Autonomous vehicles impact on real estate
-
- IDS Center
- Posts: 4617
- Joined: December 4th, 2012, 11:41 am
Re: Autonomous vehicles impact on real estate
I do not buy arguments that autonomous vehicles will magically reduce congestion and parking needs. At all. For one, to have that happen we would need perfect driving and parking, which would require pretty much ubiquitous use of autonomous vehicles and I don't see that happening in my lifetime. It may someday, but that day is so far in the future its impossible to predict what transportation will even look like. Holo-commuting could take over! And even if we do have the vast majority of vehicles (> 95%) autonomous, there will likely be regulations in place on spacing, etc. for safety reasons that will reduce the perfectness we'll be able to achieve.
I agree that on low-use/last-mile transit situations, autonomous vehicles make a lot of sense. It's also where PRT might (might!) make some sense. Autonomous vehicles are much better than PRT due to the much lower infrastructure costs. I don't know yet if such a service should be provided by a transit agency, a private company or just some distributed network of vehicle owners. Probably not the latter if it's meant to be a true reliable public service.
Simply loading disabled people isn't the biggest problem. It's getting them secured so they aren't killed in the case of an accident. Even on LRT I always feel a bit nervous about unsecured wheelchairs.
Parking is not going away, even in the core. Not until we achieve the nirvana of ubiquitous autonomous vehicles and as I said, that's at least half a century away and likely much more. Again, it's primarily a social problem, not a technical one. People are attached to their cars and they're attached to the idea that parking is an absolute necessity. Businesses are not going to go and rip up parking lots and ramps. We're building more every day.
I will say that there's a chance that outside factors might force a switchover much sooner. Cars could get prohibitively expensive for individuals or we could see the collapse of suburbia or climate change could get so bad that people finally accept that we need major changes in how we do things. But if we get into those situations, the pain is going to be enormous and autonomous vehicles and parking will be the last things on our minds.
I agree that on low-use/last-mile transit situations, autonomous vehicles make a lot of sense. It's also where PRT might (might!) make some sense. Autonomous vehicles are much better than PRT due to the much lower infrastructure costs. I don't know yet if such a service should be provided by a transit agency, a private company or just some distributed network of vehicle owners. Probably not the latter if it's meant to be a true reliable public service.
Simply loading disabled people isn't the biggest problem. It's getting them secured so they aren't killed in the case of an accident. Even on LRT I always feel a bit nervous about unsecured wheelchairs.
Parking is not going away, even in the core. Not until we achieve the nirvana of ubiquitous autonomous vehicles and as I said, that's at least half a century away and likely much more. Again, it's primarily a social problem, not a technical one. People are attached to their cars and they're attached to the idea that parking is an absolute necessity. Businesses are not going to go and rip up parking lots and ramps. We're building more every day.
I will say that there's a chance that outside factors might force a switchover much sooner. Cars could get prohibitively expensive for individuals or we could see the collapse of suburbia or climate change could get so bad that people finally accept that we need major changes in how we do things. But if we get into those situations, the pain is going to be enormous and autonomous vehicles and parking will be the last things on our minds.
-
- Nicollet Mall
- Posts: 137
- Joined: April 24th, 2017, 10:47 am
Re: Autonomous vehicles impact on real estate
Hmmm, individual owners renting out their autonomous cars via app likely will be a thing, hadn't thought about it much, but pretty sure that a fleet owner could provide lots advantages over individual owner making their car available when they don't need it. Google has dropped idea of making their own autonomous pods but they are buying fleet vehicles from others.
What you want from a fleet vehicle is different from individually-owned vehicle, not the least is the emphasis on reducing maintenance and operating costs and improving durability for tons of miles in short time. Also, I think shared vehicles may be built for niche purposes than individual cars that cater to many uses the owners may have for it. I think eating in cars will be a big thing, and restaurant on wheels will be a big option in fleets. Like, order breakfast and car to be at your apartment at 7:30 am with your favorite meal, eat on way to work. Or order a food car that is stacked with fast food/vending options in refrigerators/warming thing, such sandwiches etc.
