Bicycles being forced off roads?

>>1082750
>This is interesting. It seems that in Europe the companies developing self driving cars are attempting to keep bicycles off the roads because they aren't compatible with robotic cars. Who do you suppose will win in this fight--big multi-national companies or a bunch of poor tree hugging hippies?

Yep. This is coming. We live during the golden age of cycling

But maybe the bike lobbies and local rider's groups (which seem hyper-organized because it's all autistic guys in their 40's with good careers) could turn this development into a massive paved trail network closed to pedestrians ?

We can dream at least

>>1082754

I think the real solution will be some kind of transmitter on your bike. Like a GPS + backup system. The cars will know where your bike is and avoid you

The last thing is that a car could easily be programmed to stay out of a bike lane

I've ridden down lanes were cars are going 50-60mph next to me and it's horrifying. Once in a while one is OVER the line passing me.

I would trust a machine to do a better job than humans in that situation.

So maybe the solution would be more bike lanes on the sides of roads

>>1082756
There's a lot of situations where the self-driving car would respond better than a human--especially if the human drunk, crazy, or so old they can barely see the street. But I've had times when I've been driving and only managed to see some biker shooting through a stop sign out of the corner of my eye. Cyclists will have to be MUCH more careful and follow the rules.

>>1082750
The nice thing about self-driving cars is that it takes a while for the human to take back control after you've demolished their windshield with a crowbar.

>>1082755
This is a human's world. Machines should adapt to humans or get the fuck out.

>>1082764
>This is a human's world. Machines should adapt to humans or get the fuck out.

The majority of humans don't even ride bikes and would trade your hobby for their convenience for much less than the benefit of a self-driving car

>>1082765
>hobby
0/10 apply yourself

>>1082750
OP, you need to GO BACK: >>>/b/

>>1082766
>0/10 apply yourself

Wow, edgy.

Time to get a mountain bike if you want to live

>>1082750
you just pulled that out of your ass. not self driving cars have problem with cyclists but everyone else.

>>1082777
Geez. Do you have trouble with someone posting a link to a news article that might be relative to some people on this forum? Go back home and stick your head in the sand.

>>1082801
Have you read the article?

>>1082828
it's a bunch of nonsense. from what i know about self driving cars they can track cyclists a hundred times better than humans.

>>1082755
>I think the real solution will be some kind of transmitter on your bike. Like a GPS + backup system. The cars will know where your bike is and avoid you

>just put a tracking device on yourself at all times, it's the only way to be safe!

i was gonna take the piss out of you, but then i remembered that i'm the only person in the developed world that doesn't own or operate a smartphone.

>>1082762
>tfw we will have renegade road cyclists
>>1082754
>could turn this development into a massive paved trail network closed to pedestrians

It would kill bikepacking/touring since remote places would be unlikely to be included in this network

>>1082865

wow you're so cool

>>1082801
Didn't Tesla actually said they have a problem with tracking cyclists?

>>1082755
>I think the real solution will be some kind of transmitter on your bike. Like a GPS + backup system.

The problem with putting an electronic system on bikes is that there's no way to make sure that everybody uses one or that the batteries won't fail. A better solution might be something passive like a reflector or a big RFID card that absorbs energy from the cars transmitter and responds. In any event, autonomous cars are coming and there will be changes.

Think about this: What is stopping car manufacturers from purposefully making it dangerous for cyclists to be on the road with their self-driving cars?

Nothing, essentially. With some PR they could floss over that it's just how the technology works and if you are trying to stop it you're stopping progress. Just buy a car.

>>1082977
dunno i seen googles demo about it and it could even predict what the cyclist will do when it didn't have it on the sensors from statistical data.

>>1083062
>big RFID card
size has nothing to it. a small rfid sticker is just as effective as it gets if it's not active rfid. range would be a few meters. which absolutely sucks balls. rfid protocol is also pretty bad about reading all stickers at once by multiple transmitters. imagine the chaos no car would know to which query a tag replied and if the tag belongs to the bike it tried to query or an other one before it.

rfid is a fucking mess overall the technology has more limitations than uses outside very specialized and limited scenarios.

