestly, getting them to work to this basic level is the "easy" part. Real world driving conditions in urban environments are suboptimal at best. You're not only going to need a system that can navigate around and not hit things. You need it to be able to dynamically route around road-construction, do clever things to find parking.

Roads in most places aren't that well standardized. It takes a lot of small judgement calls for a taxi to actually do its job. Uber and Lyft still have trouble dispatching people to pick you up at your actual door!

Re: Driverless Hype Collides with Merciless Reality

#208

Post by dgreensp » Thu, Sep 13, 2018, 9:18 PM UTC

photos_victim wrote: post #189

100ideas wrote: post #167

Earlier quoted context omitted.

Does anyone have back-of-the-napkin estimates of total passenger throughput for central public transportation vs ride share transportation for any metropolitan area? If public transit i.e. Muni serves 10x or 100x riders each day vs ride sharing than I see your point... I think. Wait. Maybe you could spell it out?

Don't bother with napkins. The data is open. TNC trips average about 150k / weekday [1], BART carries about 415k trips / weekday [2], and Muni carries about 750k [3]. Thus, public transportation carries about 8x more people, with the caveat that not all BART journeys start or end in SF. On the other hand, 99% of vehicle trips in SF are either private cars or TNCs. Neither of these was my point, though. My point was t…

If we’re comparing a bus or train to a Lyft or Uber carpool or line of some sort, wouldn’t you expect it to be more cost effective to drive people around in buses rather than cars, even if the app coordinates things so that there are two people in the car? It sounds like buses are not very cost effective.

I think what’s actually happening is Uber drivers are paid peanuts and provide and maintain their own vehicles. There’s a technology innovation (eg the efficiency of knowing if and where someone needs a ride) coupled with a labor innovation in paying drivers less than minimum wage.

Re: Driverless Hype Collides with Merciless Reality

#209

Post by linuxftw » Thu, Sep 13, 2018, 9:22 PM UTC

Fricken wrote: post #202

jimmy1 wrote: post #152

Earlier quoted context omitted.

> Driverless cars are coming along just fine. I am curious why you believe this to be the case? Driverless cars have yet to even come close to being workable in adverse conditions, to my knowledge, please correct me if I am wrong. > This is mostly about sensors and geometry. And mostly about the limitations of sensors, and the limitation of available algorithms to overcome geometry. A algorithm a driverless car uses…

The infrastructure driverless vehicles rely on is their own internal maps, which have lanes and signage marked out. Being able to visually detect these things is important, but it is a measure for added redundancy. Great strides have been made in using machine learning to filter out obscurants such as snow. Perception for autonomous vehicles is effectively a solved problem. The biggest technical challenges are relate…

> Great strides have been made in using machine learning to filter out obscurants such as snow. Perception for autonomous vehicles is effectively a solved problem.

Great strides is not 'perfected' and I don't think perception for autonomous vehicles is a solved problem whatsoever. What is a driverless car to do if it slides out of position on the road and is now facing the wrong direction? Does it have the ability to know whether it needs to call for help, can safely maneuver, it's occupants are in immediate danger and should exit the vehicle (or not exit the vehicle)?

Re: Driverless Hype Collides with Merciless Reality

#210

Post by Fricken » Thu, Sep 13, 2018, 9:24 PM UTC

salty_biscuits wrote: post #192

Fricken wrote: post #107

Earlier quoted context omitted.

From Sterling Anderson, former director of Autopilot who oversaw the implementation of HW2: >Perception is a game of statistics.Crudely speaking, if we have three independent modalities with epsilon miss-detection-rates and we combine them we can achieve an epsilon³ rate in perception. In practice, relatively orthogonal failure modes won’t achieve that level of benefit, however, an error every million miles can get b…

He is basically just saying that the errors need to be independent, so you can get that with different capture mechanisms. I think that is wrong. For example, small, fast and poorly reflective will have similar detection performance for lidar and cameras. I think that the real problem is getting consistent statistical estimators and fusing them at a useful update rate. Has anyone seen the lidar from the uber crash? I…

The radar, Lidar and cameras on the Uber that struck Elaine Herzberg all detected her, it wasn't a sensor failure that led to her death, it was a software failure, the autonomous OS tuned to not respond to positives below a certain threshold.