Those that the run efficient fleets might not be same ones that own the transportation OS - but people that provide the fleets and charge the OS people for the vehicles, might be a sort of like Foxconn is to Apple with mobile phones. The ones that rule in terms of managing cars, making good user interface software (see Uber, Lyft) may collaborate on design of hardware or not. Apple designs hardware for phones and then hires Foxconn to make, so maybe car OS will do same. Google only does software for phones, and many companies make the hardware however they want. So maybe ride-share cars go the way of Apple did with phones, the Ubers and Lyfts design the cars and order them from manufacturer. Or Tesla maybe does it all, even more integrated than Apple has been with phones. Or maybe ride-share OS do like Google is doing with Waymo, just buy cars from manufacture.
What you want from a fleet vehicle is different from individually-owned vehicle, not the least is the emphasis on reducing maintenance and operating costs and improving durability for tons of miles in short time. Also, I think shared vehicles may be built for niche purposes than individual cars that cater to many uses the owners may have for it. I think eating in cars will be a big thing, and restaurant on wheels will be a big option in fleets. Like, order breakfast and car to be at your apartment at 7:30 am with your favorite meal, eat on way to work. Or order a food car that is stacked with fast food/vending options in refrigerators/warming thing, such sandwiches etc.
Those that the run efficient fleets might not be same ones that own the transportation OS - but people that provide the fleets and charge the OS people for the vehicles, might be a sort of like Foxconn is to Apple with mobile phones. The ones that rule in terms of managing cars, making good user interface software (see Uber, Lyft) may collaborate on design of hardware or not. Apple designs hardware for phones and then hires Foxconn to make, so maybe car OS will do same. Google only does software for phones, and many companies make the hardware however they want. So maybe ride-share cars go the way of Apple did with phones, the Ubers and Lyfts design the cars and order them from manufacturer. Or Tesla maybe does it all, even more integrated than Apple has been with phones. Or maybe ride-share OS do like Google is doing with Waymo, just buy cars from manufacture.
Re: Autonomous vehicles impact on real estate
One very big difference between Apple and Google phones (Pixel is contract manufactured for Google, btw) is that the ride sharing part already exists, and one of its large advantages was not owning and operating fleets (along with ignoring local regulations).
I'm not sure why privately owned vehicles don't want reduced maintenance and operating costs or improved durability, and I don't think specialty vehicles - like one stocked with food - are likely to be a meaningful part of the business for a very long time if ever. Why not just have the car stop at your favorite food vendor? And if you own the fleet, you probably don't want people creating messes you need staff to clean and stock.
Anyway, I can see Uber (which already has I think) and Lyft, collaborating in the design of autonomous vehicles for their use, and even investing in starter fleets, but I don't see why they'd want to own every vehicle on their system if they can avoid it. For a nationwide network, we're talking about trillions of dollars of capital, especially in the early stages, that they may well be able to avoid, as they largely do right now.
Leaving aside long distance trucking, the beginning of autonomous vehicles will be in personally owned cars that can drive themselves (tech that sort of already exists). It stands to reason that's also where autonomous ride sharing starts, for the very simply reason is that will be the first available tech. Transitioning to a corporate fleets will need very compelling reasons for it to happen and I can't think of any that outweigh the capital costs.
ETA: Except, perhaps, if they tech winds up needing centralized control for it to actually work. In which case, I don't know if we'll ever get to where people are willing to fully adopt it.
I'm not sure why privately owned vehicles don't want reduced maintenance and operating costs or improved durability, and I don't think specialty vehicles - like one stocked with food - are likely to be a meaningful part of the business for a very long time if ever. Why not just have the car stop at your favorite food vendor? And if you own the fleet, you probably don't want people creating messes you need staff to clean and stock.
Anyway, I can see Uber (which already has I think) and Lyft, collaborating in the design of autonomous vehicles for their use, and even investing in starter fleets, but I don't see why they'd want to own every vehicle on their system if they can avoid it. For a nationwide network, we're talking about trillions of dollars of capital, especially in the early stages, that they may well be able to avoid, as they largely do right now.
