Self Driving Cars?

[explanade]

Sure because current Ubers are personal properties of the drivers who keep it clean.

In fact that's one of the draws of Ubers over taxis.

But what happens when SDCs are taking a dozen passengers or more per day with no cleaning in between the rides?

 

[Music Lover]

Imagine if most people didn't have to own a car or pay for repairs and fuel, didn't need a garage or to shovel your car out of the snow, didn't need a driveway, and didn't even need a driver's license.

Everyone will have to still have to pay for depreciation, repairs, fuel, insurance, maintenance, parking, etc. It will just be charged on a per mile basis. Nothing is free.

If no one really "owns" the car, then all the things that car owners now do must be done by someone else:

Who authorizes maintenance and repairs?
Who pays for maintenance and repairs? Tires, brakes, etc.
Who decides where the maintenance and repairs will be done?
Who takes care of insurance, licensing, etc.?
Who cleans the car...both inside and out?
What happens if it doesn't start one day maybe due to a dead battery? Will it just sit there unattended?

 

[Midpack]

Everyone will have to still have to pay for depreciation, repairs, fuel, insurance, maintenance, parking, etc. It will just be charged on a per mile basis. Nothing is free.

If no one really "owns" the car, then all the things that car owners now do must be done by someone else:

Who authorizes maintenance and repairs?
Who pays for maintenance and repairs? Tires, brakes, etc.
Who decides where the maintenance and repairs will be done?
Who takes care of insurance, licensing, etc.?
Who cleans the car...both inside and out?
What happens if it doesn't start one day maybe due to a dead battery? Will it just sit there unattended?

While we'll still pay, shared vehicles will have considerably higher utilization than all the privately owned vehicles today. So while we'll still pay for all those expenses, it will be less. Cars will turnover faster too, so the average age of cars on the road will be 'newer.'

Of course someone will own the cars and authorize work. Owners may be private Uber/Lyft, automakers, taxi services and/or public municipal public transportation or others.

 

[Music Lover]

While true, shared vehicles will have considerably higher utilization than all the privately owned vehicles today. So while we'll still pay for all those expenses, it will be less. Cars will turnover faster too, so the average age of cars on the road will be 'newer.'

The only cost that goes down with multiple users would be insurance/licensing as it's spread out over many users. However, as far as I can see (and I may be wrong), is that every other cost will remain the same.

For example, a $25,000 car that lasts 4 years and has a residual value of $5,000 needs to expense $20,000 in depreciation, a couple dozen oil changes, 3 or 4 sets of tires, a few trips to the shop, etc., in just a 4-year period as opposed to about 12 years.

 

[NW-Bound]

One thing I'm wondering is what a self-driving car will do if it can suddenly not see what's around it. The EyeSight feature on my Subaru disables itself when I drive into very dense fog or near-white-out snow conditions. If the car can't see, it can't react. It can't pull over the the shoulder if it can't find the shoulder. It can't just stop, because it might get rear ended. When I've encountered these situations, I've been able to see well enough to proceed. But maybe my vehicle is just being conservative when it warns me, or maybe a fully self-driving car will have more sophisticated vision and will be able to see better than a human in any condition.

Some companies are working only on cameras while others are using LIDARs and other sensors.

People like to say self-driving cars are safer than human drivers because of the indefatigable and fast-acting computers. They miss the difficulties developers have been having: the limitations of the sensors that allow the computer to see.

All self-driving cars use vision cameras. They are cheap, and are necessary for the computer to read traffic lights, road signs, hand gestures by a bicyclist, a traffic cop, or a road maintenance crew, etc... They are needed to read colors painted on the curb for parking permission, parking lanes, etc...

Supplementing the vision cameras are other sensors, such as the Lidars always visible on Google car roofs, millimeter-wave radars, and ultrasonic sensors. Each of these has different limitations. For example Lidars are excellent to map out the obstacles and other vehicles in the surrounding, but they have problems under certain weathers, such as fog, rain, dust storm, etc...

Ultrasonic sensors are dirt cheap, and good for close-range detection. Hence they are found in cars with self-parking features. A self-driving car should have these, because a Lidar on the rooftop cannot look down to obstacles close to the wheels.

Millimeter-wave radars are good at determining speed relative to other vehicles, but do not have good resolution like a Lidar.

