Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Will bikes on the roads create a problem for AV's?

This site may earn commission on affiliate links.
Will cyclists feel safer and so there will be more of them?
I anticipate that road bicycling, and running and walking on roads that lack sidewalks, will grow in popularity as the roads become safer due to autonomous vehicles. It will also help that diesel vehicles will be replaced with EVs, thus eliminating some particularly noxious fumes that are commonplace today. In addition, bicycling, running, and walking may then be the only permissible ways to enjoy the open road under manual control.

Will this cause problems?
Having more cyclists, runners, and walkers will be a very good thing, on balance. The US and other nations are currently facing an epidemic of obesity, and a great many people are totally sedentary. Making the roads safer will take away a primary excuse that many have for not getting out and moving. This will improve overall quality of life, productivity, and health outcomes.

Yes, having more cyclists and pedestrians on the roads may cause relatively minor delays for car users. On the other hand, interconnected autonomous vehicles should be able to do better than human drivers at optimizing the use of road surfaces when passing cyclists. I can imagine vehicles driving within 5 cm of one another, width-wise, in order to free up one meter of space around a cyclist that they are passing at speed. (California state law now requires that motorists keep at least three feet away from cyclists.)
 
  • Like
Reactions: ReddyLeaf
I think that other drivers will be annoyed by that AV driving so very, very law abiding. Yes, it follows the rules strictly which in theory is good, but I don't know if it mixes well with people driving as well.


And btw, nice link! As a person living in the Netherlands I always love to see our country being THE example of well designed roads and bike lanes :)
 
  • Like
Reactions: ReddyLeaf
Google Autonomous Cars have been able to recognize bicycles and even hand signals as well.


2861D8CB00000578-3070672-image-a-6_1430934746565.jpg



I don't expect it's a problem for Google (now called Waymo) but I haven't seen any studies from Tesla.
 
Mixing modes of transportation on the same pathways most always results in more accident, fatalities and injuries. The smaller and slower traveler most always loses out.

For safety, there should be a separate pathway for cars, trucks, airplanes, boats, skaters, runners, walkers, pet walkers, motorcycles, rockets, helicopters, and kids on wheels.

The greater the closing speeds, the greater the injuries.
 
  • Like
Reactions: ElectricTundra
I think the issue here is if people riding bikes will create congestion problems for people in AV's if the AV's are being extremely cautious. And worse if more people begin riding bikes because they know AV's won't hit them.
 
Mixing modes of transportation on the same pathways most always results in more accident, fatalities and injuries. The smaller and slower traveler most always loses out.

For safety, there should be a separate pathway for cars, trucks, airplanes, boats, skaters, runners, walkers, pet walkers, motorcycles, rockets, helicopters, and kids on wheels.

The greater the closing speeds, the greater the injuries.
Separate pathways would be nice, but it won't be practical to have them everywhere, particularly in semi-rural areas like where I live. As a bicyclist, I will not consent to being excluded from multi use roads that I currently have access to. Every sport carries some element of risk. It should be considered a basic human right to be able to travel around on one's own power, if not by bicycling at least by running.
 
Yeah. Big problems. The cars will run over and kill the riders.


But when has the r and d ever stopped so that the current functionality will never improve? It hasn't. This and other problems will be solved and I'm sure you are only asking this because you are assuming that the development of both electric and self driving cars will follow the same timeline of the iCE cars. Which it won't.
 
Separate pathways would be nice, but it won't be practical to have them everywhere, particularly in semi-rural areas like where I live. As a bicyclist, I will not consent to being excluded from multi use roads that I currently have access to. Every sport carries some element of risk. It should be considered a basic human right to be able to travel around on one's own power, if not by bicycling at least by running.

