Wal-Mart seeks to test drones for home delivery | Hacker News Reader

Wal-Mart seeks to test drones for home delivery

reuters.com

37 pointsrobandrews44 comments

I'm curious because every battery operated "drone" I've seen has a pretty limited range and flight time. Some of the higher-end multicopters built for videography can carry a decent camera for maybe a half hour but it seems like for deliveries, you'd be really limited in terms of weight carried and distance. The thing still has to get back to the warehouse even if it manages to land, drop the package, and fly back without hitting anything.

I say this as a big fan of hobbyist multicopters as well. The more automated the solution, the less useful and feasible it seems to me. I guess you could have someone at the "home base" watching a camera feed over LTE and maybe the latency wouldn't be so bad but just in terms of logistics, I can't see how some battery-powered DJI multicopter would be of much use outside of some edge cases.

Still, I guess it's cool to see companies experimenting. Sometimes you learn one thing by trying another. And on the surface it really does seem to make more sense to fly a small, battery-powered craft to drop off small items rather than having to drive a car back and forth across town to deliver a t-shirt or a roll of TP.

According to the article, walmart wants to test a drone delivering a package from a truck and then returning to the truck. If that is the case, then they could use an automated truck to deliver packages to the general area first. Then the drone would activate and deliver the package the last few feet to the door. Presumably, if they did something like this the drones would have their batteries charged while the truck was driving to each location.

In the various combinations, I wonder if there's a place for a mini-autonomous van, too small for a person, but can carry a few packages.

The main truck stops regularly along a main road, and releases a few of these tiny vans. The last few feet to the door (tricky to navigate, usually require a "off-road" person) are by drone, as here.

Wow, that could work. A self driving van. Of course, the drone would charge from van.

How about the optimization: as the van is slowing down or looking for a parking spot the drone takes off. Van begins unparking as the drone gets near.

Or in cities, the drone could take the subway.

I worked as a "holiday helper" for UPS during a Christmas season when I was in college. My only job was to take the package to the door for the truck driver while they queued up their next package.

I could see these drone systems working a lot like what I did at that job. Even if the driver still had to drive almost right next to the house, if they could reliably have that "last 100ft" delivery taken care of by drones it would be a big efficiency boost.

Just remember that there are a lot of economic factors pushing batteries to be cheaper and smaller. Elon Musk may actually make drones economically viable, thanks to Tesla's need for the exact same thing- cheap, dense energy storage.

I think Google's experiments with drone delivery have been with fixed wing for that reason. Quadcopters' range is just too bad.

I'm ok with quiet, unmanned, self-driving delivery trucks that automate the package delivery process. I am not looking forward to delivery drones buzzing around overhead becoming a normal occurrence.

They won't. They're an absurd impracticality anywhere outside of Silicon Valley (suburban homes, no weather ever), and completely unnecessary even in those ideal conditions.

People are excited because they're cool toys, not because it actually solves a problem.

When the naysayers abound, you know things are about to get off the ground. Parcel delivery starting in NZ (http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&o...). My bet is that this technology will most likely start small (see: unsexy) and incrementally improve the technology (battery, airframe, autonomy) and regulations, while Amazon and Wal-mart keep promising the moon.

In cities it seems like a good idea. Millions of customers are within drone range in Manhattan. Drop packages on a building's roof, a doorman brings them down to the front desk, there you go. It can take several hours to drive a delivery across Manhattan at rush hour.

As for the range, electric cars mean that batteries are receiving a lot more development and economies of scale, plus people have been working on package switching networks for years already now.

Is it going to happen tomorrow? No. But Walmart can afford to be in it for the long haul.

I can't disagree with you enough. This will be a huge game changer. Sure, it won't change things a ton for the consumer (many times do you really need that item the same day or is 1-2 days okay to wait?) but for the retail companies that adapt this is going to be HUGE.

