Hacker News new | past | comments | ask | show | jobs | submit login
Self-Driving Cars Have a Bicycle Problem (ieee.org)
68 points by type0 on Feb 25, 2017 | hide | past | favorite | 110 comments

"Now Johnny, when you ride your bike you must wear your I-am-a-bike vest and follow these patterns or the cars are likely to kill you." :-)

Summary is that "people riding bikes" (PRB) is a much denser image set than "people in cars" (PIC) and "people walking or jogging" (PWJ), and the PRB objects have a much higher dynamic angular vector capability (they can change direction extremely quickly) combined with a wider dynamic velocity vector to PWJ they strain the ability of the predictive filters to reliably asses the collision threat. As a result you need either faster/better hardware or better algorithms to deal with that particular group.

I'm not particularly surprised by that, I expect the cars to also end up with small animal issues, as they can appear suddenly and aggressive evasive maneuvers to avoid hitting them may injure passengers at the cost of saving a squirrel's life.

And all of that adds up to some of the many things that one has to think about when claiming victory here. It is going to be a long hard engineering slog to get to full autonomy. My question is you can build a computer that can drive a car fully autonomously, what other missions could you create for it? Some of those are kind of scary.

This is when the humans strike back against the machines.

Every example of a NN I've seen is good at one thing, the thing it is trained for. I haven't yet seen a NN that can beat people at chess, recognize faces, drive, and make coffee, or more importantly, decide when to do one or the other. The closest thing I can think of is Watson from IBM.

People can drive (poorly) but they also recognize the value of life in a biker or a squirrel, and in some cases, override their usual learned behavior, for example, instead of continuing to drive, swerving, basically improvising.

Computers and programs again are good for what they have been input and now, what they already have learned. The synthesis part, I'm not quite sure even NN have reached that part yet. I may be wrong, the answer is probably "learn more" or "learn faster" (which is what you suggest), but it's easier to synthesize like this when you have a mind with general knowledge, at which point the mind is less like a NN and more and more like a person.

There's a saying that neural networks are the second-best way to do everything. Don't be disappointed when their trajectory flattens out.

And, as for their role in autonomous vehicles, I don't think they play a primary role (although I don't design autonomous cars so I wouldn't know for sure.) You can see in this Tesla video that a lot of the computer vision isn't even reliant on neural networks: https://www.youtube.com/watch?v=BLlwm5Dq7Is

Why do you think that video shows that Tesla isn't reliant on neural networks? Neural nets outputting bounding boxes is pretty normal.

Despite NVIDIA's papers, end-to-end neural nets for self-driving cars are dumb, but for the initial perception based on sensor data, neural nets are the only game in town.

I was looking at the optical flow indicators and the road-tracking. Afaik any NN solution to those are immature if they work reliably at all. Your probably right that that the bounding boxes are probably NNs.

There's a saying that neural networks are the second-best way to do everything. Don't be disappointed when their trajectory flattens out.

Outstanding quote! I'm going to reapply it.

Frankly you're overstretching. If you've ever driven in India you'd be questioning the very basis of humanity.

> The closest thing I can think of is Watson from IBM.

What are you imagining Watson is? Watson is not a singular entity that decides whether to play Jeopardy or analyze medical records or whatever. It's a brand.

The comparison that comes to mind is Colonel Sanders.

Sometimes with the accidental (and non-accidental) encounters between bikes and cars show that the value of the life of the biker isn't really valued that highly.

People often take the minimum as the representative statistic. The article implies the mean event for self-driving cars is much worse than that for human drivers.

Yes, I bike, so I know the worst of the worst. I also remember a few courteous encounters, but probably many fewer due to negativity bias.

I think NNs would work well to solve the bicycle problem with some more data -- maybe a lightweight sensor of some sort on bikes, or a device that could communicate on a public VANET - That would be a safe long term solution, but obviously many years out.

I haven't yet seen a NN that can beat people at chess, recognize faces, drive, and make coffee, or more importantly, decide when to do one or the other.

These are the sorts of things humans struggle with too. It could be argued that we, also, don't have one neural net that does all these things but rather a variety of specialised brain regions. It would then make sense that a computer might run specialised programs to deal with different tasks.

Anyway, self driving cars and bicycling can be fixed with a simple engineering solution: separate the two with barriers. This is what we should do anyway.

