# How would the road network change, after all cars are self-driving?

In a future where all cars and public services are self-driving the way, I am assuming traffic congestion would be seriously limited, if not removed completely, because the road network would also have a companion communication network between vehicles, allowing the implementation of some optimal bandwidth occupation algorithm.

Besides the technology that will need to be installed along the road, how do you imagine the road network to be affected by that? (for instance: less multi-lane roads, or more single lane roads, with no traffic signals? or with different ones? what about parking; there are just examples, please don't refrain from adding your thoughts).

• Not enough for a complete answer, just an idea: I think, the cars needs to be controlled from an higher authority. If every car optimizes its own path, they interfere with others and the global solution diverges from the optimum (see for example en.wikipedia.org/wiki/Braess%27s_paradox ) – Peter Schneider Nov 14 '14 at 15:16
• @PeterSchneider That assumes you must reach the global optimal. If "close enough" is fine, then it may be better to use the resources available to analyze traffic and close roads where the paradox occurs. It would probably be much cheaper on the whole. – Geobits Nov 14 '14 at 18:18
• @PeterSchneider: Aaargh! "A higher authority." H is not a vowel! – Mason Wheeler Nov 15 '14 at 0:56
• @Twelfth: As I understand it, solar panel roadways are more of a gimmick than a serious technological possibility, for two fundamental reasons. First, the point of a solar panel is to generate power by sunlight striking it, but the point of a road is to have large vehicles (that cast large shadows and block sunlight) continually driving over it. Second, solar panels have to be very smooth for optimum light gathering, but roadways have to be very rough for optimum traction, especially in wet or otherwise slippery conditions. A smooth road is a safety hazard; a shadowed solar panel is useless. – Mason Wheeler Nov 15 '14 at 1:01
• @Guido: Yeah, it's a mistake I'd expect from native Italian or Spanish speakers. But I see it all too often from people who grew up speaking English, and it's really annoying because H is not silent in English. – Mason Wheeler Nov 15 '14 at 1:13

Currently with people driving there is a max capacity of around 1800 vehicles per hour per lane (if the 2 second rule is adhered to).

The following is the formula for stopping distance:

$$D_{\text{total}}=D_{p-r}+D_{\text{braking}}=vt_{p-r}+\frac{v^2}{2 \mu g}$$

$t_{p-r}$ is the reaction time (time from something happening in front of you to putting the foot on the brake), usually taken to be around 1.5 s; with automatic driving this can be nearly nullified. This would result in a much denser road use.

How intersections are dealt with depends on how cars can communicate. If each intersection has a computer that the car has to "log into" then it can reply by telling the car to slow down a bit so it can slot neatly in the gap in the other stream. This would make roundabouts much more efficient than they are now.

They would potentially reconfigure themselves depending on the traffic volume, a simple yielding configuration during low traffic, (turbo)roundabout as traffic increases all the way to full on stop light. This stop light can be more efficient than today's system because all cars slated to depart in the green slot can accelerate at the same time and at the same rate.

As for routing instead of always going for thé optimal path each car should instead use a weighted random selection of possible paths. The best paths has the biggest weight but less optimal paths are also possible. The weights can be compiled from data of travel times. Planned construction would then be artificially penalized, accidents and breakdowns as well. This will result in traffic distributing itself over the network.

Also here reconfiguration of the network can happen, turning a street into a one-way street to get the volume out of the city in the evening for example.