Yes drones are impractical right now but we have just begun using them (they're still incredibly new). Battery technology will improve, drones will get better. By investing in the technology now you can work through it, find what works best, iterate and repeat with small subsets of customers. When drones become more cost effective and efficient you'll be ready and already serving products this way. Years later you may even close many of your retail stores. Why would you need them? People can use their phone, order something and have it just come to them. Logistics can be reworked incredibly so they can hold more inventory at less locations (thus lowering the overall inventory hold).

It's not going to happen today or tomorrow but 5-10 years? Drones are going to save companies billions of dollars on retail locations. Sounds crazy today sure but don't bet against technology.

> many times do you really need that item the same day or is 1-2 days okay to wait?

How many people use Amazon Prime? Clearly a few days makes a big difference to a lot of people.

> Clearly a few days makes a big difference to a lot of people.

It certainly does hence why I mentioned 1-2 days assuming everyone was someone who does 2 day at the most :)

I'd imagine more people care about getting it in 1-2 days versus 3-5 days but I'm not sure how many care about the time difference in getting it same day. Sure people care but I feel like the bigger business opportunity is slimming down the inventory and retail presence.

"Airplanes are interesting toys but of no military value." -- Marechal Ferdinand Foch, Professor of Strategy, Ecole Superieure de Guerre. 1911

"Anyone who fights, even with the most modern weapons, against an enemy who dominates the air, is like a primitive warrior who stands against modern forces, with the same limitations and the same chance of success." -- Field Marshal Erwin Rommel

When I can get a product faster by ordering online than driving to a store it will be a game changer.

Imagine: saving the 30 minute to 1 hour round trip to a store, or waiting half the time for a driver to wade through traffic at peak hours.

I already save the 30 minute trip to the store by ordering online. So all the drone would offer me is to save a few days shipping time (or, back when I lived in the UK, one day. Oh and supermarket delivery is common there too.), which frankly isn't something I'm going to pay much for most of the time. I only drive to the store if I don't know what I'm going to buy and I want to browse in person.

"all the drone would offer me is to save a few days shipping time"

That's exactly what some (many?) of us want!

> which frankly isn't something I'm going to pay much for most of the time

The drones are going to be cheaper than the UPS guy, not more expensive.

I would honestly be surprised if this was the case, at least for the end customer.

At best we'll see immediate drone delivery as an incremental surcharge or a value-add for a subscription service (Prime).

drones can only get cheaper over time.

employees need to eat in 2015, and 2016 and 2017, etc.

So we get our new gadgets on a few hours for less cost to the company, more dollar-cost to us, and in return we have more unemployed people so pay more taxes.

I'm really not a Luddite; we need to find ways to put the tech development focus on things that benefit society as a whole.

/rant

The cost savings will get passed on to the consumer; retail is a very competitive low-margin industry.

We get our gadgets for less cost and the people that used to deliver them get paid by the government to retrain to do something else that's actually fulfilling and productive.

Win-win.

Amazon are already offering 1 hour delivery [1] in certain locations. I think just London and a few cities in the US. Argos have also rolled out same day delivery [2] for £3.95, if you order before 6pm. This seems to be available for quite a bit of their inventory across the UK. It's not one hour delivery but getting close and the price is pretty good.

1. http://www.amazon.co.uk/b/ref=pn_uk_surl_lp?node=6584642031 2. http://www.argos.co.uk/static/ArgosPromo3/includeName/fast-t...

Indeed. Even ignoring the total impracticality of them, you can also look at just energy issues.

In order for drones to be cost effective, they need to stay in the air for minimum the full trip out and back from the home, but if you look at the location of most big warehouses (e.g. Amazon, Walmart) they're built outside of big cities for cost reasons, so now you have to fly twice as far to deliver packages onto the opposite side of the city.

So either the FAA will need to allow liquid fuel drones (gas, diesel, etc) which have their own safety issues, or they'll need to be extremely large to hold enough batteries. Plus the recharge time of batteries will also be an issue.