As someone who had their bike destroyed in a high speed car vs. bicycle collision, and was lucky to walk away with a broken thumb and sprained ankle: cycling on roads is insane.

> cycling on roads is insane.

How is it that people in developed countries across the world have learned to do this, and we haven't?

That said, the relative health benefits of cycling outweigh any risks 7-to-1 for an average human. Cycling for an hour, even with American road infrastructure, actually results in a net increase in the average person's lifespan of over an hour. Seems like maybe it isn't so insane after all.

Do you have a source for your claims in the last paragraph? It would be an interesting read.

There I googled it for you, so spend at least 10 minutes reading it now. Please.

But it is common sense, we know that excerise prolong your life, doing it regularly in modest proportions is best. Using a bicycle for transport to work means you are exercising twice daily, maybe 10-20 minutes for lunch as well. It is obvious that this kind of exercise will do you good.

For me it's one to two hours of exercise per day, it's hard to beat those health benefits.

http://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2015.3... https://www.uu.nl/en/news/dutch-bikers-live-six-months-longe...

I haven't yet to be hit may be out of luck, but in my city, I don't have the option to be separated if I want to get everywhere. Instead, there are a plethora of "share the road" signs everywhere.

I'm okay with the separation if I get the same access as everyone else...and I am not blocked by slow bikers. Of course, I don't need to ride on highways/freeways (nor do I want to).

> Summary is that "people riding bikes" (PRB) is a much denser image set than "people in cars" (PIC) and "people walking or jogging" (PWJ), and the PRB objects have a much higher dynamic angular vector capability (they can change direction extremely quickly) combined with a wider dynamic velocity vector to PWJ they strain the ability of the predictive filters to reliably asses the collision threat. As a result you need either faster/better hardware or better algorithms to deal with that particular group.

Nice analysis of why bicycles are so dangerous even now - computational issues like these are just as much an issue for brains as for computers. Not surprising that, in many cities in the US, biking is insane. Nearly everybody I know that bikes to work has a close call or two in their history.

> Nearly everybody I know that bikes to work has a close call or two in their history.

Nearly everybody you know has had a close call or two with driving, I'm sure. And not just an "almost a minor fender-bender where everyone would have been okay" close call either. Cars are one of the leading causes of death for people of all ages. I've had close calls with cars as a pedestrian, as a cyclist, as a motorcyclist, and as an occupant in cars. I've been hit by a car while on a bicycle.

Cars as an entire category are dangerous, as are the humans who operate them.

It's rather funny that the cyclist who rides downhill at 40mph outside my home every day classifies cars as the hazard and not himself, since he must ride on the road and not the footpath. He's no different to the guy who was riding at a similar speed on the footpath outside of my work. Apparently that was OK, you see, because if he was on the road he would have been riding the wrong way on a one way street and there was no way he was going to ride around the block to get on the southbound road. He nearly hit me every day after work for a month, because he was cutting blind corners at speed.

What about the two who were occupying a whole traffic lane, weaving back and forward to stop cars overtaking, while riding at under 15mph in a 40mph area? They'd start screaming like lunatics at anybody who legally overtook them - the cops wanted a word with them, but they disappeared when they saw me on my phone.

Five or six times a day, cyclists run a red light at a busy intersection in town. I see many near-accidents in my fleeting times in that intersection, all of which are caused by the cyclists - and I'm not counting the number of times they've nearly hit pedestrians crossing with the red light. I've seen one car run a red light, three years ago, since I started crossing there in 2012.

Years ago, I was overtaken by a cyclist and his three children, just as I had begun my turn. I nearly ran over the father, and they took their sweet time to move through the intersection a car coming the other way (who legally had the right-of-way) had to stop and wait for them. I could have got out of my car and walked, and I would have got to the other side before they did. They bumbled along as if they'd done nothing wrong.

It's not uncommon for cyclists enter intersections on foot, and so legally become pedestrians for the duration of the crossing, and then jump on their bikes and dodge through traffic while hurling obscenities at vehicles who have right-of-way because they didn't enter as pedestrians. The law here doesn't change your status because it's convenient for you.