Leaving aside long distance trucking, the beginning of autonomous vehicles will be in personally owned cars that can drive themselves (tech that sort of already exists). It stands to reason that's also where autonomous ride sharing starts, for the very simply reason is that will be the first available tech. Transitioning to a corporate fleets will need very compelling reasons for it to happen and I can't think of any that outweigh the capital costs.
ETA: Except, perhaps, if they tech winds up needing centralized control for it to actually work. In which case, I don't know if we'll ever get to where people are willing to fully adopt it.
-
- IDS Center
- Posts: 4617
- Joined: December 4th, 2012, 11:41 am
Re: Autonomous vehicles impact on real estate
https://techcrunch.com/2015/03/03/in-th ... interface/
"Uber, the world’s largest taxi company, owns no vehicles. Facebook, the world’s most popular media owner, creates no content. Alibaba, the most valuable retailer, has no inventory. And Airbnb, the world’s largest accommodation provider, owns no real estate."
"Uber, the world’s largest taxi company, owns no vehicles. Facebook, the world’s most popular media owner, creates no content. Alibaba, the most valuable retailer, has no inventory. And Airbnb, the world’s largest accommodation provider, owns no real estate."
Re: Autonomous vehicles impact on real estate
Yeah, better wayfinding tools like the screens metro transit is testing out on the A Line (maybe even some nice kiosks where you could get directions!) would help. Adjusting stops to avoid snowbanks/puddles could be done via the AI and sensors in the literal sense, but most importantly avoided in the first place by having better stops (fully-level-boarding aBRT stops have better drainage). Snow removal on sidewalks during the "last mile" to/from the bus stop would be (and is) a far bigger accessibility problem. We really need municipal snow removal *to the sidewalk*. We already have systems for safety and dealing with unruly passengers on our LRT system.Right, losing the driver could bring down costs marginal costs a lot, as long as we don't worry about things like helping disabled passengers board, strapping down wheelchairs, adjusting stops so people aren't trying to climb over snowbanks or exiting into puddles, giving directions, providing supervision for unruly passengers, enforcing decorum, etc.Labor is the highest cost for providing transit service, I think there's a lot of opportunity for driver-less bus services.
Some of those things could obviously be replaced with better stops/stations, maps and route finding, better designed buses, but not all.
Now the part I'm most interested in is how to help passengers with disabilities. I recently ran across this grad paper on the topic of precision docking at level-boarding bus platforms. The Red Line uses a guide strip, which isn't the most functional solution (since Red Line buses still need wheelchair ramps). http://cardinalscholar.bsu.edu/bitstrea ... sequence=1
If we wanted to, the technology now exists for reliable ADA-compliant level boarding. While it would be best to roll out on BRT/aBRT routes first, and then upgrade everything else, these options are pretty affordable (especially if we had a pile of driver-less bus money). Magnetic Guidance would probably be best for us since it works great in all weather, works well with curves, and was the most precise option. The only complaint was that during general travel (not when docking at a station) it can mildly jerky. If we were to use it beyond docking, it could be reserved for bad weather, while Optical Guidance (more traditional self-driving) would be the default.
Using magnetic guidance would cost $100k per vehicle. That's around $90M total for every bus we have. At $10k-$20k per lane-mile, the cost is negligible for using it near stations for level-boarding purposes. If we used it everywhere for bad-weather autonomous driving, it would cost $10M-20M per 1k lane miles (how many miles of road do we actually have bus service on?) For reference, all of Minnesota has 70k miles of municipal and county roads (according to MNDOT).
The biggest question is passive-restraint technology for wheelchairs on buses. There would have to be a public awareness campaign to make sure people know how to use it, and we would have signage on the bus that helps. There are rear-facing passive restraints available on the market that might suffice, and having no driver would free up space for ADA seating. In my experience parking your wheelchair perpendicular to the direction of travel and next to a divider works the best on the light rail (since tipping over is a concern for accelerating and braking, usually people just focus on braking).