• Could you give a little more explanation as to why roundabouts would be more efficient than traffic lights? If cars could communicate, presumably they could also communicate with the lights. Also, are you implying that roundabouts would be more efficient than letting cars travel through intersections unimpeded by stop signs or lights? I would think that would be most efficient. – B T Nov 14 '14 at 22:12
• traffic lights require you to stop even if there is no other traffic in your way. In theory a roundabout would allow moving at lower speeds continually. – Oldcat Nov 14 '14 at 22:33
• @BT because the roundabout has less conflict areas, especially the turbo variants, and traffic lights require the conflicting traffic to halt. – ratchet freak Nov 14 '14 at 22:40
• Turbo roundabouts are interesting. And what about stopless intersection in the case of automated-only traffic? – B T Nov 14 '14 at 23:51
• If the crossing has an intelligenmt system anyway, that intelligent system could also switch the traffic lights to green for an approaching car if there are no other cars approaching. OTOH, if all driving is automatic, the traffic lights would not be needed at all; cars which communicate by radio don't need lights to tell them whether they may drive. – celtschk Nov 15 '14 at 12:15
• Roads would likely have some sort of embedded short-range lane and navigation markers, as cheap as possible. Perhaps something like reflectors with QR codes.
• Control would be mostly distributed in the cars themselves, with approval of the overall route going through a central coordinator, but a peer to peer network for coordinating with adjacent cars.
• More long-distance traffic would be routed over surface streets.
• The driving age / licensing requirements would be reduced or eliminated.
• People would have weak navigation skills / get lost easily.
• People would tolerate longer commutes because they could sleep, eat, shave, perhaps even shower along the way.
• The slightest mechanical defect would not be tolerated. A check engine light would result in immediate routing to the nearest mechanic.
• Deliveries would be fully automated and therefore faster, meaning you might shop online even for groceries.
• Cars would tend to be smaller, because it's easier for families to travel together in separate cars.
• You wouldn't need disabled parking spots, because everyone would get dropped off and picked up at the door.
• Poorer people might share a vehicle between extended families, even if they don't live close by.
• People would take more cross-country vacations without air travel.
• Politicians would not be able to resist appropriating the technology for their own ends to:
• track people's movements
• effect arrest warrants
• impose quotas
• ease congestion by forcing staggering of travel
• tax based on usage
• give preferential travel times to special interests
• I doubt self-driving cars will significantly increase the number of people making cross-country car trips. As uncomfortable as a 5 hour plane ride is, a 40 hour car ride would be worse, even if the car did the driving. – Johnny Nov 15 '14 at 0:50
• Also if you are to make a 5 hours family trip, you will want to have the family with you more than travelling split. Your point about the progressive weakening of navigational skills is extremely interesting though. – guido Nov 15 '14 at 1:06
• The car ride would be preferable with planned stops, and the nav skill can be seen today with GPS ;) – ratchet freak Nov 15 '14 at 1:15
• Btw. I am shopping for groceries only online ;) – Pavel Janicek Nov 15 '14 at 16:00
• @erdekhayser honestly I don't see that possible in the foreable future; car sharing like that you describe may (and will, probably) increase, perhaps a lot, but a system 100% based only on that would assume a perfect society. – guido Nov 15 '14 at 21:15

I don't want to be the nay-sayer in this, but probably the answer is "nothing much" - and here's why:

1. Traffic signals would probably exist, both on the road and in the cabin. Knowing the reason for a delay is important to both the riders and the people around them. While in-cabin may be sufficient for drivers and even passengers, pedestrians still need to be aware of traffic flow signals. They may be smaller or less conspicuous, but still necessary.

2. Speed limits would probably change, depending on the time of day and current traffic conditions. Maybe not as much as is imagined because if your vehicle doesn't know the mechanical capabilities of the vehicles around of it (for example, the state of disrepair) it will need to follow less closely than a human might who does not have that knowledge - or be aware of their lack of knowledge. Cars in disrepair may be able to report their impaired status which does not hinder traffic flow today. Today's driver is left to judge the safeness of their vehicle - and other vehicles - without much measurement. Additionally, what liabilities exist when your car communicates deteriorated equipment to other vehicles and authorities? Its unlikely this will be adopted by the majority of drivers any time soon.

3. More roads and lanes can be "re-purposed" depending on traffic loads and tolls. More roads could have part-time tolls. Really, this just sounds like a way to collect more taxes, but the roads will be more costly due to the embedded technology. Also, does it seem appropriate that some lanes that travel faster be available to the "wealthy" that can afford the tolls. Again, very slow to adopt, if ever...