Back in 2013, one of my ex-workmates told me quite a tale, about the time she pulled up at a set of lights. When she moved off she signaled to change lanes and only noticed by chance that there was a cyclist holding onto her car to get a free ride. He couldn't have seen her signal, the indicator was near his foot, and if she had turned he would have toppled in high speed traffic, perhaps ending up under her wheels.

Back in the early 1990s, a friend of mine was pulling out of his garage and was hit by a cyclist at some speed - the cyclist threatened to call the police, but stopped short he would have been charged, not my friend.

Don't lump it all on drivers, cyclists are dangerous, too.

I'm not sure how throwing out a bunch of anecdotes about people doing stupid things that might injure themselves counters the notion that people doing stupid things in multi-ton vehicles traveling at tens of miles per hour regularly injures and kills an unacceptable number of people per year.

As a cyclist, I hate seeing cyclists flout the rules like this because it makes the people driving those multi-ton metal weapons lose respect for all cyclists.

However, dangerous on a bike is not even in the same league as dangerous in a car.

While clearly inexcusable, how many of those incidents of bad cyclist behavior do you imagine could have ended with a fatality or catastrophic injury?

Is a diferent type of danger.

> how many of those incidents of bad cyclist behavior do you imagine could have ended with a fatality or catastrophic injury?

Several. My neighbour had around 75 y/o when a cyclist guy passed too close and broke her hip. Result: A long post-operatory, big scar in her leg, and she never could walk properly again. Died a few years later.

That sucks... Yes I wasn't thinking about the frail aged. I'm assuming she was on the sidewalk. What a horrible person to cause an elderly person to fall.

Of course, a car would have done much more damage to her in the same situation, and there are many more instances of real injuries from cars to cyclists than cyclists to pedestrians.

Did the cyclist actually make contact with her or just pass too close?

I'm not condoning the cyclist's actions in any way, just asking.

Should we all list personal anecdotes of the stupid stuff we have seen motorists do as well?

The difference is a "close call" on a bike has a far higher chance of being lethal than a "close call" in a car.

Not if you're the unfortunate pedestrian or cyclist who finds themselves on the wrong side of the "accident with a car" equation.

> "Now Johnny, when you ride your bike you must wear your I-am-a-bike vest and follow these patterns or the cars are likely to kill you." :-)

Um, this is true for humans, as well, you know. Every single person I know who decided to start commuting to work on bicycle in the US was in the hospital within a year.

After training, I suspect that computers are going to be MUCH better at avoiding bicycles. They're also going to be much better at avoiding the stupid things that bicyclists do, as well.

Maybe the US is different, but I've been commuting on my bike, on public roads in a major city, for literally two decades.

Admittedly I've had a few accidents over the years, but none that drew blood (as far as I remember), and certainly not yearly.

Either you live in a dangerous place, or you know bad riders :-)

What do you think about large animal issues? In some rural areas collisions with deer, moose, and elks are a serious problem and aggressive evasive maneuvers may be necessary. Are current autonomous driving systems able to recognize them quickly and accurately?

I sometimes wonder how autonomous cars would fare in very rural areas. In the country, there's usually nothing on the road at all. But when there is, it can sometimes be something strange like, as you said, a moose, or a piece of plywood, or a slow driving tractor.

When I was a teenager growing up in the rural Midwest, I drove a small sports car. I always wanted to see if I could fit under one of these things (a crop sprayer) when I came up behind one on the highway:


I wonder if this kind of vehicle would even register on some cars' systems, or if autopilot would just charge full speed ahead underneath to my teenage heart's content.

Re your last point, this leaped immediately to mind.


David: The Robo-Cup Federation hope that by 2050 a team of fully autonomous humanoid robot soccer players will win a soccer game, complying with the official FIFA rules against the winner of the most recent World Cup of human soccer. That terrifies me. If they can run around and kick a ball and beat us at football, what else can they do?

John: Bring us breakfast in bed!

David: Yes, and then strangle us.

I don't understand this logic at all.

We are supposed to be terrified by robot soccer players strangling us in bed? Sure it could happen, but there seems to be easier ways to die.

Apparently one of them is riding a bike around self driving cars.

It is the unintended consequences of things.

One of the better lines in 'Real Genius' (another comedy) is Lazlo's "All you's need is a tracking system and a, a large spinning mirror and you can vaporize a human target from space." where the consequence of the main character building a multi-GW laser is suddenly clear to that character.