The Tesla that drove under a semi-trailer, had its top sheared off and killed its inattentive driver, had only a vision camera that failed to distinguish the white trailer against the white cloudy sky. It has a millimeter-wave radar, whose output was ignored because it could not tell a tall trailer from an overhead highway sign.

So, a combination of sensors will be needed, plus very sophisticated software to combine all sensors together to paint a picture of the surrounding environment. A fast computer is essential, but the limitations right now are seemingly in the sensors.

 

[Midpack]

The only cost that goes down with multiple users would be insurance/licensing as it's spread out over many users. However, as far as I can see (and I may be wrong), is that every other cost will remain the same.

For example, a $25,000 car that lasts 4 years and has a residual value of $5,000 needs to expense $20,000 in depreciation, a couple dozen oil changes, 3 or 4 sets of tires, a few trips to the shop, etc., in just a 4-year period as opposed to about 12 years.

In your post, you mentioned "depreciation, repairs, fuel, insurance, maintenance, parking, etc. It will just be charged on a per mile basis. Nothing is free."

• The rider will pay full price for fuel, one way or another.
• But depreciation, repairs, insurance, maintenance and replacements will be spread among all riders. Some expenses may be higher per vehicle (insurance I'd guess), but it'll still be less per rider. OTOH you won't bear full depreciation, only your fair share.
• Parking will vary, future riders who pay parking on a private vehicle now will pay less, but other riders will bear some of the parking costs.

Overall riders will pay less than they would if they owned the car themselves, significantly less if there's competition. Otherwise, they'll just buy their own dedicated self-driving cars.

 

[Music Lover]

^^

I'm not trying to be difficult, but I only see a savings on insurance. Maintenance, repairs, and depreciation, etc., although shared will still cost the same "per mile" regardless of whether there is 1 user or 20.

 

[Rustic23]

Put me in the can't wait column. But I'm not sure I go for the no pedal or wheel. I see it can get me to a point, but how do I get in in the garage? Or move it to one side of the garage or the other?

 

[meierlde]

Put me in the can't wait column. But I'm not sure I go for the no pedal or wheel. I see it can get me to a point, but how do I get in in the garage? Or move it to one side of the garage or the other?

Or how do you go offroad? (like perhaps to load something) Or how does the self driving car work with 4 inches of clean snow cover?

 

[Music Lover]

How would one distinguish between a puddle and a pothole filled with water?

 

[NW-Bound]

When we have a real self-driving car, it will have to be smart enough to read traffic signs, parking signs posting permitted hours to park, to brake when waved down by someone in distress by the side of the road, etc... You can converse with the smart car, and it will understand you.

Same as with Android phones which you prompt with the command "OK Google", you will start with "OK Car".

You say "OK Car, take me to the shopping mall on Broadway ".

Car: "Which entrance do you want me to drop you off at?"

You: "The one closest to Sears"

Car: "Sears closed down long ago. Tell me what you want to buy, I can make some recommendations"

Put me in the can't wait column. But I'm not sure I go for the no pedal or wheel. I see it can get me to a point, but how do I get in in the garage? Or move it to one side of the garage or the other?

You: "OK Car, park to the left side of the garage"

Car: "You have junk on the left side. First, you need to go out to move it"

Or how do you go offroad? (like perhaps to load something) Or how does the self driving car work with 4 inches of clean snow cover?

You: "OK Car, go off road here"

Car: "Are you crazy? It rained hard yesterday, and the shoulder is muddy. I will get stuck and you will have to get out to push me back on the road. Are you sure you want me to do this?"

How would one distinguish between a puddle and a pothole filled with water?

No, a self-driving car cannot, the same way a person cannot. To decide to drive over it or not depends on many factors. Maybe a smart car can eventually consider all these factors in making its decision.

 

[GrayHare]

"Open the garage doors, please, HAL."

 

[mh]

I agree it will start out that way, but it won't end that way - nearer the end of the evolution to self driving [-]cars[/-] vehicles. For the reason you state in blue above, people who want to drive themselves will eventually be forced off the road because they will be 10 times (estimated) more dangerous than self-driving cars. And parking spaces will be scarcer eventually. And when manual driving is discouraged (or outlawed), fewer people will want to own cars - it will be a niche activity like horseback riding is today, maybe not allowed on public roads in the distant future (speculating?).