4. A technology will need to be developed for local traffic optimization based on the current vehicle locations, speeds and capabilities. The location and maintenance of these systems will probably prevent any savings from reduced visual signal cost (i.e. traffic signals and signs), road maintenance or public property damage. But over time their value will probably be in reduced accidents and safety. Much like today's traffic signal was an improvement over traffic officers even when they weren't obviously economical and it wasn't clear if the general population would "obey a light."

5. Global optimization is incredibly unlikely for a very long time. It invades privacy - vehicles and government entities will know where you intend to go and then communicate it. Law enforcement, judges, insurance companies, traffic planners and other drivers rarely have access to that information (do they ever even ask for GPS route data?). Communicating intent is completely different from communicating status. But without explicit intent, much of the current infrastructure would have to remain in place (only minimal reduction to traffic lanes and road maintenance). Local optimization is possible for one-way roads and some intersections, but a "higher power" cannot re-route you if all it knows is that you are waiting to go forward or trying to turn but nothing else. That's still local optimization. This situation also leads to lengthy trials over privacy, rights and public protection.

6. Residential areas will likely try to enact laws to reduce or eliminate "non-local" traffic. While "short cuts" and "traffic route optimization" sound nice, people living on quiet roads where traffic may be re-routed frequently will certainly not appreciate the change. Again, many "gains" in automated systems make assumptions, and "reduced traffic congestion through the use of alternate routes" doesn't mean that alternates suddenly materialize or even can be planned any better than they are today. But existing residential roads that are not congested could suddenly become the frequent target for automated traffic route alternatives. The residents on those roads will likely fight back, as they do today with planned changes in traffic patterns.

7. While it is nice to imagine time savings for people that do not have to drive, and can then "use that time" in other ways, it's probably not as beneficial as imagined. First, the cost of the technology will be a burden that a consumer must bare. Second, like early 20th century household appliances that "eased household chores" which led to increased expectations of cleanliness, so too will automated driving probably lead to longer commutes and no addition to overall leisure time. Although it may be possible to consider in-cabin time "leisure" time, doing so means larger cars, less energy efficiency, more equipment, higher operating costs, etc.

8. Last, just as the internet did not end "brick and mortar" shopping in places like grocery stores, etc. automated driving will not reduce or eliminate "errands" - in fact, because it will be easier to perform (less work on a human) it is likely to happen more frequently. People can more easily get to and from stores. And it would be less of a burden on a homeward bound household member to "stop by the store" than ever. So, it's likely that with a person's mind freed from the burden of navigation and driving, they will likely shop more in physical locations, not less, possibly requiring more roads, increased traffic congestion, higher energy usage, etc. Ease of navigation and transportation does not automatically improve the planning capabilities of a population - and could make impulse route changes more frequent since the "driver" is no longer focused on the activity of driving and can suddenly see a store or advertisement that changes the route of the vehicle.

I realize that most answers here talk about efficiencies and improvements. I believe automated driving will result in safety improvements more than anything else. Its unlikely to improve much else for a very long time unless those safety improvements of more value than freedom and privacy. Conversely, improving safety may be more important than the "freedom" to harm or "privacy" of criminals. It would be a vastly different future either way.

• Hello Jim, and welcome to WB! This is my second review of a first post, and it seems that I only get the really high-quality ones ;-) – Shokhet Nov 16 '14 at 2:27
• Why would you think that self driving cars would lead to high speed lanes for the wealthy, when we don't have high speed lanes for the wealthy right now? I don't see how their implementation depends on self driving cars in any fashion. You just need to prevent people from using the lane without paying, which can be achieved just as well with a barrier and a toll station (which already can be operated automatically). – Cubic Nov 16 '14 at 5:45
• @Cubic, where I live, we have exactly that: a high speed lane for the wealthy (well, not that wealthy, but there's a toll not everyone is willing to pay). The first level is free but slower, and the second level has a toll, but is considerably faster. en.wikipedia.org/wiki/Anillo_Perif%C3%A9rico – Arturo Torres Sánchez Nov 16 '14 at 16:38
• I think it's slightly too early to say that the Internet didn't end brick and mortar stores. We're only just over two decades in, and the number of stores is shrinking. – RemcoGerlich Mar 23 '15 at 10:09

Very little to none in first years is my best guess idea, because thats how the self driven cars are driven now in nowadays system and it is very pricy to adapt the road system because of new technology. Good example on this is the reason why there are no super fast trains in USA. Because updating nowadays system is pricy.