I'm all in favor of self driving cars, and I look forward to the day I can use them to commute and work at the same time, but cars are already used as bombs in the middle east, it isn't a large step from self driving taxi to self driving car bomb is it?

I feel I should point out at this juncture, that this was a quote from a BBC Radio 4 comedy, and that it's not to be taken as seriously as you seem to be.

I think his broader point though, as is the point of the person I was replying to originally, that when we design robots for highly adaptable, sophisticated tasks, we design a potentially very dangerous robot.

Right. Horses can beat humans in most running races. Gorillas and chimpanzees can beat humans in wrestling. And yet who's in charge?

Because we can outthink them. We're actively trying to build something that can at least think as well as us.

There are deep cnn models for doing joint recognition detection at around 60fps (at 300x300) at reasonable accuracy 75%.

This is arguably the most expensIve part of the pipeline, but we sHould have atleast two more architectures out by NVDA/AMD before any product comes out.

I was thinking about this a while back, wondering if these self-driving cars were any better at "seeing" bicycles (and motorcycles and pedestrians) than humans are.

I haven't rode a bicycle since I was a kid but I do put ~10k miles/year on my Harley. In the last 10 years, I've been hit head-on twice, ran off the road once, and had too-many-to-count "close calls" by drivers who weren't paying attention (not to mention the friends that been killed) -- and that's on a loud, relatively visible (compared to a bicycle), 900-pound piece of steel. I can't imagine the average bicyclist is "seen" better than I am, especially in busy traffic areas/large cities/etc.

I'm all for anything that can makes the roads safer for those of us on two wheels!

The public policy and PR question I'm curious about - how will the conversation be structured around self driving cars and their difficiences? When someone on a bicycle gets killed, they'll be able to point to the limitations of the system and say "this code can't handle this and it'll kill others - we should outlaw it". While others can point to stats to say self driving cars are safer, I'm afraid the dialog won't be that rational and they'll get banned on a state/city level.

I had this conversation with a friend the other day. "If an automatic car will only detect a red light or a cyclist or a child on the road accurately 99.99% of the time, should it be allowed on the road?" - his answer was "if it's safer than a human on average, then yes". My answer is - absolutely not. System malfunction is one thing, but if there's a non-zero chance of not detecting something that it has to detect to operate safely, then it shouldn't be allowed to be sold.

So by your logic, human-piloted cars shouldn't be sold either, as humans are not able to detect things 100% of the time.

A human is not a component for the car - the car itself is not smarter than a knife or a gun - it's how you use it that defines the responsibility here.

Or to put it in other words - if I run a red light by accident, then that's what it is - an accident. I'm still responsible for causing a crash if one happens, but there's nothing more I can do to improve my perception of surroundings(and we already do ban humans who are unfit for driving).

If a self driving car runs a red light, because its sensor is only 99% accurate, then I'm pretty sure the company making it will be sued to oblivion, because they released something that does not work all of the time(and I am talking about normal operating conditions, not sensor malfunctioning).

Radar and LiDAR are exceptionally good at seeing bicycles and motorcycles and pedestrians. Predicting their movement is a separate problem.

Didn't stop Uber's cars from doing the dangerous right hook turns instead of merging in the bike lane EVERY single. time.


I guess they've succeeded in acting like most human drivers. I've almost gotten right hooked a few times while running (and wearing bright highlighter-colored high-viz reflective clothing).

Waymo's cars do a pretty incredible job of detecting bikers (even erratic ones) - and even understands hand signals

One of Google's demos was recognizing a hand signal made by a bicyclist in front of the car.

I suspect that motorcycles are the harder problem. They aren't that much larger than bicycles. They use the road like cars, which only pushy bicyclists do. And they're faster than cars, and almost as maneuverable as bicycles.

Pushy? Since bicycles are defined as motor vehicles they have just as much right to the road as cars.

Ah yes, but you see I'm in a hurry home to catch Game of Thrones reruns and I had to wait twelve entire seconds for oncoming traffic to pass so I could get around some inconsiderate guy trying to stay fit and/or reducing the effects of traffic.


The sad irony is it's those "pushy" cyclists who stay alive, by making themselves more visible, being a part of typical traffic patterns, and refusing to be pushed off to the side by cars trying to squeeze into the same lane.

