Hacker News Reader

I'm also "intimately familiar" with multi rotors, and I don't think any of these issues you've identified are valid, with the exception of regulations (which Amazon admits won't happen).

> Battery recharge time - out for 15 mins flight? Recharge for 2 hours.

So? Swap out batteries. The whole drone doesn't have to be grounded for 2 hours.

> Flight time / payload - pick one. (20 mins with no payload is a stretch. Add bigger batteries? Too much weight. Add bigger motors and props for more lift? now you need bigger batteries...

That's wrong. There are many off the shelf platforms which can easily handle 5+ lb payloads for 20+ minute flights. Here's one: http://www.freeflysystems.com/products/cinestarHL.php

> No good in 'weather' of pretty much any kind.

That's wrong, too. Multirotor flight speeds are 30+ knots. Modest wind is not a big deal.

> Unstable with a failed motor / prop. Will fall out of the sky under numerous failure scenarios.

Most robust multirotor platforms have 8 or more motors/props, and can trivially handle the failure of one or two. It's true that there are "numerous failure scenarios" which will result in a UAV falling out of the sky, but the same is true of passenger airliners: this will happen if the wings fall off. To suggest that a UAV can't be made as reliable as a typical manned helicopter is short-sighted.

> The blades are lethal - can be shielded, but will add weight and reduce efficiency.

This is only true in the absence of failsafe mechanisms: e.g. a device that brakes the motor when below a certain altitude. Also, as you mentioned, they can be shielded.

> GPS is unreliable at best, and is easily interfered with.

No, GPS is perfectly reliable "at best", and utterly nonfunctional at worst. Fortunately, there are lots of alternatives for GPS.

> Current regulations in countries that have them typically require line-of-sight operation; away from people and property (which rules out landing outside someone's front door); no over-flight of persons; restrictions around distances from the operator when taking off and landing. (And remember the battery / weight issue? Go over a certain weight [~8KG], and the regulations get even more onerous.)

This is 100% valid. Regulations are a concern.

What are other good alternatives to GPS?

Two guesses: cellular location services and custom radio telemetry systems. Maybe that's one guess.

http://en.wikipedia.org/wiki/GLONASS

Inertial navigation systems[1] have been used for more years than GPS has existed. Even now almost all military drones / rockets / ICBMs use very precise inertial systems.

[1] http://en.wikipedia.org/wiki/Inertial_navigation_system

Regarding the flight time/capacity issue, I'd be quite interested to see how a system like [1] would perform.

[1] http://spectrum.ieee.org/automaton/robotics/aerial-robots/ir...

How far can they fly in 20 minutes?

The multirotor you cited has a no-load hover time of about 15 mins - hover n/e forward flight. It has a loaded flight time of about 5 mins. Like I said - pick one.

I didn't say wind, I said weather. That said, forward flight in even moderate wind draws considerably more current.

The motors are open for cooling and weight efficiency. They aren't waterproof. Try flying in snow, or dusty / sandy environments and see what happens!

If GPS accuracy randomly ranges between perfect and nonfunctional, then you have described unreliable. Please cite all these alternatives.

What do you imagine might happen if one were to 'brake the motors below a certain altitude'?

And I didn't say 'can't be made', I said the technology isn't there yet, nor by 2015 as Amazon suggest. I also said it will be eventually.

I'm the complete opposite - I've never even seen a drone-like machine.

Could you squeeze in any more minutes with solar panelling? I mean if they have to fly in decent weather anyway they may as well try to get a bit more charge in them while flying

I am certain any panel you could possibly add would cost more energy to lift than it might collect.

> 20+ minute flights

That's 10 minutes each way. Assuming they fly at 30 mph, you get a range of 5 miles. That might work for big cities, assuming they can fly for 5 miles in a straight line, which they can't.

> No, GPS is perfectly reliable "at best", and utterly nonfunctional at worst. Fortunately, there are lots of alternatives for GPS.

GPS got hacked a long time ago, and even military drones got stolen using a fake GPS transmission.

There are not that many GPS alternatives. GLONASS. Maybe if you have some local towers, but that's a huge project in itself.

1. New LiPo batteries recharge at up to 10C (ten times their rated hourly capacity), which means a full charge under 10 minutes (including final balancing stage).

2. Octocopters, unless loaded to the hilt, can more or less fly uninterrupted even when missing a propeller.

3. Multirotors fly well even in strong winds (20mph winds and more).

I don't see any technical difficulty preventing a company with the resources of Amazon to introduce this service in a year. The major problems are legislative.

And social. I'm curious what sort of utility such a service could yield, but I'm not interested in filling our airspace with service drones. The low-elevation sky is relatively uncluttered by our works.

I think it can work with minor interruption to everyday life if the drone are confined to the paths of major highways. It will reduce their noise pollution (and be marginally safer in a crash).

The bigger impact would be on smaller retailers, that don't have this service. This service is an absolute killer, making it faster to order a toothpaste from Amazon compared to picking it up at a shop outside your house.

This would be an incredible benefit to human economy. Imagine the millions of retail jobs completely obsoleted by drones. Consider the price difference between wholesale and the corner retail shop -- a cost this could largely eliminate. Consider all the hours wasted in retail-facing interactions.