I am inclining to "shared ownership" in next years. Say 20 years after having every single car self driven, the system of owning your own car will become more and more obsolete, because car sharing will be really easy. And I can imagine there will be companies offering pool of cars for a fee.

Just imagine "taxi" service which will take you anywhere you want, at any hour, and will cost reasonable low amount of money

And then - movement to automated public system if every car is automated, imagine Uber - like system where you "rent" your car while driving to work, and with bonus, still able to drive you to work on time

I am torn apart about between car communication If automated car should be able to communicate with other is big unknown to me. Yes, I am using Waze to calculate me best home route and the navigation is working great thanks to community, but also, machine to machine communication is hackable and may be made incompatible

• Companies offering pool of cars, interesting idea. Specially if the cost is easy to afford. – Hatoru Hansou Nov 14 '14 at 16:49
• I don't think it would be made obsolete - people like owning things. People like ownership as a status symbol, but they also like owning things for the convenience - for instance, if you have your own car, you can store things in the trunk, you can leave things you might want in the glove compartment or in the dash, and you'll never have to wait for one to show up when you call it. I imagine shared ownership systems will become increasingly popular, but I also imagine there'll always be a place for individual ownership as well, same way that libraries are awesome, but people still own books. – neminem Nov 14 '14 at 17:12
• @HatoruHansou - this already exists in many cities -- car share services are becoming more and more common for city dwellers that don't want to own a car. Users buy a membership for the service and have access to a shared fleet of cars. In many cities, the cost of membership and reasonable use of a car is cheaper than the cost of parking alone. – Johnny Nov 15 '14 at 0:46
• @neminem actually I think the concept of trunks will be obsolete. Instead you will have multiple small self-driving cars - not all of them will be able to transport passengers. You can go grocery shopping on the way to work. You fill one of your storage cars with the groceries. It drives itself home. While you go onto work. Ideally it should be able to let itself into your house. – emory Nov 15 '14 at 1:08
• @neminem Ownership also means responsibility. I think the idea of not having to care about repairs, fuel or even acquiring a car in the first place is going to be quite attractive for many people. – Cubic Nov 16 '14 at 5:40

It probably won't be long into it before personal car ownership becomes rare and either taxi-type services become more common or you timeshare vehicles with friends or even neighborhoods.

Once you have a self-driving vehicle, there is going to be less reason for it to be just sitting somewhere, when it can be working. There is a husband and wife who have developed a solar powered road system and it does several things, it does convert sunlight to electricity, it is a very decent driving surface, and many other things. Including having LED lights that can be used to make road markings, (ie switch lane direction during rush hours) give warnings and possibly talk to a smart car.

Combining that technology with self-driving cars will be truly revolutionary. At that point electric cars could actually recharge as they are driving down the road, or at least have recharging stations scattered along the way. It would help reduce carbon foot-print and the estimate was that if all driving surfaces in a town to small city it would provide enough power to mostly take the city off the gird, total green.

You also wouldn't need big vehicles for most people, since if you buy a TV it can be delivered by the store in a self driving van, would different security so the right person gets the delivery however.

• For personal deliveries, an identification card (could be the credit card you pay with) would probably need to be swiped to open the car door and get the package. – celtschk Nov 15 '14 at 12:24
• @celtschk That is what I was thinking – bowlturner Nov 15 '14 at 15:05
• @bowlturner I suggest you look into the criticisms of solar roadway. It's basically a bunch of really inefficient systems thrown together with a bunch of hyperbole. – NPSF3000 Dec 29 '14 at 11:32
• @NPSF3000 even if their design is poor, I think the idea has a lot of merit, someone will come up with something similar that actually works. – bowlturner Dec 29 '14 at 14:39
• @Bowlturner, that's the really great thing - we already have. All the good things about solar roadways already exists, just in far, far, far more efficient ways. – NPSF3000 Dec 30 '14 at 0:18

Congestion won't be eliminated, it might even increase. Having self driving cars can in theory make far better use of existing roads (see ratchet freak's answer for example).

