Cringely Predicts: Professional Drivers With Drone Landing Platforms (cringely.com)

In what may be his final year of technology predictions, columnist Robert X. Cringely argues aerial delivery drones "are definitely coming just as fast as regulators will allow them, but I don't think they'll be implemented in the way people expect." As soon as autonomous systems can be shown to be as safe or safer than human pilots, they'll take over most drone piloting duties... Here's the problem with Pizza-to-the-Home: where does the drone land at your house that won't risk hitting a child, pet or vehicle and also won't risk losing the delivery to theft or damage? We can't economically mandate a drone landing tower for every house that's above obstacles and with a guaranteed clear approach.... But we CAN mandate such a landing platform on top of every pizza delivery vehicle.

Using GPS, the drone and car can find each other with the drone landing only when the car is stopped and the approach is clear... [F]or that driver each delivery will take five minutes or less. Pizza is delivered faster and hotter and the driver, instead of making 2-3 deliveries per hour, can make 10-12. This is what we'll shortly see proposed for drone delivery, not just for pizza but for everything else...

Now here's where Internet-style disintermediation comes into play. Such a drone delivery network still costs money to build but that money will be instantly available if the class of goods that can be delivered expands beyond food to anything weighing under, say, 10 pounds. This means prescription drugs and even Amazon Prime or walmart.com packages can arrive on the same car, delivered to that car by multiple drones and drone networks. All it requires is WAAS GPS and a standardized car rooftop landing platform, which I am sure we will shortly see.

Comments

Re:Fuck Drones. Target practice

[Real Data Collection]

Considering that the drone pilot will most likely be a ex-CIA drone pilot with years of experience in blowing up Afghanistan wedding parties, they will use YOU for target practice and still deliver the pizza in 30 minutes or less.

Aspiring to stopped-clock accuracy

[radarskiy]

Has anything printed under the "Robert X. Cringley" nom de plume ever been correct?

You will print a target

[drinkypoo]

The target will be a QR code. You will agree on the rough location of the target during the process. The process of downloading and printing the target will also include agreeing that you're responsible for putting it someplace sensible.

If the drone gets there and it doesn't look like a good place to drop a pizza, you will have to go somewhere else to get your pizza.

Nowhere in this process will there be a driver, except to pick up failed drones. That person can be the assistant manager.

Re:You will print a target

[raftpeople]

Or you will buy a sturdy-ish reusable mat with a preprinted QR code. You scan your landing pad when placing order.

Re: You will print a target

[Lanthanide]

What training? Just have a human from the company visit each house once to confirm the spot is acceptable.

Then no more humans are needed ever again.

Re: You will print a target

[Jane Q. Public]

No.

People keep forgetting probably the most important aspect of any kind of flight: knowing where the other planes (aerial vehicles) are.

The major limitation of aerial delivery systems is not the landing zones, it's the flight path and avoiding other aerial vehicles.

Such a system will not only need a sophisticated shared information system with real-time, accurate GPS coordinates from each drone, but be able to communicate information about potential collisions to drones in flight.

Unlike most car crashes, people in the area can't just step on the brakes to avoid the collided vehicles. They'd be falling out of the sky.

Rarely, maybe, but very dangerously.

Re:You will print a target

[drinkypoo]

Or you will buy a sturdy-ish reusable mat with a preprinted QR code. You scan your landing pad when placing order.

I like this, it should be larger than most people can print anyway. Maybe they could make a pizza box that unfolds into the target. That reduces manual deliveries to one. Include an insert with a foil or plastic protective layer to contain grease.

Re: You will print a target

[drinkypoo]

The major limitation of aerial delivery systems is not the landing zones, it's the flight path and avoiding other aerial vehicles.

This could be solved with any combination of transponders, with centralized traffic control managed via cellular network, and actual sensing and avoidance hardware. The combination of all three (and maybe something I haven't thought of) would be the most intelligent, but also the most expensive. It takes relatively little logic for GPS-aware devices to avoid one another if transponders are used. Granted, that information can be used to attack drones, so centralized management is probably the best answer — even though it can be used to attack drones. And if I were spending any significant amount of money per device, I'd want them to be capable of making some decisions on their own. For example, is the landing surface really flat?