There was something that might have been posted on this forum somewhere that I read a while back. It was about increasing the number of wheelchair spaces on buses overseas (Europe? Japan?) and I think they had booths you could back your chair into with a divider on either side to keep you in place (it should also have some stanchions/grab bars). Assuming it works, it could be a simple, space-efficient, and cost-effective solution.
Re: Autonomous vehicles impact on real estate
Right, I thought I alluded to that. But anyway, we have many, many stops. There's no way we're building a station for each of them. And if we do eventually find the capital to invest in building many of them, we'll be adding back some labor costs to maintain them (which could also go to your next point, snow removal).but most importantly avoided in the first place by having better stops (fully-level-boarding aBRT stops have better drainage)
We have cameras and officers that ride the trains and patrol stations. Again, there are way more bus routes and stops, with longer duration routes. Maybe it scales for the actual safety issues, but they won't be there to tell the kid to turn down his music or the unruly person to pipe down unless we're again adding back a lot more of our labor cost savings.We already have systems for safety and dealing with unruly passengers on our LRT system.
Maybe we can get by without it, though.
-
- Wells Fargo Center
- Posts: 1138
- Joined: June 1st, 2012, 8:03 am
Re: Autonomous vehicles impact on real estate
https://www.theverge.com/2018/3/19/1713 ... pe-arizona
Uber is pulling all of its autonomous vehicles off road. Add five years to any scenario you may have in your head involving the adoption of driverless cars.
Uber is pulling all of its autonomous vehicles off road. Add five years to any scenario you may have in your head involving the adoption of driverless cars.
-
- Wells Fargo Center
- Posts: 1138
- Joined: June 1st, 2012, 8:03 am
Re: Autonomous vehicles impact on real estate
Self-driving cars have driven about 4 million miles on U.S. roads and one person is dead.
That's a WAY worse record -- at this point anyway -- than for human driven, which is ~1.25 deaths per 100 million miles.
That's a WAY worse record -- at this point anyway -- than for human driven, which is ~1.25 deaths per 100 million miles.
Re: Autonomous vehicles impact on real estate
I was guessing someone tried to cross between parked cars, but it does not appear there was any street parking in that area. Very, very large swath of pavement at the stated intersection of North Mill and East Curry in Tempe. Perhaps this really says more about road design than autonomous vehicles?
http://goo.gl/maps/SwEzqLdU1y52
http://goo.gl/maps/daJt74MG3yq
http://goo.gl/maps/SwEzqLdU1y52
http://goo.gl/maps/daJt74MG3yq
Re: Autonomous vehicles impact on real estate
Thankfully, there will be extensive video documentation of what happened.
Re: Autonomous vehicles impact on real estate
Given everything that's been happening over the last couple of years, I probably wouldn't hold up Uber as the leader in autonomous vehicles.
Re: Autonomous vehicles impact on real estate
Arizona is ground zero for hostile urban design, so it won't be a surprise if that turns out to be the biggest factor.I was guessing someone tried to cross between parked cars, but it does not appear there was any street parking in that area. Very, very large swath of pavement at the stated intersection of North Mill and East Curry in Tempe. Perhaps this really says more about road design than autonomous vehicles?
http://goo.gl/maps/SwEzqLdU1y52
http://goo.gl/maps/daJt74MG3yq
Still, it's very relevant to AVs because proponents claim they will be better able to handle crowded and complex situations, and because their impending arrival is frequently cited as a reason to further expand auto-centric urban designs like the one that killed this woman.
-
- Nicollet Mall
- Posts: 137
- Joined: April 24th, 2017, 10:47 am
Re: RE: Re: Autonomous vehicles impact on real estate
The car was driving 40 mph when it hit her, persumably the legal speed limit and the car speed this road was designed for.Arizona is ground zero for hostile urban design, so it won't be a surprise if that turns out to be the biggest factor.I was guessing someone tried to cross between parked cars, but it does not appear there was any street parking in that area. Very, very large swath of pavement at the stated intersection of North Mill and East Curry in Tempe. Perhaps this really says more about road design than autonomous vehicles?
http://goo.gl/maps/SwEzqLdU1y52
http://goo.gl/maps/daJt74MG3yq
Still, it's very relevant to AVs because proponents claim they will be better able to handle crowded and complex situations, and because their impending arrival is frequently cited as a reason to further expand auto-centric urban designs like the one that killed this woman.