I think weather and safety are their biggest detractors, but I think energy problems are next on that list. All these companies keep doing is pumping out press releases but without any specific details about how far their drones can travel, under what conditions, etc.

I'd short sell anyone who announced they're going to invest in making this work. It is fine as a pet project and to get some publicity, but if they start throwing millions behind it, I'm betting against them and it.

> but if you look at the location of most big warehouses (e.g. Amazon, Walmart) they're built outside of big cities for cost reasons, so now you have to fly twice as far to deliver packages onto the opposite side of the city.

90% of all Americans live within 15 miles of a Walmart. If they were able to launch drones from their store locations it might address this concern.

Wow, I assumed that was a guesstimate, but it's the actual stat! Crazy.

http://www.statisticbrain.com/wal-mart-company-statistics/

but if you look at the location of most big warehouses (e.g. Amazon, Walmart) they're built outside of big cities for cost reasons, so now you have to fly twice as far to deliver packages onto the opposite side of the city

Big warehouses are dependent upon current transportation technology (trucks and big cargo ships). If drones become feasible, there's no reason we couldn't move to decentralized (retail-scale) warehousing serviced by drones.

I don't know - removing trucks from the roads sounds great to me. In theory wouldn't it be better to have less on the ground and more in the air? Since we exist on the ground and there's a lot more we can do with ground space.

This assumes the drones would be sufficiently high enough that it wouldn't feel like a bunch of bees buzzing around overhead. Not true today, but eventually...

Granted, I wouldn't want the sky filled with them either even if they were perfectly quiet and efficient and safe. I feel like there could be a reasonable balance though.

Seems pretty inefficient to fly things, because the energy needed to keep it floating in air. Although history tells us never to say never about technology, it seems like having small autonomous robots on wheels. They could travel on the sidewalks in cities, and also go the fastest way straight thru forest (for the many who live outside of cities).

Flying has got to be more efficient than traveling through a forest.

As a person who spends a lot of time in the forest, there's a ton of impassable obstacles. Cliff faces, streams, briar patches, downed logs, etc.

The robot on the ground would spend more energy in the forest than the robot in the sky.

I think being in the sky also helps keep them from being a target for looting.

>I think being in the sky also helps keep them from being a target for looting.

But then a target for shooting :)

While I agree with you that wheels would be more efficient energy wise the problem there is obstacle avoidance (especially if it travels down a long path that is suddenly blocked and it has to return) (thus maybe it's not as energy efficient except for very direct and easy routes) and thievery. Flying solves both of those issues but is expensive energy wise.

I think we'll see this little by little in testing and, maybe in 5-10 years, one of this laboratory "break throughs" in batteries will materialize and drones will be able to fly very far away and this will become cost effective.

I think flying over a forest would be way easier on a robot than weaving it's way through underbrush. Although, a slithering robot would be pretty awesome!

You could say the same thing about aeroplanes.

On a related note Camden Real Estate has suspended free holding delivery packages for the 60K apartments it manages across the USA. They said they stored a million packages at cost of 150K employee hours.

Since the younger computer savy apartment crowd tends to order more online, I expect this issue to grow. Maybe apartments will compete on package amenities.

Walmart seeks to test press releases for free advertising.

They could also use cannons and shells with parachute for delivery in Silicon Valley area. Well, looks like I missed another startup opportunity...

I'd certainly be willing to put up with a few broken windows in exchange for the novelty of having my books arrive via field artillery.

http://www.damninteresting.com/one-small-step-for-mail/

This title would make a great Onion article, where 'drone' is a human being forced to run to each delivery point for 7 dollars an hour.

Amazon must have them absolutely terrified.

My guess would be that they wrote anything they could think of into the application so that they can play around without doing another one.

Given their scale, a few million spent on this or that doesn't say much about what they will actually end up doing.

I'm imagining some of Walmart's clientele ordering stuff via drones in order to have something moving to shoot at.

Reply on news.ycombinator.com