Re:Swing and a miss

[smoothnorman]

My wonderment exactly, (i think "Cringley Predicts" has slipped a gear). Makes only slightly more sense to have the homeowner's drone meet the pizzeria's drone and pass the pizza like an aerial lateral pass, (Then an intercept! Then pass interference! uzw...)

Yeah right

[Anonymous Coward]

"Pizza is delivered faster and hotter" - sure. It is on a drone, flying outside in freezing temperatures. It is no longer in a nice warm vehicle. Sure, they will put one of those insulator things around it. But it will not be warmer than it would have been in that same insulator in a warm vehicle. That's just incorrect. The whole idea is stupid though. Who is going to put up with the noise pollution from all these damn drones?

Re: Yeah right

[Elfich47]

That is why cars have strict laws for muffler sizing.

Pizzas aren't drugs

[petes_PoV]

The problem with this guy's guess is that pizza delivery is time-critical (while it's still warm - measured in minutes) whereas all the other stuff can take an hour, two or all day. It is very unlikely to matter. And since a ground based vehicle can carry a hundred times as much weight as a drone, it only needs to "fill up" once and then do its rounds.

But the density of pizza deliveries is the limiting factor. At any given time there are not likely to be more than 1 pizza per square mile (as different people will order at different times) so what takes the greatest amount of time is getting the delivery van to the correct location. Whether to deliver the pizza directly or simply to receive it from the drone.

The scheme fails.

No.

[Gravis Zero]

Here's the issue, vehicles are owned by the drivers themselves, not the pizza place. Don't expect drivers to install a huge-ass drone platform on their car and provide their own drone or for pizza places to suddenly invest in vehicles and/or drones.

There is no driving incentive to reduce delivery time or do away with driver tipping (that's less money for the driver, so yo have to pay them more), so it's not happening. The only way this makes sense is if they can reduce the number of pizza places needed to serve an area but this is countered by the fact that traffic limits the area that can be served.

These guys ("Robert X. Cringely") know nothing about running a pizza business.

Re:No.

[Waffle Iron]

Don't expect drivers to install a huge-ass drone platform on their car

Well, thanks to the miracle of magnetic fields, many of them already put huge-ass lighted signs on top of their cars.

Re:No.

[FrankSchwab]

So let's explore this. The Pizza parlor invests in nine 1.5 meter square pieces of plywood painted with a high-contrast landing pattern, with suction cups and straps to tie them onto drivers cars. Let's call that $500. They invest in a dozen big drones capable of carrying, say, four extra-large meat lovers pizzas in an insulated pouch - Let's call that $25000, dwarfing the cost of the landing platforms.

Now, the Pizza parlor hires nine drivers for a Friday evening, straps platforms onto their cars, and sends each to a different area surrounding the parlor. An order comes in, the pizza comes out of the oven and gets popped into the heated box under a drone, the drone goes and finds the closest driver. The driver may be in front of the desired house at the moment, or may be at the previous house - the drone lands, driver moves pizzas to his front seat, and delivers them to the desired house and collects his tip. If the driver is in motion, he pulls into the nearest parking lot, waits for the drone to land, and collects the pizzas.
From the Customer's point of view, nothing changes in the current pizza-delivery model except their pizza arrives in 15 minutes instead of 45, and is likely hotter when it gets there. From the Driver's point of view, they deliver more pizzas per hour with fewer miles driven. From the Pizza Parlor's perspective, they've made a huge capital investment, but they're delivering 2-3x the pizzas they used to. From everyone else's perspective, there's a constant stream of annoying drones flying overhead (and occasionally crashing into their neighborhood) destroying their ability to peacefully enjoy their backyards. I guess that's an externality that just doesn't need to be considered.

Re:No.

[MrL0G1C]

"From everyone else's perspective, there's a constant stream of annoying drones flying overhead "

I live on a busy road, I bet these things would be no more noisy than the constant traffic, vehicles are noisy even when only travelling 25mph. Drones could fly a bit higher to cut noise pollution.

Separate note, the summary is dumb, landing pads wouldn't need to be mandated, you just say: If you don't have a landing pad available that meets specifications xyz then you can't legally have drones delivered to you and/or drones can only deliver to sanctioned landing pads. And who can't stick up a flat bird table like thing for the fast food to be dropped on to, not difficult.

Re:No.