There are still a lot of people who are imagining self driving cars will be less safe, and offering up exceptions as "proof." I'm still waiting for one of them to offer up data to support self driving will be less safe. I know being effectively forced off the road doesn't seem possible to today's avid car owner/drivers, but...

...undoubtedly there was a period when horses & carts and automobiles shared the roads. But I'm not seeing horses on the roads much anymore.

And folks in remote areas just aren't going to have the same level of services as urban dwellers, just as they don't have access to taxis, Uber, Lyft, etc. today. It has to cost more to serve folks in remote areas, should urban dwellers absorb that cost?

This will probably be served by formation of local coops much like telephone and internet services were done in many remote / rural areas in the earlier days of those technologies

 

[meierlde]

I agree it will start out that way, but it won't end that way - nearer the end of the evolution to self driving [-]cars[/-] vehicles. For the reason you state in blue above, people who want to drive themselves will eventually be forced off the road because they will be 10 times (estimated) more dangerous than self-driving cars. And parking spaces will be scarcer eventually. And when manual driving is discouraged (or outlawed), fewer people will want to own cars - it will be a niche activity like horseback riding is today, maybe not allowed on public roads in the distant future (speculating?).

There are still a lot of people who are imagining self driving cars will be less safe, and offering up exceptions as "proof." I'm still waiting for one of them to offer up data to support self driving will be less safe. I know being effectively forced off the road doesn't seem possible to today's avid car owner/drivers, but...

...undoubtedly there was a period when horses & carts and automobiles shared the roads. But I'm not seeing horses on the roads much anymore.

And folks in remote areas just aren't going to have the same level of services as urban dwellers, just as they don't have access to taxis, Uber, Lyft, etc. today. It has to cost more to serve folks in remote areas, should urban dwellers absorb that cost?

Re Horses on roads it depends on where you drive in Amish areas you see a lot of buggies. In Tx around rodeo time you see folks out riding their horses. Or other times when the want to give the Horse real exercise, and don't live on a big ranch.

 

[meierlde]

Note that one area where the self driving car has problems is seeing bicyclists. as reported on in the media.

 

[growing_older]

When we have a real self-driving car, it will have to be smart enough to read traffic signs, parking signs posting permitted hours to park, to brake when waved down by someone in distress by the side of the road, etc

There may also be several rounds of learning for cars to distinguish between those friendly people waving down a car for assistance, and the future equivalent of the highwayman or protesters who learn what tricks will get self-driving cars to stop when a human driver might not want to. Lots of difficult areas here.

 

[Midpack]

^^

I'm not trying to be difficult, but I only see a savings on insurance. Maintenance, repairs, and depreciation, etc., although shared will still cost the same "per mile" regardless of whether there is 1 user or 20.

Depreciation is not the same cost per mile. Mileage influences depreciation but the correlation coefficient isn't 1.0, not even close. If you don't drive your car for a year, it will still depreciate. If you drive your car 24K miles this year, it will depreciate a little more than the same car driven 12K miles in a year - but not twice as much. There's data all over the internet.

The relationships may well change in the self driving era, but there's no precedent to suggest it will be as simple as 'per mile.'

I'll let you think through the others, also not strictly correlated to miles.

 

[savory]

Note that one area where the self driving car has problems is seeing bicyclists. as reported on in the media.

I have also read this but expect it will be solved. At minimum, i expect a device could be attached to a bike to improve its visibility.

Also road configurations will change dramatically since narrow lanes will be easier to manage with computer systems. Ultimately, and if required, a Dutch inspired bicycle road system could be introduced separating motor vehicles and those using Active Transportation.

 

[Amethyst]

You have just pointed out why auto-autos are not going to be a panacea for "the elderly and infirm" to continue doing things independently, as everyone keeps saying.

They will face the same frustrations that keep many of them from using smartphones ("I can't get this $!^* thing to work!") and then there's still the "last tactical mile" (getting out of the car and navigating to the desired place and back to where the car will pick them up.)

We'll still need robot carers.

You say "OK Car, take me to the shopping mall on Broadway ".

Car: "Which entrance do you want me to drop you off at?"