However currently population is going up, and desire to spend on infrastructure is not going up. Assuming those trends continue you will start seeing reduced congestion as self driving cars are adopted. That'll make people happy, but it won't last. Population increase will start eating into the gains, and unless we start spending more on roads congestion will start to raise again.

I would even guess that as long as the amount of congestion climbs slowly it will be allowed to climb above current levels because it is much more productive time:

• eat a meal
• watch TV (or for long commutes a movie)
• if you own an RV (small RVs might get more popular):
• sleep
• take a shower
• get dressed
• knowledge workers could do some of their job during the commute (what do you think Apple/Google/other employees do on the buses now?)

I'm also not 100% sure how much congestion will reduce short term as a lot of things might end up causing more miles to be driven:

• people may be willing to tolerate even higher commute times in exchange for lower housing costs
• if cars self drive there is less reason for young people not to have access to them (or have more after schools activities then their families can currently drive them to)
• elderly people who are no longer safe to drive themselves can regain that sort of mobility, and I expect many of them will
• disabled people (blind, or even just folks who have poor night vision, folks with reduced motor control, and many more)
• dog delivery for lunch time walkies, why have a strange walk your dog when you could get a little light exercise? You just need a dog door with a timer and a garage (no, I doubt this will be a large number of miles, but I think this is the kind of thing in the "long tail" of "nobody does it now, but a few people will do it in the future")
• parking lots can move far from buildings (live in SF, it might cost less to have your car parked in another city overnight and come pick you up, that'll add a lot of road miles!)
• car sharing inside a family can have the car taking someone to work, driving back home, and taking other family members to work (or social activities)
• if most people drift away from individual ownership of cars and start renting time from a fleet of cars they will spend time driving from the last drop off spot to the next pick up spot
• the cost of "same day delivery" will drop a lot, and it might not all displace purchases you would have driven to a store for anyway (then again, if it is "same day", not "ASAP" shopping could be combined from multiple stores to multiple people, so maybe it could be less traffic for that)

Some other things I think might be interesting, if congestion starts to get bad again with all self driving cars you can have lot more gradations of lane limits. Rather then just "carpool" lanes that also allow low/no emissions vehicles during specific hours you could end up with surge rates where as roads get more congested you only get to use some roads/lanes/routes if you are willing to pay more. (and in a utopian psudo-libratarian future that money would pay to add lanes or roads to reduce congestion, but I think in the real world that'll be an extra tax only part of which will go to roads, and counter productively may encourage local movements NOT to do anything to fix congestions because that is how they are funding schools, police, and the governor's reelection campaign!)

Emergency vehicles are currently very hampered by congestion. The sirens can only be heard so far away, and while most people try to do the right thing, some might think the right thing is to get out of the traveling lanes so the cops/fire/ambo can use them, other might think they need to stay out of the "break down lanes" so the cops/fire/ambo can use them. Some might decide to do either of those things "in a second or two, after I take advantage of the gap!". With self-driving vehicles you will get a consistent behavior, and traffic can be alerted along an entire route as soon as needed. This could save a lot of lives.

Most traffic accidents are caused by poor driving, or drivers medical issues. They will go away. Vehicle failures will still cause accidents. They might cause fewer (self driving cars might refuse to drive if they haven't been maintained properly, and even if they don't, it is a whole lot easier to keep to a maintenance schedule if the car takes you to work, and takes itself in for service, and picks you back up after work as per normal, not waiting until you can find time for it). They might cause more (sensors add more points of failure). They might cause less (if programmed to deal with a mechanical failure, and properly tested a self driving car won't forget how to deal with correctable failures like flat tires, and even ones that have limited correction ability like break system failures self driving cards could far faster and more effectively warn other cars to get out of the way)

Accidents cause more accidents (people gawking at an accident aren't paying as much attention to driving as they should). That won't happen with self driving.