[jdschulteis]

From everyone else's perspective, there's a constant stream of annoying drones flying overhead (and occasionally crashing into their neighborhood) destroying their ability to peacefully enjoy their backyards. I guess that's an externality that just doesn't need to be considered.

That's why Amazon is going to have their drones fly in at altitude, balloon down to complete the delivery and back up, out of the annoyingly loud range.

Re:No.

[angel'o'sphere]

These guys ("Robert X. Cringely") know nothing about running a pizza business.
Perhaps he should read Snowcrash?

But it is funny, to watch how the "drone industry" is searching for a problem to solve.

I doubt we ever will have something that is worth building up a drone delivery infrastructure.

Re:No.

[CanadianMacFan]

And who can't stick up a flat bird table like thing for the fast food to be dropped on to, not difficult.

People in apartment buildings and condos? I'm not going to order and then head downstairs with a landing pad to wait for the delivery which could take 10 to 30 minutes (or longer if the place is busy). Then you have to bring the order back up along with the landing pad. Maybe the building would have a permanent landing pad to save you the bother of having to take one down with you. But you would still have to go down early because who knows how long it takes for an elevator to come so you can't go down when you get a message saying the drop is just about there. It's a pain in the butt when right now it gets delivered right to your unit.

Every time I see one of these articles it assumes that everyone lives in houses.

Re:No.

[drinkypoo]

Amazon is going to have their drones fly in at altitude, balloon down to complete the delivery and back up, out of the annoyingly loud range.

No they aren't. It's a cute idea, but impractical for a variety of reasons. Quads really aren't that loud anyway.

Re:No.

[Gravis Zero]

They invest in a dozen big drones capable of carrying, say, four extra-large meat lovers pizzas in an insulated pouch - Let's call that $25000, dwarfing the cost of the landing platforms.

$25000 for a dozen? That would be a steal! To carry that much weight, you would be lucky to get one drone for $25000. That doesn't even cover the cost of insurance for the drones.

From the Pizza Parlor's perspective, they've made a huge capital investment, but they're delivering 2-3x the pizzas they used to.

Except they are already meeting demand (if they need more drivers then they hire more), so they aren't going to get more sales, just reduce the number of drivers. So your investment in drones is really just a way to have fewer drivers. Drivers are cheap but these drones are expensive. One lost/damaged drone is going to cost a LOT of money. It's a risky proposal.

Re:No.

[Gravis Zero]

I doubt we ever will have something that is worth building up a drone delivery infrastructure.

They are actually good for inspections, so they do have legitimate uses. However, they have been found to be very useful for illegitimate purposes. It's currently being used to reach nearly inaccessible locations like prisons. However, in the future I could foresee them being used to drop off drugs to avoid having a physical presence.

Re:Swing and a miss

[Brett Buck]

Cringely?

https://en.wikipedia.org/wiki/...

Professional Drivers == Pizza Delivery?

[TexasDiaz]

When did Pizza Delivery people get classified as "Professional Drivers"? How long do you have to be delivering pizzas to be considered "professional"? I drive to work every day to my job - does that make me a professional driver, too?

Re:I don't see it happening.

[aix tom]

I'm sure there is a way to fix that.. Perhaps put tubes underground for the drones? Perhaps the droned could then even be propelled by compressed air? I'm sure that will be the NextBigThing(TM)

https://en.wikipedia.org/wiki/...

a car ? isn't the point to eliminate the car ?

[cats-paw]

if the vehicle is remotely piloted, then the "pilot" can very easily check to see if your beloved pet is in the way using the camera feed.

also too, a landing pad high up clear of obstructions is not necessary. a very plan target, in white or other solid color, so that on object would stand out. Even an automated check system, will probably work. AI can give you a definite "no", but a person is required for a "yes".

Drones are coming, that's for sure.

So is incredible unemployment. The 2nd gilded age is here, and it's got climate change as a backstop. This is not going to end well.

Re:Swing and a miss

[smoothnorman]

There are multiple definitions. You simply chose an inappropriate one. Try this one from the OED: "(n) a feeling aroused by something strange and surprising". As in having drones visit a car which then completes the drone's delivery [shrug]

Re:Swing and a miss

[Slugster]

Yea his reasoning is kind of odd... especially since the self-driving cars that have already been tested for food delivery... are technically 'drones' themselves...