I think there's a big difference in most rural or even semi-rural and urban/suburban. In most rural areas it is quite easy to pass cyclists as there is less oncoming traffic and roads are often wider and straighter than many urban/suburban areas. But what about rural roads with lots of curves or hills that make it more difficult to find safe places to pass cyclists? At what point will AV drivers begin to complain that their cars won't pass cyclists and so they are constantly being slowed? It's a 50 MPH rural highway but they're constantly being slowed to 13 MPH for long stretches?
 
I think there's a big difference in most rural or even semi-rural and urban/suburban. In most rural areas it is quite easy to pass cyclists as there is less oncoming traffic and roads are often wider and straighter than many urban/suburban areas. But what about rural roads with lots of curves or hills that make it more difficult to find safe places to pass cyclists? At what point will AV drivers begin to complain that their cars won't pass cyclists and so they are constantly being slowed? It's a 50 MPH rural highway but they're constantly being slowed to 13 MPH for long stretches?
like noted in other threads where this has come up, this is just one of many, many variables that will have to be addressed before autonomous cars are ready for real time usage.
 
  • Like
Reactions: ElectricTundra
I think there's a big difference in most rural or even semi-rural and urban/suburban. In most rural areas it is quite easy to pass cyclists as there is less oncoming traffic and roads are often wider and straighter than many urban/suburban areas. But what about rural roads with lots of curves or hills that make it more difficult to find safe places to pass cyclists? At what point will AV drivers begin to complain that their cars won't pass cyclists and so they are constantly being slowed? It's a 50 MPH rural highway but they're constantly being slowed to 13 MPH for long stretches?
It should be possible to program/train AVs to pass cyclists both safely and effectively. This will probably take time to perfect, but the market should ultimately favor AVs that are capable of avoiding getting stuck behind slower-moving road users.

One aspect of this will be making better use of the available road surface, as I previously suggested. A significant fraction of human drivers aren't all that confident about maneuvering their vehicles around cyclists and won't pass until there's way more room than necessary or legally required. As a cyclist, I appreciate their consideration for my safety, but I'm not keen on having cars stuck behind me.

There may, however, be an issue if AVs insist on following the letter of the law with respect to encroaching on the oncoming lane, causing them to not pass even when safe opportunities exist. This brings up an interesting side point that goes beyond cyclists and pedestrians: the advent of strict, law-abiding AVs may increase society's motivation to seek revisions to traffic laws.

Also, in terms of safety, AVs should be able to do much better than humans at gauging the velocity of a given cyclist. You mention "13 mph", but the reality is that there's a great deal of variation depending on terrain and the cyclist's fitness level. Some drivers tend to underestimate the speed of conditioned cyclists and fail to accelerate enough to pass effectively. Or they pass unnecessarily and end up cutting off fast-moving cyclists.

Finally, with "overly polite" AVs on the road, we may find a need to more rigorously enforce some rules of the road for cyclists. For example, the law (at least in California) calls for cyclists to ride as far to the right as practicable, except when traveling at traffic speeds, passing, or turning left. A minority of riders aren't very considerate about this and tend to hold up traffic unnecessarily. Some cyclists, generally those who aren't as educated or serious about riding, tend to ride on the wrong side of the road which can be a real safety issue for everyone.
 
At what point will AV drivers begin to complain that their cars won't pass cyclists and so they are constantly being slowed?

Always amazes me when I am following a car, on a twisty rural road, and we come up behind a cyclist and the car in front is quite happy to overtake them on a blind bend. Whereas I'm holding up the traffic until I have clear sight of the road ahead. AP, with MegaMusk Sensors :), may well have better information about the road ahead than I have, and thus be able to decide to overtake safely and sooner.

V2V communication will, at some point, augment the car's understanding of whether the road ahead is clear.
 
Always amazes me when I am following a car, on a twisty rural road, and we come up behind a cyclist and the car in front is quite happy to overtake them on a blind bend. Whereas I'm holding up the traffic until I have clear sight of the road ahead. AP, with MegaMusk Sensors :), may well have better information about the road ahead than I have, and thus be able to decide to overtake safely and sooner.