• Are you suggesting people would have their dogs driven to their workplace so they can walk them during their lunch break, and then have them driven back home? – Azor Ahai Apr 10 '17 at 23:48
• Yes, but as I stated I don't think a lot of people will do that, but ZERO do it now. It is there more as an idea of "stuff that is impossible now will be possible in the future, and individually those things are unlikely to amount to much, but take together they may well..." – Stripes Apr 10 '17 at 23:53
• Haha oh I was just clarifying what your wrote – Azor Ahai Apr 10 '17 at 23:58
• people will have no idea where anything is anymore, and would not recognise their city on a map.
• people will try to trick you into clicking on something to change your destination and drop you off at some nasty place.
• today people plan less and look up the route after leaving. Continue that trend and people won’t have a clear idea where they want to go until underway, searching for the right kind of establishment on the map-guide. With ads.
• dogs will take the car on their own once they learn that they can. They’ll go joy-riding instrad of walking, or visit you at work minutes after you punch in.
• cars will be less self-owned dedicated and more like trains and elevators: call for a ride, tell it where to go. You can have small cabins for personal transit, limo-bus’s for groups, and cargo handlers of any size. You can use a basic 2-seater, or pay extra for luxury or mini-RV’s for longer trips.
• or, more modular: the roomy seats-plus cab will rendezvous and dock with locally sourced expansion modules for meals, sleeping, entertainment, etc. on demand. The car will grow and shrink back down, reducing what needs to transport over long distances and offering versatility.
• party busses and reconfiguring social trains will offer another possibility for docking for a journey, and will become a destination in itself for teenagers (and dogs).
• people will stop commuting as much, or going shopping in person, and reduce car usage. They will use ideas like above to make it fun and interesting to get out there, just like today movie theaters have to work hard to attract business as home theaters are nicer than “plain” movie theaters.
• surface vehicles using roads will combine with rails, containers, flying drones, and other forms of transportation to make a seemless whole. You can address a package and toss it outside and it gets to its destination, somehow. Same with people: if you enter a destination a thousand miles away, the system will use local neighborhood roads, freeways, land-train superfreeways, pack the cars into containers and load on locomotives, as fits.
• I already address a package and toss it outside and it gets to its destination somehow. No need for self-driving anything for that. – Michael Hampton Oct 11 '16 at 22:02

The road network itself would probably become smart. I can picture realtime traffic flow manangement, where the cars know of the fastest route at any time. Stop signs and traffic lights would be obsolete if all vehicles were self driving. Speed limits would increase to about 160km/h on controlled access highways, and decrease to about 20km/h in cities. Highways would be mostly road trains of freight and buses. I'd expect far fewer medians, and roads with three lanes being the norm instead of two. Inbound/rush hour direction having an express and slow/stopping traffic lanes, and the other direction having just one. Benefits of more lanes would be the same.

The actual path of the roads would stay roughly the same, new roads would be built in much the same way: the path of least resistance.

This assumes vehicle manufacturers are willing or forced to work with one another. If they're not, nothing will change.

If you had perfect routing and better traffic management, what you could probably end up with is a system where the number of intermediate sized roads are less than today.

Assuming every house needs road access, you still need the itty bitty neighborhood streets. And for high volume, long distance routes you need superhighways. But with better management all roads could hold more cars without congestion, and merging could in theory be managed without adding congestion.

So the need for "middle sized roads" of a variety of sizes is where the squeeze could come. I'm not sure if quick merge into a big fast road or smallish roads with high speeds would win out.

In most systems: consumption grows to fill bandwidth. In other words, if people think that the roads aren't clogged then they're more likely to go somewhere. They're more likely to purchase a car for their teenage child. Old people will feel empowered to drive places. And then suddenly--even with automation--the roads are still effectively clogged.

Even if the roads themselves are automated this doesn't mean that when you're arrived at your destination there will be adequate parking. Real estate prices increase over time so the value of a parking lot will become too great not to develop it in some other way.