Before you bother with flight at all, note that there is an immediate efficiency to be gained, by building driver-less motorcycle trikes for food delivery. Less weight, better fuel efficiency than a car. And we haven't even gotten self-driving cars to work 100% of the time yet...

The entire desire for airborne commercial/residential drone delivery is terribly misguided IMO. It is very hazardous from a liability standpoint.
It is somewhat like saying "since helicopters are easy to build now, we should get rid of all the cars and just use helicopters instead".
Helicopters are wonderful things, but only if you 1) have a very high priority task that can bear a high cost, or 2) if you are really really rich and like burning money.

UPS/FedEX should work like this too. Do Residential deliveries at night, and instead of someone walking the package to my porch, the UPS truck just calls me as it rolls up to my place, and I walk outside and get my own mail.

Yea but--even if they got an advance warning--a lot of people would be watching the end of their favorite tv show instead, taking a dump, talking on the phone or whatever and not show up at the curb.

And anyway (where I live, midwest US) DHL drivers sometimes do this, if they have your phone number and are delivering a package that is signature required. They call or text you a few minutes ahead of their arrival.

Re: Drone pirate

[TheMeuge]

That's a stupid idea. 22LR has a huge range and can potentially kill someone from 1000 ft away if you miss. Use birdshot out of a 410 shotgun shell. If the drone aims > 30 degrees above the horizon there's no chance of collateral damage. In a couple hundred feet it'll be basically sand.

Re:A consummate gentleman and absolute professiona

[Anonymous Coward]

Robert X. Cringely. *facepalm* How is this clueless, know-nothing douchebag still getting attention from anyone?

And he's not even the original Robert X. Cringely, although he somehow managed to wrangle some sort of legal agreement allowing him to claim that he is.

Customer car target

[manu0601]

And what about using the customer's car rooftop as landing station?

Re:Customer car target

[CanadianMacFan]

Not everyone has a car and not everyone has a house to go along with their car.

This system is designed for people who have houses and now you want to limit that to people who have houses and cars. I would say that the people without cars should be the ones that need delivery more than the ones with the cars.

The Powers That Be..

[betsuin]

So how in F do these drones fly legally?

Here in the Land-of-Oz a certain department (CASA) mandates where you can fly your favorite muti-rotor-copter.
There's even an App for it. Trust me, it's bloody restrictive as to where one can take off as "they" fret over real helicopters getting struck.
Got a helipad within 5 klicks? Forget it. Flying over private property - no way! Just as a couple of examples.

Where are we going to land?

[Elfich47]

As the observation was made to me: If the car is not parked on a flat surface, how is the drone going to land on the car?

I can think of plenty of places where city streets are on hills. Plus throw in weather (rain, sleet, snow, freezing rain) and a drone landing on a car that is parked on a slope is a drone that is crashed in the street and run over by a car and then everyone has lawsuits and insurance payouts.

This need to go back to the drawing boards with a bit more reality attached.

Re:Where are we going to land?

[CanadianMacFan]

Electromagnets. Wouldn't need to be that strong so they wouldn't eat up too much battery life. All controlled by the drone. Turned on when it lands and as it takes off the electromagnets are turned off.

Since there's a driver any major accumulation of snow, freezing rain, etc would be taken care of by the driver so that's not a problem.

Re:Where are we going to land?

[Elfich47]

That is very case by case basis. There are vehicles using aluminum now as parts of their exterior. So the drone landing and expecting a magnetic grapple is going to need all sorts of safeties in the landing software to ensure that there is a firm grapple before the rotor blades spin down.

I actually see the snow being a real problem: target recognition. Requiring drivers to get out of their car during a snow storm to clear the targets mounted on the roof so the drone can see the landing target? During a heavy snow fall the driver would have to be out of their car constantly clearing it to ensure the drone has a clean target. That is part of the reason why helicopter landing pads use automatic snow melting systems to ensure their pads are clear.

Re:Where are we going to land?

[CanadianMacFan]

Aluminum in cars isn't a problem since the driver puts a landing pad on the roof and the landing pad would have some light metal in it in order for the electromagnets to attract to.