You: "The one closest to Sears"

Car: "Sears closed down long ago. Tell me what you want to buy, I can make some recommendations"



You: "OK Car, park to the left side of the garage"

Car: "You have junk on the left side. First, you need to go out to move it"



You: "OK Car, go off road here"

Car: "Are you crazy? It rained hard yesterday, and the shoulder is muddy. I will get stuck and you will have to get out to push me back on the road. Are you sure you want me to do this?"



No, a self-driving car cannot, the same way a person cannot. To decide to drive over it or not depends on many factors. Maybe a smart car can eventually consider all these factors in making its decision.

 

[Totoro]

Who authorizes maintenance and repairs?
Who pays for maintenance and repairs? Tires, brakes, etc.
Who decides where the maintenance and repairs will be done?
Who takes care of insurance, licensing, etc.?
Who cleans the car...both inside and out?
What happens if it doesn't start one day maybe due to a dead battery? Will it just sit there unattended?

The fleet owner - just like the bus company today. It's no different.

 

[Midpack]

When we have a real self-driving car, it will have to be smart enough to read traffic signs, parking signs posting permitted hours to park, to brake when waved down by someone in distress by the side of the road, etc... You can converse with the smart car, and it will understand you.

I'm not pretending I can predict the future, but eventually there won't be any need for traffic signs or parking signs, that'll all be built into the nav system or there will be sensors at intersections we may not even see. If signs still exist, it'll be as a courtesy to human riders, but eventually riders won't even need to know or care.

When horses and carts still dominated the "roads," I'm certain most people were convinced cars would be dangerous and unworkable because they viewed the world from the perspective of horses. Thank goodness there are some people who don't let all the unimaginative naysayers discourage them.

It takes some imagination to even begin to see the future, and there will still be many big, unanticipated surprises.

 

[Totoro]

We'll still need robot carers.

Yes, but a handful of them and they will be remote. Car doesn't understand instructions => get connected with an operator. Human in some instances, AI in others.

Parking in a specific spot? You don't verbally instruct the car, you point on the display exactly where you want it. Visually, the car will point out the obstacles and/or whether it accepts the instruction.

Off the top of my head.

 

[Totoro]

I'm not trying to be difficult, but I only see a savings on insurance. Maintenance, repairs, and depreciation, etc., although shared will still cost the same "per mile" regardless of whether there is 1 user or 20.

Cost of current gas car 100%, typically roughly breaks down as follows:

• Depreciation: 35%
• Fuel: 35%
• Insurance / repairs: 15%
• Maintenance: 10%
• Tires: 5%

All of these have a variable portion and fixed portion. Fuel is 100% variable, depreciation roughly 50/50 with current cars at typical mileage (midrange cars), tires is more 90% variable.


With an electric self-driving vehicle and very high mileage the following can happen:

• Depreciation: fixed part goes to 0% with high use. The variable part will go up (battery downgrading outweighs fewer mechanical parts), 30% less depreciation isn't out of bounds. 23% left.
• Fuel: probably -50%, potentially even -75%. 18% left.
• Insurance: -90%. 1.5% left.
• Maintenance: -50%, electric is simpler and a fleetowner can drive efficiencies. 5% left.
• Tires: Probably drops a little bit: less careless driving, better maintenance (keep the tire inflated). Say -20%. 4%

So your cost goes from 100% to 23% + 18% + 1.5% + 5% + 4% = 51.5%, call it half. Before actual sharing, parking costs, speeding tickets etc ..


Two caveats apply: this is a very rough approximation, I know that. And we didn't consider a big factor: Taxes!


Regarding depreciation specifically: I think it's probably underestimated how far the depreciation factor will decline. Vehicles can have a very long lifetime, trains, planes and buses (shared fleets) last 30+ years easy. Most current depreciation in a car is more fashion driven than utility driven.

 

[meierlde]

I'm not pretending I can predict the future, but eventually there won't be any need for traffic signs or parking signs, that'll all be built into the nav system or there will be sensors at intersections we may not even see. If signs still exist, it'll be as a courtesy to human riders, but eventually riders won't even need to know or care.

.

But what about folks on bicycles, walking, running, on skates, scooters etc.
the signs are still needed.

 

[Totoro]

How many road signs do you see in an area where no cars are allowed? Usually zero. Almost all road signs are for use of motorists only.

Traffic lights and crossings probably will have more staying power.