Why would a small store serving a local population not be able to offer this? There is a lot of open source drone stuff out there, so the technology should be readily available. At least available within a year of Amazon putting it into service.

If my local mom-n-pop store has an app and a drone, I would shop there.

I can think of a few problems:

1) Scale: a small store might not have enough sales to justify buying a single drone and training the staff to use it.

2) Regulation: licensing and certifications to be allowed to use these - particularly in a completely hands-off way - may have high fixed costs that a small store can't afford.

3) Lack of knowledge: most small store managers probably don't have any idea of how to set this up, or that it is even possible.

And in the end, why would you buy from the local mom-n-pop store? Nostalgia? Helping the people you know? I think that'd wear out pretty soon or just not work for many people.

Could you get your pizza or Chinese food delivery via drone?

Yeah, for food I think it could actually work, if you had a startup like GrubHub handling the drones, which would just fetch the food from the chosen restaurant and deliver it.

Amazon's page reads like a Google April 1st post.

Or maybe someday we'll look back at these comments like we do at http://slashdot.org/story/01/10/23/1816257/

How right, so very very right you are.

Quoted from the thread you linked.

"Raise your hand if you have iTunes ...

Raise your hand if you have a FireWire port ...

Raise your hand if you have both ...

Raise your hand if you have $400 to spend on a cute Apple device ...

There is Apple's market. Pretty slim, eh? I don't see many sales in the future of iPod.

~LoudMusic -- No sig for you. YOU GET NO SIG!"

I love this one, could be written today...

Apple is being distroyed by the rumors that are being created. When they announce that they are going to have a new product, everyone thinks it's going to blow their worlds. Rumors start flooding in about even the most outragous products ( I even heard a few "sources" mention teleportion) This is getting plain stupid.

Apple is a normal company. Why does the public constantly expect them do the impossible?

Actually, these comments are not as discredited as one might think in hindsight. iPod sales weren't spectacular until the 4th generation of the iPod classic and until iPod Mini was introduced 3 years later. http://upload.wikimedia.org/wikipedia/commons/3/34/Ipod_sale...

I think it's valuable to point out the problems with the current technology and regulations like scoot did, even though we all realize that in a few years things could change.

One could say the same about the iPhone - http://static4.businessinsider.com/image/50104de669beddc2110...

Or really any 'disruptive' product.

  Battery recharge time - out for 15 mins flight? Recharge for 2 hours.

Short trips to/from a fixed distribution point is the canonical use case for battery swapping in electric vehicles.

It is, but keep in mind that for weight efficiency the LiPo batteries are only protected by a heat-shrink skin, no physical or fire protection at all. They will spontaneously combust quite robustly in a crash, or during charging of they are faulty or have been stressed in flight.

So they have to be charged in a fire safe manner, and (in a commercial setting such as a distribution centre) away from the main facility. Not beyond possibility, and again with changes in battery technology this will improve (LiFo is looking hopeful relative to LiPoon the safety front), but another logistical hurdle, so still worth mentioning.

In a city, it would be awesome to see automated rooftop drone charging systems (solar powered of course).

Well, you know, Bezos is a pretty smart guy. So maybe he knows something about this technology that the rest of us don't.

Maybe he has an upper hand over the current state-of-the-art.

Or maybe he's just orchestrating a little extra PR for Amazon on 60 minutes?

I have no particular love for Bezos - last I heard, Amazon sucks as a workplace - but this seems unlikely. There are better ways to drive PR than reveal one of your company's secret projects to the public.

Or he wants to create public demand for this service to use as leverage for getting the regulations through.

Yeah, right. The privileged information available only to the mega-rich...pfft.

Or advances in the available technology driven by a great R&D team with a significant budget instead of hobbyists?

I don't want to knock hobbyists, because they often do great work. But if Amazon thinks this is the future of package delivery, they could bring an awful lot of resources to bear on the problem.

Bezos has access to massive R&D resources because he created massive value for consumers.

Now, those resources can be leveraged to generate even more value for consumers.

The privileged information available only to the mega-rich...pfft.

Don't dismiss the value of insider information re: government policy.

There are two ways to present this type of criticism: lay it out and hypothesize how they are going to overcome these obstacles, or get up and say they will certainly fail because these obstacles are impossible to overcome. You can guess which one of these makes you look like a short-sighted un-imaginative cynic.

Having to have several sets of batteries seems like on of the simpler problems to deal with.

Why not just put the battery in the payload lid?

>We hope the FAA's rules will be in place as early as sometime in 2015

Well, it is carefully crafted to say the rules will be in place, not the technology. The FAA could say Amazon needs to do X first where X is any one of the challenges you mentioned.

I'm glad they put that in, though, because otherwise the threads are always full of "The FAA will never allow this" comments. The wording gives me a reassuring feeling that Amazon knows how much hard work it will take to accomplish.

> We will be ready at that time.

Except for, you know, that bit (the bit you cut out)

Holy crap, you're right. My eyes skipped completely over that. I wasn't trying to be a jerk or anything. Upvotes for you, shame for me.

Sounds like a lot of fun problems to solve, and I'm sure they have many competent engineers working on it.