>>1083074
>What is stopping car manufacturers from purposefully making it dangerous for cyclists to be on the road with their self-driving cars?
well the driver will be responsible for the car for a good while even if it's in self driving mode. until that changes the customers will not buy a car that routinely endangers people since they are liable.

>>1083081
This could be the case. But as I see it, people don't like liability so buying a car that they don't control but which they're liable if it hits someone won't be popular either. It's more likely that the company will adopt some kind of black-box recording shared liability setup, and so their motivation would be to minimise liability- and if they can get more people to buy cars as a side effect it will turn into the primary motivation.

>>1082750
I think it might be resolvable by installing some chips onto bike? Or just separate grade traffic

>>1082865
Nah, if the system is done without memory function and without ability to send its signal to remote server, then it won't be able to be use as tracking device

>>1083095
let cars self drive on highways first, no bikes are allowed on highways and it's boring as fuck it would be nice to relax a bit less tiring on long journeys.

>>1083098
That's the step they are taking

After WWII, the US automobile manufacturers (lead by GM) managed to remove streetcars from most cities in favor of buses. They had a lot of money and power and weren't afraid to use it. Now days they have ever more money and power. Every auto manufacturer in the world along with Google, Apple, etc. are working to develop autonomous cars. It will happen and they will establish the rules. I'm sure that bicycles won't even be a minor annoyance to them.

We need to build self-driving bikes as a counter.

File: carshole.jpg (8KB, 261x140px) Image search: [Google]

8KB, 261x140px

Fuck cages. Bring back ubiquitous predictable rail.

File: narrow-streets[1].jpg (58KB, 409x500px) Image search: [Google]

58KB, 409x500px

There may be a shoving match at some point, probably based on insurance and liability, but any car that can't handle bicycle traffic also can't handle pedestrian traffic, and a good chunk of europe in general.

>>1082757
>>1082756
Right, remember that we're comparing it to the current status quo, which is already terrible in most of America. Video related.

https://www.youtube.com/watch?v=Dvs3xrO9_gw

>>1083074
Liability issues. And being MORE dangerous or reckless around peds and cyclists isn't ever going to be a design priority. It's more like not caring.

>>1083079
I think the problem is a little easier because even knowing that there's ANY cyclist around you helps with detection. It's not as important as if there's five or one, as long as the car knows to be more conservative there.

A side benefit is automated traffic lights.

>>1083095
Grade separation is the gold standard, but cost and space make it unworkable.

Cyclists, take to the seas!
https://www.youtube.com/watch?v=NlRmvTebbxI

>>1083077
>could even predict what the cyclist will do from statistical data
"Act like a complete moron".

Anyway, not Tesla themselves, but still:
http://fortune.com/2017/05/29/tesla-danger-cyclists/
http://road.cc/content/news/223386-never-use-tesla-autopilot-feature-around-cyclists-warns-robotics-expert
Cheers.

>>1083268
tesla will catch up, they don't have the expertise of the big players yet. but knowing musk that will not stop him.

File: jeremy-corbyn_1.jpg (39KB, 620x387px) Image search: [Google]

39KB, 620x387px

>>1083074
>What is stopping car manufacturers from purposefully making it dangerous for cyclists to be on the road with their self-driving cars?
among many other things, the 90000000000000 lawyers riding 10k bikes

if a self driving car can't track a cyclist how can they track pedestrians jay walking? or an animal crossing the road?

>>1083356
I am not an expert in the subject but I think it come down to
- Pedestrian are not suppose to walk along with cars, so whenever a pedestrian is detected on the road in front of you or approaching the space in front of you, then you know there's something wrong and you need to stop or change direction. However, it is normal for auto traffic and bike coexists, so they cannot simply treat bikes as obstacle, and must learn to deal with it.
- Also, bikers aren't always biking in a straight line on the road. Sometimes they would also switch lanes or change in speed and such. Nowadays drivers react according to visual clues from the movement of those bikes, but I imagine it would be hard enough for car's system to identify those visual clues.

>>1083376
>but I imagine it would be hard enough for car's system to identify those visual clues.
it's a rapidly evolving field that didn't even exist 5 years ago. when self driving cars become mainstream it will be solved long time ago. it's possible the todays usual attitude for biking will be gone by then like going rapidly on the sidewalk switching to a pedestrian crossing ignoring the red light and not even slowing down then turn off it and join the traffic. this sort of behavior would indeed be confusing to self driving cars.