40 mph is so much more lethal than 20 mph.
To me Uber is just being stupid doing 40 mph on surface street even if it is legal. Going 20 mph might even too difficult because how fast tither cars are going but you don't have to go 40. I'm assuming most riders would still value a cheap driverless ride if it took a bit longer...because it is cheap and you can do tiher things because you are not driving.
In my mind, surface streets of any kind should all be 25 mph max. How much time does 40 mph save when you have traffic and stop lights.
Keep the fast car speeds to expressways, freeways.
The woman was walking her bike into the road at 10 pm at night and may have been homeless and was jaywalking. So like and combination of faiures, car going too fast on surface street, dark, person walking a bike with maybe lots of stuff on it (maybe messed up AI software somehow), etc.
Sent from my Pixel 2 XL using Tapatalk
-
- Nicollet Mall
- Posts: 137
- Joined: April 24th, 2017, 10:47 am
Re: Autonomous vehicles impact on real estate
She was on a median and walked out into lane.
They are saying it was her fault, but aren't you supposed to stop for someone standing in a median waiting to cross?
It was 10 pm at night, probably not that much traffic so seems like it would be easy to avoid cars if need be. No way a pedestrian in Phoenix should ever to think to step in front of a car at night, just reality but don't know if car has no blame.
Sent from my Pixel 2 XL using Tapatalk
They are saying it was her fault, but aren't you supposed to stop for someone standing in a median waiting to cross?
It was 10 pm at night, probably not that much traffic so seems like it would be easy to avoid cars if need be. No way a pedestrian in Phoenix should ever to think to step in front of a car at night, just reality but don't know if car has no blame.
Sent from my Pixel 2 XL using Tapatalk
-
- Nicollet Mall
- Posts: 137
- Joined: April 24th, 2017, 10:47 am
Re: Autonomous vehicles impact on real estate
If the car cannot handle unexpected pedestrians, it should not be on the road.It was 10 pm at night, probably not that much traffic so seems like it would be easy to avoid cars if need be. No way a pedestrian in Phoenix should ever to think to step in front of a car at night, just reality but don't know if car has no blame.
Yes, human drivers are bad at it too, but we don't send them out there knowing they can't handle relatively common occurrences.
Re: Autonomous vehicles impact on real estate
I think we need to know how visible the pedestrian was and how far out the car was when she stepped out. I generally agree with you amiller92 that these things should be able to handle unexpected pedestrians, but there is a limit to how much can be done when high-speed streets are involved. Which is a good reason to slow down the street, but maybe not ban autonomous cars.
I am a little disturbed that the car was going 38 in a 35mph zone, though. These things shouldn't be designed to mimic human speeding.
I am a little disturbed that the car was going 38 in a 35mph zone, though. These things shouldn't be designed to mimic human speeding.
-
- Capella Tower
- Posts: 2869
- Joined: June 1st, 2012, 9:19 am
Re: Autonomous vehicles impact on real estate
Well, I think the most interesting aspect of all of this is that Political leaders are considering letting these future AV companies (Google, Uber, etc.) include some kind of disclosure that prohibits any potential accident victims from filing a lawsuit against the companies at fault. They'd be allowed to follow an arbitration process, but that process would be led by the company in question in most cases.
I'm sure somebody on here knows more about this subject than I do, but I believe this is the gist of the situation and I don't know how people would sign their rights away just to jump into an autonomous vehicle (however, as I understand it people already do this when they step into Lyft or Uber).
How much of a potential issue are lawsuits/litigation to the success and implementation of AVs?
I'm sure somebody on here knows more about this subject than I do, but I believe this is the gist of the situation and I don't know how people would sign their rights away just to jump into an autonomous vehicle (however, as I understand it people already do this when they step into Lyft or Uber).