Yes, I've been a pushy cyclist, for that very reason. Getting pushed off the road, or broken on opened doors, is no fun.

Yes, it's strange how motorists perceive the cyclist limiting their speed to 15-20 mph for a whole city block when most of the city is crawling at 5-10 mph due to all the cars.

Certain streets in SF are unspoken "highways" with timed green lights to get across the city quickly: Fell/Oak, Sutter/Bush, Geary, Lincoln Ave, 19th Ave, Guerrero, Howard/Folsom (though this seems to have changed with the crazy wide bike lanes).

While it's completely in one's rights to ride a bike on those streets, it's pretty stupid unless you're hauling ass because the speed differential between the car and bike is too great.

Urban planners can talk all they want about how these streets are bad design and cities shouldn't have them, but it's best for both parties if bikers didn't use those streets. I don't know how someone rides on those streets without thinking it's a bad idea because most of the time, it's someone on a slow mountain bike going 10 mph.

They're vehicles, sure, but not motor vehicles. Unless they have motors, anyway. And sure, "they have just as much right to the road as cars". But being right isn't much use if you're dead.

I don't think bicycles are that unique of a computer vision problem. If self-driving cars have a bicycle problem, they probably also have a children problem, a wildlife problem, and a road debris problem that's simply masked by the relatively low frequency of those things compared to bicycles on the road.

Another comment mentions motorcycles, i'm very curious to see the detection data for them: they should have roughly similar characteristics to a bicycle, but at a similar or higher frequency.

Human driven cars have a bicycle problem as well.

Humans aren't very good at predicting or seeing them either in dense urban areas. Grade separation is the way to go!

Human drivers don't even follow the most basic bicycle safety rules, such as merging into the bike lane before making a right turn to prevent "right hook" collisions. From my observations, probably no more than one out of ten drivers does this.

I guess they think it's illegal to drive in the bike lane at all, even though the opposite is true: at least in California you're legally required to merge into the bike lane in the last 200 feet before the intersection when making a right turn.


If you travel (and drive) regularly between 2 or 3 states, it will be almost impossible to "intuitively" remember all these rules. What you describe is outright illegal in my home state, but required in a large city I also split my time in.

Road rules for bikes and bike lanes are wildly different depending on state, and heck - even city. There are states where you have one set of rules within city limits, and entirely different throughout the rest of the state.

It's a huge mess, and doing things like putting unprotected bike lanes to the right of traffic lanes is almost criminally stupid and dangerous. You cannot undo decades of muscle memory ("things don't pass me in the right lane especially on dense city streets") and then point to "there is a rule that says you have to!" when the expected tragedies start occurring.

I used to think drivers and bikers were just "assholes" to each other in the US - then I spent time in the Netherlands and Belgium and realized these two modes of transportation are purposefully being put at odds against each other in the US for political reasons. It's quite possible to combine these modes of transportation into a network that works for both types of users.

By physically separating bicycles and automobiles?

A perspective from a writer who favors separating bikes and cars:


Oregon law actually requires the unsafe behavior, though I expect that has very little to do with how drivers behave.

And, I'd like to see the evidence that shows California drivers are scoffing at this law.

To put it another way, I don't think anyone knows this law.

I don't know if it's a recognition problem, IME it's a not knowing how wide their vehicle is problem.

much as I wish humans could share the road, separation is the way to go. People are too territorial on the road, and drivers and riders alike have too little attention available to reliably avoid each other. The road makes people stressed and leads to mistakes

Did you just make that up ?

Make what up? Ask any cyclist, and they'll be able to recite to you a long litany of horror stories involving human-driven vehicles. I routinely deal with multiple cars driving in the (non-grade-separated) bike lane on my commute every single day.

Yeah, I started commuting via bike recently. 5sh miles, all on busy city streets. Up valencia, then market, then about a mile along embarcadero.

We really need to do something about people parking in bike lanes and in my opinion should completely close market to cars. Allow chariots, public buses etc. Anything with 5+ passangers, but close it to everyone else. There is just too much going on to have that mix.

We also desperately need more grade-seperated bike lanes and bicycle only traffic signals that are aggressively policed.

Sorry if I offended you or anyone else.