>>1083406
No matter how this is initially set up, there's going to be accidents at first and a lot of lawyers involved. Not only are bicycles a problem but these automated cars are going to have to use the same streets as cars with human drivers. Should be interesting times.

So I suppose they aren't compatible with slow moving motorcycles eh?
What a load of bullshit

>>1083719
they are already doing it and most of the accidents are caused by drivers ramming the self driving car from behind because nobody expects anyone to actually follow traffic rules.

>>1083964
*to the letter

>>1083719
>>1083406
There may be accidents at first. However as long as the automated car doesn't break traffic rules and the cyclist manual car is the one at fault there really isn't much to fight over. If it's the automated cars breaking traffic rules and being at fault you can expect massive settlements and court cases.

>>1083969
i fucking love this automation boom i'm an engineer working in the industrial automation business and it's awesome. but what will it leave behind is probably not so good for most people. lot's and lot's of jobs that our fathers took for granted will not exist in a few years. professional drivers will be gone soon enough. on most roads you will not be allowed by the law to turn off the drive assist in a decade. you want too drive manual? get off the public roads!
this will be the natural consequence of almost all accidents will involve manual drivers including bikers soon enough.

>>1084331
You say it like it's a bad thing

>>1084332
it will hurt a lot of people who can't keep up. it's very good for me, nice challenge at work good money. we try to make our systems complicated so lot of people are needed in maintenance and administration. also better business more support hours and all.

>>1082762
>takes a while for the human to take back control

Why? That shouldn't affect the autonomous pilot in the least. The car will continue with no weaving or flinching.

This will give the human time to draw, aim and fire.

>mfw first to get that multi-million "hit by an auto tesla" lawsuit

>>1084331
>on most roads you will not be allowed by the law to turn off the drive assist in a decade

maybe after trump and some psychotic faggot control freak like zuckerberg becomes dear leader

>>1084487
Or you know, someone looks at traffic accident statistics.
There are 2-3 9/11s worth of deaths on the road every year. Thousands and thousands of deaths, and tens of thousands of injuries, many of which people never recover from. (And factor in health insurance which has a fun habit of going 'oh you're not going to recover? Well in our terms and conditions it says that if you're actually sick we don't cover you').
That makes it easy politically to ban kill boxes from the roads. It'll be like seatbelts.

Besides, there won't even BE a media counter-narrative of 'but I like driving', because the auto industry will be selling MILLIONS AND MILLIONS of cars with self driving technology.

Lifelong cyclist here.
Bicycles really do not belong in the road.

Closed courses are fine.
MUPs are okay-ish - though pedestrians don't really belong on them.

>>1082750
Just fuck everything about this. If your "self-driving" car can't handle every day situations that human driver handle with ease, the solution isn't to outlaw things that are hard for your car to handle. It's to make your car better. If your team sucks too much to do that, tough titties. Find another job.

>>1082977
Yes. Quoting Elon from a meeting that a friend sat in: "Cyclists aren't our customers, I don't give a shit".

Tesla Autopilot gives me the willies.

>>1083077
Yep. Google has about an order of magnitude more data than any other self-driving car organization out there. Recognizing cyclists and predicting their path has been trivial for them for a the past ~2 years or so.

Driving on public roads is a never-ending stream of edge cases. If you can't recognize and deal with cyclists, I don't trust that your programming team has thought enough about the problem. Can you deal with road construction? An un-mapped temporary stop light at a construction zone? A construction worker with a hand-held slow/stop sign? A crosswalk guard with a handheld stop sign? These are all things that your system will have to regularly encounter.

>>1083352
Lolno. There's so much fucking turnover on the autopilot team. Morale has been through the floor for a loooong time now. Musk micromanages the hell out of it and fires critical team members on the regular when they have the guts to tell him that what he's asking for is unreasonable.

>>1083133
>hey had a lot of money and power and weren't afraid to use it. Now days they have ever more money and power.
Do they? Isn't the automobile industry around the world in sharp decline because people simply aren't buying/can't afford new cars like they did 20/30 years ago?