V2V communication will, at some point, augment the car's understanding of whether the road ahead is clear.
Makes two of us. I've heard there's another somewhere in Belgium :)

Will V2V be of any use in this case until 100% of vehicles are AV? Even then would it need to account for a vehicle with a disabled xmitter? Or how would an AV account for a non-motor-vehicle in the oncoming lane or entering from a side road - like a bunch of cyclists?

Thinking about it from an AV developer standpoint, what all do I need to account for before I have this vehicle that I'm controlling go in to the oncoming lane? How much risk can I expose my passengers to? How much risk can I expose others to?
 
Will V2V be of any use in this case until 100% of vehicles are AV? Even then would it need to account for a vehicle with a disabled xmitter? Or how would an AV account for a non-motor-vehicle in the oncoming lane or entering from a side road - like a bunch of cyclists?

Thinking about it from an AV developer standpoint, what all do I need to account for before I have this vehicle that I'm controlling go in to the oncoming lane? How much risk can I expose my passengers to? How much risk can I expose others to?
For the reasons you gave and others, an AV should not ever rely exclusively on V2V communication for such critical decisions. Besides other road users, there could be hazards such rocks or debris on the road, or animals such as deer. No passing should take place without adequate visibility.

Practically speaking, though, even in no-passing zones (double yellow line in North America), there is often enough visibility to permit encroaching on the oncoming lane in order to pass cyclists. Because cyclists occupy less of the road than cars/trucks and they usually travel more slowly, passing cars' exposure to the oncoming lane is generally quite limited. While this is not (yet) reflected in our laws, it seems that most law enforcement officers tacitly allow this kind of passing. AVs may force us to revisit the relevant laws here.

Yes, there's always an element of risk because other road users may abruptly change speed or direction, rocks may fall from a hillside, a stationary animal or human may suddenly jump into the road, etc. And those risks will vary by location, of course.
 
Will V2V be of any use in this case until 100% of vehicles are AV?

My WAG is that if it does make a significant difference it will be mandated for retro-fit. Presumably only needs power and then GPS can say "I'm retrofit, may not be very inaccurate, but here I am, this is my heading and speed" and my end, with accurate onboard maps etc., would then have to interpolate that.

but Bikes (on the other side of the road), horse-riders and any other road-users (Deer?!!) would have to have V2V too ..

... used to terrify me when I used to ride a horse on the road, thinking about it now I would have been very happy to pop a V2V device in my pocket, no different to putting a hi-viz vest on - but there again most of the Horsey People around here think they own the blessed roads so fat-chance that they will bother ...

Speed differential, to a bicycle, is such that overtaking does not take much distance, so maybe less of a challenge to AP as it only needs to have clear visibility for a relatively short distance ahead
 
Speed differential, to a bicycle, is such that overtaking does not take much distance, so maybe less of a challenge to AP as it only needs to have clear visibility for a relatively short distance ahead
Yes, possibly. But... it depends. There are plenty of racer types who can average 30-35 in a pace line. If an AV is behind a bunch of 27 MPH racers on a 45 MPH road then how long will it take to pass and at what speed (been a few years since 8th grade math)?

My guess is that we're going to see a lot of AV's stuck behind cyclists waiting to safely pass them.
 
Practically speaking, though, even in no-passing zones (double yellow line in North America), there is often enough visibility to permit encroaching on the oncoming lane in order to pass cyclists.
I'd think that AV's will have to strictly obey all laws, simply from a liability standpoint. A lot of stuff that we do as drivers today an AV will not be able to do. So, the passing laws will need to change first and I'd not hold my breath. Then there's all kinds of liability even if the law changes. How large of a vehicle is in the oncoming lane? Or vehicles if multiple? Will they see you encroaching and move over? What if they don't see you? Or there is something preventing them from doing so? I just can't see an AV manufacturer taking the risk.