I too commute a lot via bicyle and also drive, which is why I disputed your comment. I find motorists often see me fine...when they're paying attention and I'm using common sense, ie lights when riding at dawn, not running reds etc.

We're talking about vision / detection here, not people being unattentive, aggressive or inconsiderate, which seems to be the actual problem with motorists and cyclists.

If a self-driving car is 100% dedicated to driving and still has problems detecting bikes 75% (at best) in optimum conditions, that's a different and important issue that I don't believe exists in human drivers today.

I actually hope for both our sakes that if self-driving cars do take over, it's for our benefit as cyclists.

No offense taken, I just couldn't make heads nor tails of your comment.

I too am quite optimistic that self-driving cars will be vastly better to deal with than human-driven ones. The constant 360 vision and 100% adherence to turn signaling will be vast improvements. Every day I deal with drivers making unexpected turns, and not checking for the presence of bikers before turning across a path that bikers might be traveling in. Self-driving cars won't have these problems.

> Self-driving cars won't have these problems.

My concern is that they will, but policymakers decide a few dead cyclists is a price worth paying. In the UK, we're already ignoring an air pollution crisis (mostly from car fumes) which is killing c. 40k people each year. On top of the KSIs from actual road accidents. Uber have shown all too clearly that some companies will hide or ignore safety issues in this area.

Pro-cyclist advocacy seems to have reached a point of equilibrium against apathy and anti-cyclist sentiment, thus bicycling amongst human-driven cars is as safe as it's ever going to be.

With robot-driven cars, the bicycle problem is an area that engineers can focus on and iteratively improve—essentially forever—even if cycling amongst robot-driven cars right now is as dangerous as cycling amongst human-driven ones.

Really interesting article... This is why I'm considered annoy by some drivers - I take up too much space on the road for them. However, better to be seen and be considered "annoying" than be killed.

The solution is likely a device that cyclists can wear that self driving cars can detect - I'm going to bet the AI for bikes is intractable because they are fast, virtually invisible at some angles and travel in between the traffic in strange ways.

An alternative could be all self driving cars near by sending each other information about cyclists via some means. Or self driving bikes ;-)

> I'm going to bet the AI for bikes is intractable because they are fast, virtually invisible at some angles and travel in between the traffic in strange ways.

If this is true, do you think humans can handle it any better? What you just described would be a nightmare for any human driver as well. Is it reasonable to allow self driving cars that are only marginally safer than human drivers?

These are good questions, I have no idea about answers to them. Turning it around a second, do you think it's acceptable that all self driving cars could be as bad as current humans are?

Not OP, but as long as it's not worse than current humans and still much better in other areas, I personally find it acceptable. But the goal is for it to be better in all areas.

I think self driving cars should be allowed if they are marginally safer than human drivers. If the car would be just as likely to get into an accident if I was at the wheel, why begrudge me the extra time to read or work?

The idea of a "notice me" transponder is an interesting one, but runs into problems with reconciling your own margin of error for location with the receiver's margin of error for location.

It's also already extremely difficult to get people to wear a helmet, not to mention making them use lights, reflective surfaces and having tyres/brakes in good condition. A lot of people just buy the cheapest $15 bike off craigslist and use that. And we would make them wear an expensive transponder than most likely needs charging and maintenance? I don't believe that for a second. Also - what happens if a car hits a cyclist who didn't wear one? Are we sure it's still the cyclist's fault?

Not to mention humans being assholes and using them to abuse autonomous vehicles.

You will be able to do that anyway by making yourself impossible to overtake, or walking out into the street for example; by definition the car would know where the transponder was as well.

This actually raises a good question that may be even more problematic for autonomous cars; people deliberately fucking with them in a way you wouldn't/couldn't before. Maybe this won't happen but who knows.

The question then is who to blame legally for the results of dangerous driving, the human or the car?. Could the police draw a line between both?

It seems like it would be useful if there was some standard for actively notifying autonomous vehicles of something they need to be aware of. For example, a bike light might also transmit a radio signal to the car, or perhaps it includes lights operating with a specific color or pattern that self-driving cars can look out for (maybe it doesn't even need to be in the visible spectrum).

Obviously this needs some thought. For example, how would you prevent people from messing with the cars - maybe it's very short range, or line-of-sight only?