>>1084511
2016 was a record breaking year for car sales...
http://www.businessinsider.com/2016-was-a-record-breaking-year-for-global-car-sales-and-it-was-almost-entirely-driven-by-china-2017-1

File: ludwig_xiv_ap_2017.jpg (190KB, 2064x1370px) Image search: [Google]

190KB, 2064x1370px

>>1082750
Most people will stop buying cars once they are self driving, especially in urban areas, nobody wants to pay for something he cant use himself. The only good market for them will be car rental systems on subscription base. If you need a ride you order one to your door. Bikes will get their own regular lanes everywhere because self driving cars are far more efficient and need less space to move around.

File: cb54158d8954ff3afbbb2892b2c7e.jpg (394KB, 2400x1480px) Image search: [Google]

394KB, 2400x1480px

>>1084561
Also there will be less cars around because they will be used far more efficiently, unlike now, standing around most of the day.

>>1082756
>I've ridden down lanes were cars are going 50-60mph next to me and it's horrifying
yeah, try cars and buses. living in a 3rd world shithole without bike lanes i do this pretty much every day

>>1084561
>self driving cars are far more efficient
i don't get where this idea comes from
the problem with cars is their size and weight
they're inefficient with space because a car is much larger than its cargo
they're inefficient with fuel because they weigh much more than their cargo
self driving cars are just normal cars with computers jammed in them
they're just as heavy and large as normal cars
they have similar acceleration and braking profiles as with human drivers
they still take up space when not in use
they're still loud, smelly, and hot when idling

>>1084653
Compared to normal cars.

>they still take up space when not in use
Not if they are regularly used and constantly on the move, which will be the case with driver less cars.
>they're still loud, smelly, and hot when idling
not electric ones
>they have similar acceleration and braking profiles as with human drivers
Only as long as there are still human drivers around. They have to swim along.

>>1084491
most people are not in love with their car and driving it's a tool a means to an end. i think anyone that really loves driving is already racing in some form.

>>1084889
Not really.. there are lots of guys that are into modifying/fixing/restoring cars with no interest in racing them.
>most people are not in love with their bikes and riding it's a tool a means to an end. i think anyone that really loves riding is already racing in some form.
Compare to this.

>>1084911
>and driving

>>1084496
>every day situations that human driver handle with ease
Yeah, about that.

>>1084653
>>1084659
Autonomous cars are more efficient only if there's a whole, complete network of them handling all of the wheel-based traffic. A few of them, or even a lot, won't really change a thing if there's still "normal" cars around, with retarded humans operating them.
I imagine city traffic composed of self-driving cars would resemble something more like a factory assembly line than your normal day on the street, but whatever.

>>1085037
>Autonomous cars are more efficient only if there's a whole, complete network of them handling all of the wheel-based traffic. A few of them, or even a lot, won't really change a thing if there's still "normal" cars around, with retarded humans operating them.

We're only at the very doorstep of self-driving cars and yet these few cars with early algorithms, surrounded by crazy monkey drivers are already safer per travelled mile.
Imagine how it'll be in 20 years...

>>1085127
He wasn't talking about safety, he was talking about overall efficiency- less traffic jams and things like that. Smoother traffic, mostly gained from the stop and start time reductions.

>>1085141
in theory cars can follow each other more closely with self driving removing human reaction time from the equation, this allows convoys to reduce power consumption by a significant margin as cars can follow inside each others wind channel.

>>1084659
>Not if they are regularly used and constantly on the move, which will be the case with driver less cars.
but they won't be
look at when normal cars are used
people all go to work around the same time in the morning
they all go home around the same time in the afternoon
at night and during the day, almost nobody is driving around
if self driving cars are going to move everyone then there have to be enough for rush hour
and if there are enough for rush hour, most are going to be parked for the rest of the day
>Only as long as there are still human drivers around
there will always be human drivers around
imagine the gov taking away everyone's car
and the only people who could have one were people who could afford autonomous
or people dumb enough to pay for taxi service
that would not go down well

>>1085127
Roads evolved from wagon tracks and as the vehicles changed, the roads adapted too with smoother paving, stop lights, and traffic rules. I would guess that autonomous vehicles will ultimately have a major effect on the ways streets and highways are constructed. Even manual drivers will be affected. For instance, speed limits will be a thing of the past as cars will automatically control their top speed. Stop lights might become a signal on the dash. And all cars will be tracked.