How much of a potential issue are lawsuits/litigation to the success and implementation of AVs?
Re: Autonomous vehicles impact on real estate
We're going to have to figure out what the rules are, just as we did with cars. There's some quantity of liability/harm that we need to distribute among (1) the manufacturer/programmer, (2) the vehicle owner, (3) the victim and (4) the state. Once we decide who's on the hook in the which proportions, they can all go get insurance to cover it and away we go.
ETA: With cars (as with trains and other things before cars), we mostly decided to use negligence rules - whether a person acted reasonably - that won't really work when a vehicle acts exactly as it's programed to act. We can still litigate over whether a particular programming choice was reasonable, but there will be crashes that happen even where everything works as it's supposed to and the supposed to isn't unreasonable. We have to decide who covers that.
ETA: With cars (as with trains and other things before cars), we mostly decided to use negligence rules - whether a person acted reasonably - that won't really work when a vehicle acts exactly as it's programed to act. We can still litigate over whether a particular programming choice was reasonable, but there will be crashes that happen even where everything works as it's supposed to and the supposed to isn't unreasonable. We have to decide who covers that.
-
- Nicollet Mall
- Posts: 137
- Joined: April 24th, 2017, 10:47 am
Re: Autonomous vehicles impact on real estate
Negligence will likely be the standard. In this case, fact they had a safety driver still in car, may give them some cover on that.
If you judge this by the way they judge human drivers, then given what police are saying, they would not have faulted or charge with a crime the human driver and don't think a civil suit would get very far with human driver if they were reasonably close to speed limit, not impaired, not texting etc.
I really want to know how long this woman was into the street before the car hit here - especially given the car didn't even try to stop, if she darted into the street at last second as car drove by, that's one thing, but given she was walking her bike which may have had several things packed on it (they said she may have been homeless - so that's probably what would give them that impression) - I have a hard time believing she just quickly moved several feet from a curb into traffic lane.
Seem more likely it was dark, she wasn't clearly a pedestrian to the cars sensors, she was crossing where there was no light and stepped into traffic, but still had to have been some short period of time where the car should have "seen" an object in the traffic lane with its sensors better than human eyes would in dark place. If she stepped into traffic seconds before hit by car, there may have been no time for the car to stop and avoid here, but I have a hard time imagining that she darted so quickly that the car had no ability to react at all.
I have seen people step in front of car in a way no human or machine could hit the brakes before hitting them, like running out between parked cars in a panic (a kid loose) and a saw a pre-teen get caught in center of a road walking one way and the quickly spin around and run other way into a car - but I don't think woman walking a bike at night is such a situation.
I really wonder if she had just stepped off a curb or was well into the road.
If you judge this by the way they judge human drivers, then given what police are saying, they would not have faulted or charge with a crime the human driver and don't think a civil suit would get very far with human driver if they were reasonably close to speed limit, not impaired, not texting etc.
I really want to know how long this woman was into the street before the car hit here - especially given the car didn't even try to stop, if she darted into the street at last second as car drove by, that's one thing, but given she was walking her bike which may have had several things packed on it (they said she may have been homeless - so that's probably what would give them that impression) - I have a hard time believing she just quickly moved several feet from a curb into traffic lane.
Seem more likely it was dark, she wasn't clearly a pedestrian to the cars sensors, she was crossing where there was no light and stepped into traffic, but still had to have been some short period of time where the car should have "seen" an object in the traffic lane with its sensors better than human eyes would in dark place. If she stepped into traffic seconds before hit by car, there may have been no time for the car to stop and avoid here, but I have a hard time imagining that she darted so quickly that the car had no ability to react at all.
I have seen people step in front of car in a way no human or machine could hit the brakes before hitting them, like running out between parked cars in a panic (a kid loose) and a saw a pre-teen get caught in center of a road walking one way and the quickly spin around and run other way into a car - but I don't think woman walking a bike at night is such a situation.
I really wonder if she had just stepped off a curb or was well into the road.
Who is online
Users browsing this forum: No registered users and 1 guest