In the end, the "messing with cars" problem renders almost all explicit-communication-between-vehicles moot, or at least not as useful as people first imagine. The autonomous car will still, no matter what, have to trust its 'vision' over any informative messages.

This is not any different from cars driven by humans, btw. Someone can put up a sign (say a merge left sign and some pylons) that can cause accidents pretty easily, which is why drivers still have to pay attention.

The thing is that wide, let alone 360 degree, vision should provide the computer with more and more useful 'messages' than humans are ever capable of taking in. And probably more than any attack vulnerable information messages could provide too.

Yes, though I can't recall any incidents where someone set up traffic signage to intentionally cause an accident (I'm sure it's happened, but I bet it's really uncommon).

Anyway, I'm not suggesting that the car take a remote signal on blind faith, but I still think it might be useful to have some kind of indicator. It can be factored into the other sensor data - "this signaling object does appear to be in close proximity and moving like a cyclist". Just as people don't tend to mess with signage and cause crashes now, I think they're unlikely (on the whole) to do it with these kind of signals. Especially since you'd be in serious trouble if you caused a wreck.

Well, keep in mind how much harder might be to catch someone messing with traffic with radio signals of some sort than by physically being present and visible and putting nonsensical signs in the road. I do think it would be more common.

I say to robots: welcome to the club. A bike on a roadway is a tricky problem for the best of us. They can move quickly, at or above traffic speeds in many circumstances. They dont follow the same rules (stop signs) and at any moment can transform autobot-style into a pedestrian. It's an issue. And when it goes wrong it is the biker who suffers. That is life driving a car. If you cannot get it right 99.9999% of the time, you aren't good enough to drive.

I have yet to see a car capable of reading the body language of a bicyclist or motorcyclist. As one of those, i still want to look a driver in the eye before i trust him not to squish me. Robots arent anywhere near there yet.

And when robots get there I wonder if they will resent cyclists for the computational cost they impose - and if they will find endless ways to rationalise that resentment (cyclists break the rules, cyclists don't pay road tax etc). I cycle and drive and when I drive I hate cyclists and find it very hard not to blame them for it, but it's just because they make driving harder

Really? You hate cyclists when you drive? That seems pretty selfish really.

If I drive a ferrari, should I hate all those non-turbo car peasants for making me have to drive slower? Of course not, I'm using shared infrastructure that everyone has paid for, even those pesky cyclists and slower vehicles.

I'm a cyclist 80% of the time, a driver the other 20%. I pay my taxes and rego on my car and follow the rules, even while cycling, and you still hate me.

I hate our lack of suitable infrastucture with properly separated cycle paths, not the riders.

When you're driving on a shared road you need to deal with slower vehicles and drive safely. If that's too hard for you to deal with, maybe you shouldn't be on the road.

I think self driving cars is a 95% problem. It'll be relatively "easy" to get 95% of the way there. The last 5% will be almost impossible to get right.

Car driving humans is also a 95% problem. It's almost impossible to get people to follow (or set) good laws, pay attention, not be under the influence, etc. I'm betting that cars will improve at a faster rate than human capability.

Yes. But what I said was not a comment on how well (or not) humans can drive now or in the future.

This is a good reason for self-driving vehicles to use high-performance (read as "subject to ITAR", alas) IR imaging sensors alongside LIDAR. Detecting/identifying humans on bicycles (or, more generally, humans/wildlife in/near roadways) can be done a lot more reliably with IR imagery than with just visual imagery.

Are self-driving cars really worse at noticing bicyclists than pedestrians? The first paragraph makes it sound this way, but there's no way it's harder to track them than birds, and it's hard to tell whether the quote from Steven Shladover is making a comparison to pedestrians.

Well, pedestrians don't move as fast or maneuver as well as bicycles. And birds? I've had many close calls with birds. But I've never hit one. And more to the point, there's nothing that I could have done to change the outcome. Except maybe to roll the car, or whatever.

I don't think so, but we build infrastructure for pedestrians, so that there are fewer points where they interact with cars.if pedestrians would have to walk in car lanes (which, by the way, should be called vehicle lanes), they would pose a big problem, too.

Considering the limitations of computer vision, this should really come as no surprise.