>>1085373
what i would like is if cars had a radio kinda like a digital walkie talkie where you can talk to some dispatcher who can connect you to a police officer if needed or advise the other drivers if you report some violation. you know kinda something way faster than dialing the police waiting to get a line giving your personal info and reporting something which you know they can't do shit about because they can't talk to the driver and can't spare the resources to hunt him down just because he was spotted passing cars in the emergency stop lane.

>>1085448
Would you be okay with people reporting you every time you run a stop sign or red light on your bike? People hate cyclists for a reason.

>>1085448
>digital walkie talkie where you can talk to some dispatcher who can connect you to a police officer if needed
you just described a cell phone

>>1085487
no you have to dial with a phone, it's dangerous to get ahold of then you have to wait a lot to connect and have to go through a lot of loops before you can begin to describe what you see.

instead just press a button on the steering wheel and when the light shows it connected to the dispatch you start talking.

the device ids the car automatically and gives position bearing and speed to dispatch, so you don't have to deal with all the crap like what's your name, where you are, where you live, what is your insurance number, your mothers name, your dogs name, is it a big dog in case we have to shoot it later? etc...

much less intimidating.

the difference is best described by sending an email to a company trying to get ahold of the right address that doesn't go straight to the spam folder phrasing it waiting 3 days for a response and knowing some imbecile pajeet will answer you who has no idea what's going on and has no understanding of your real issues because he doesn't really speak english, vs just asking a quick question on the online chat.

>>1084495
Here we see the effects of internalised cagerism. It really is tragic.

>>1085475
Their reasons are utterly irrational.

>>1085475
well it's really hard to report bikers as they have no license plate.
>hallo sir what's your emergency?
>some faggot cyclist just run a red in front of me.
>i see can you tell me anything about him?
>he rode a bicycle wore tight faggot pants and some faggot glasses and helmet.
>i see... well anything unusual that stands out?
>he had a faggot blinking light too under his fagseat.
>...

>>1085475
No they don't.

I run red lights when there aren't any cars coming because the red lights won't switch unless I go over and press the pedestrian button, and fuck you.

>>1086588
I run red lights when there aren't any cars coming because there aren't any cars coming

>>1086567
It's really hard to report pedestrians as they have no licence plate too! I think you might be on to something…

>>1086567
It's because the potential for damage is so low. Except for rare cases, the only way a cyclist can really fuck someone is by 'forcing' a car to fuck someone on their behalf. (As in, they pull out and a car swerves into a kindergarten.)
Beyond that, someone in a car can crash into people/property and still roll away fine with great speed.

If a cyclist hits something hard enough to where it actually matters, you'll probably find him crumpled close by on the ground.

>>1082865

And yet here you are posting from another electronic device...

>>1082750
>Ghosn worries that driverless cars have a cycle-shaped hurdle to leap: "One of the biggest problems is people with bicycles. The car is confused by [cyclists] because from time-to-time they behave like pedestrians and from time-to-time they behave like cars."

There's your problem. Just make up your mind and everything will be fine. This is a problem for regular drivers too.

>>1087511
Cyclists killed two people in Berlin, 2014. They were old, but it's not impossible.

>>1091153
I dont give a shit. Doesnt add anything.

>>1091150
No, that's what a cycle is. You dont have to chose because someone decide you're king of the road.

>>1082874
>renegade road cyclists
imokwiththis.jpg

>There are many who believe that AVs will be great for cities, that "with the right planning, they offer the potential for a better quality of life, economic growth, improved health?
How exactly does an autononymous murder cage make your healthier? You're still sitting on your ass in traffic sipping on sugary drinks and eating doughnuts

>>1091979
in theory it can lead to better organized more fluid traffic with less pollution and way way fewer traffic jams that resolve exponentially faster.

you can also relax more compared to driving. basically it's like getting up to a few hours in extra each day to watch some video read listen to music relax sleep whatever.

self driving cars can drop you off near where you want to be and go find a parking place on their own, then you can call them from your phone much like how you order a taxi except it's easier it will find you by gps not words.

it's like having your own personal driver for free.

two things i await very much: self driving cars at least on highways but cities would be even better and electric cars retaking the roads completely from the gas guzzlers.