As for the driver having to constantly clear off the landing pad, that's not an issue either. The drone can send a message when it's a minute or 30 seconds away and the driver can then clear the landing pad. Or the drone will hover when it arrives until the driver clears the landing pad. Not a big issue.

I'm not advocating for this idea. Just that these issues aren't really issues. There's bigger problems with this idea. Especially with the people that are suggesting that the drone deliver to a landing pad that the customer puts out. That's assuming that everyone that's ordering pizza lives in a house. What about people in tall residential buildings or people staying in hotels? Or people getting are at work. Never mind that drones powerful enough to deliver a couple of pizzas would be noisier than a car and I wouldn't want to hear them flying all evening. I'm two blocks from two pizza shops (different directions) and if this technology came into being my neighbourhood would be hearing a lot of drones. Plus there would be the drones from the other delivery places too.

You want UL certification for "drone platform"

[Elfich47]

Because if you want a drone landing platform on a car to be insured, I will bet dollars to donuts that all of that will have to be UL certified. And that means some sort of permanent installation on the car so the installation doesn't get torn off the car through wind resistance, weather or other problems.

insurance

[Elfich47]

This operation will have to be insured. This will be the stumbling block. Until it can be insured and shown to be reliable and not a threat to health and welfare when be operated or moved, it will not be operated. Think of it this wa: how many things have you seen go flying off of a car at high speed?

Re:Swing and a miss

[Elfich47]

To sort out residential versus commercial deliveries would be a huge issue.

Fed Ex and UPS would have to run a second delivery shift in order to have evening delivery. And that means paying adjusted rate to their drivers . And in addition their entire delivery logistics at the delivery stations would have to be rethought, likely the entire building would have to be rebuilt from the ground up. The dock doors for the morning runs have those trucks in them once they come back and are busy being unloaded from their pick ups.

Loading, unloading, and internal sorting logistics would have to be rethought. Right now the parcels come in on the 53' foot trucks, go onto the sorters and are loaded into the trucks. You are talking about having to place all the residential parcels into holding or buffering while the morning deliveries are gotten out of the way first.

And in addition you are talking about a significant cost increase: More drivers, more runs, more trucks to service, more redundant runs. Plus the companies have to figure out if each and every address is residential or commercial. Plus the complication of people running businesses out of their own homes and that means delivery companies would have to deliver to the same address multiple times in the same day - either due to home businesses in apartment blocks, or apartment blocks having public businesses on the first floor. Night delivery is not an item I can see happening right now.

Solved problem, but not this way

[lordlod]

There are a lot of people experimenting in this space. Unsurprisingly this problem has been considered with far better solutions than this suggestion.

The Alphabet Project Wing system for example doesn't land. It remains at a safe distance above the ground and lowers the load.

This avoids any direct interaction between members of the public and the fragile dangerous drone.

The risk of theft is significantly mitigated by being able to schedule delivery times to correspond to when you are at home. No need to leave a product outside for half the day, you know and can collect it the minute it arrives.

High Tech

[cstacy]

So rather than having my pizza delivered by the pizza delivery guy in his car, pizza delivery guy will drive his car to my house and wait for drone to appear out of the sky with pizza.

They will use machine learning AI to figure out where to pre-position the landing sites (pizza delivery guys). Pizzas will be 3-D printed and accounted for using blockchain. .....

Why land?

[aggemam]

Why land at all? It could hover in front of a window (e.g. outside the 20th floor in a highrise) and you grab the goods off of it.

Not pizza. Platform as a service.

[Fencepost]

People are focusing on the economics of this for a pizza place, and that's the wrong way to go.

A ton of delivery cost is "last mile" (not literally, basically from the facility to the door). This is talking about dividing that into two pieces - "to the door" from a mobile landing platform so basically the last 30-100 feet and "to the platform" which is most of that so-called last mile. The Platform As A Service driver or vehicle doesn't have to cover all that range back to the facility - just the quarter mile to the next delivery in the area.

Viability for this is going to depend on density of deliveries, but you actually see a variation of it with UPS for holiday deliveries when they switch to having a driver with a truck full of packages plus 1-2 runners who actually take things to the doors. If you think of that UPS truck as a platform constantly being refilled by drones, you have what he's getting at.

Re:Swing and a miss

[Bandraginus]

And no it's not going to hover over you and drop the pizza.

I beg to differ