Which is why LIDAR seems so essential. Computer vision has made some amazing steps forward but it's still not perfect.

lidar solves some of your depth of field issues, but there's still a challenge in seeing a bike at a relative velocity of 50mph and reacting appropriately.

I wonder what they do about more exotic vehicles like my recumbent tricycle. They’ll see it alright, but I wonder what they classify it as and thus what evasion model they employ.

It'll just be the standard bicycle "kill all humans" program. :)

But seriously, you bring up a good point. Maybe both cyclists and yourself will come up with ways of gaming the system in their favor.

I can't find a link now, but I'm reminded of an experiment where baby chickens would prefer a ruler with three lines painted on it to their real parents.

Perhaps you could have a design on the back of your seat that tricks the self-driving cars into thinking you were a gas tanker truck or similar large, dangerous vehicle, and get yourself a wide berth.

Automatic driving is totally possible, but self driving cars have a huge problem that automated planes and trains don't. They share the road in close proximity to human controlled vehicles and humans.

The huge, insurmountable problem is that people will NEVER accept autonomous transport that regularly kills people and causes accidents, even if the computer driver is better than an average human.

As long as self driving cars share roads with humans they will never stop killing people. Humans are unpredictable by nature, and there will regularly be situations a self driving car encounters where death is just unavoidable.

When self driving cars start killing people what is going to happen? You can't sue a machine. What if the machine has to decide who dies? This is an insurmountable problem. Cars may drive automatically but there will always be a qualified human sitting behind the wheel.

Look at planes, it's a lot easier than driving since there's nothing to run into up there. We've had fully automated flying systems of various kinds for decades. Hell, even the space shuttle could land itself... from space. To this day every commerical flight with humans on it has not just one, but two human pilots.

The other big problem with self driving cars is that hardly anyone needs them. Cars are built to transport people, and it turns out people are pretty good at driving. So the vast majority of the time there is zero need for the car to drive itself. In cases where it would be convenient we already have cruise control type inventions and automated lane following.

So what about trucks? For one, if people won't tolerate a 3000 lb piece of automated steel killing people occasionally they sure as hell aren't going to tolerate a 80,000 pound piece of steel doing the same.

So why don't we make special roads for the automated cars and trucks and hitch them together so they can't hit each other? We've had this for over 100 years and it's called a train.

Basically, automated cars with no humans behind the wheel will never happen. This gets rid of most of the advantage of a car driving itself in the first place. Also, if humans need to be able to take back control there goes all the advantages of packing cars close together on highways and stuff too.

i really want to make a self driving bike. a car is so big and serious and expensive but a bike is so cheap and comparatively harmless. adding retractable training wheels would make self driving considerably easier too. once full autonomy is achieved, it can be sent out to collect food orders and bring them back to your house.

Half of the joy of biking is the riding aspect. For example, I get my exercise for free.

Who said you wouldn't still need to pedal?

Okay, consider this, why would you want to ride a horse if you just press a button and you get to point B from point A?

For frequent commutes where I don't feel like it, I just take public transport. A lot of time, I bike for the fun of biking, although many times it is also for the sake of transit.

ok i don't get it. Why not develop a general obstacle-avoidance system ? simply make the car be repelled by any objects, whatever shape or speed, and make it give more margin for obstacles that have a higher entropy of motion.

Yes, people love to have these endless discussions about how the cars will decide about who gets to live and who dies, and about how people won't want to buy a car that has a chance of making the moral decision to save someone other than them.

Literally the only thing the self-driving car needs to do is be able to stay on the road, and if the road is blocked, use the brakes. There's no moral dilemmas and no grey areas.

if {obstacle} then {brakes = 1}

Detecting what's an obstacle and what isn't is of course a difficult problem, but I'd think it'll be the legal issues that will actually be the largest barrier to self-driving cars. Someone has to be responsible when it fails. If the human isn't driving, then the car maker has to be responsible. This liability will have to be added to the cost of the vehicle.

How about motorcycles?

No they have a problem with people on bicycles not following the rules. A problem human drivers encounter on a daily basis.

A problem human drivers believe they encounter on a daily basis because themselves do not know 50% of the road laws (in the best case).

I encounter human drivers exceding the speed limit on a daily basis as well. Is that not also failing to follow the rules? At least the robots won't do that.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact