Amazon Drones Don’t Go Far

If you are like us, you’ve wondered what all the hoopla about drones making home deliveries is about. Our battery-operated vehicles carry very little payload and still don’t have a very long range. Add sophisticated smarts and a couple of delivery packages and you are going to need a lot more battery. Or maybe not. Amazon’s recent patent filing shows a different way to do it.

In the proposed scheme, a delivery truck drives to a neighborhood and then deploys a bunch of wheeled or walking drones to deliver in the immediate area. Not only does that reduce the range requirement, but there are other advantages, as well.

For one, you’d expect the truck will keep the drones at full charge. In addition, most of the smarts for the drones live in the truck itself. The truck can track the drone position and calculate routes, commanding the relatively dumb drones to do what they are told. There’s even a provision for launching an aerial drone to help monitor the delivery robots.

Of course, you assume the main vehicle could also be autonomous, but too large to, say, drive up to your doorstep and drop off a box. So, when you think about it, it makes sense to have one robot bring a smaller robot or robots to handle the actual delivery. Is it really patentable? Don’t ask us, we aren’t lawyers.

What would the little drones look like? Maybe like this? We wonder if they’d communicate via Sidewalk?


41 thoughts on “Amazon Drones Don’t Go Far

    1. I was thinking about the same thing, except with manufacturing.
      As they improve internet speeds and cut latency, we may see this happen.
      People in countries such as India would control robotic work stations in the US or other first-world countries..
      They would preform tasks that are currently to difficult for AI.
      Kind of a “Tele-operation” that is cheaper than hiring a US or other worker.

  1. I wonder how large percentage of Amazon deliveries are to areas where this would make sense. It would have to be several packages at a time to a small area. Without ability to handle stair cases, door phones and elevators, it’s mostly limited to dense suburbs, where there would be maybe 20 houses within range. And not everyone orders from Amazon every week.

    1. Yeah, considering the potential obstacles in its path like the ones you mention, plus vandalism and theft, this is as stupidly impractical of a business model as “drone deliveries.”

      1. Indeed, though aerial drones annoying and stupid as the idea is are at least in a space humans don’t really get to interact with easily – so deliberate effort would be needed to bugger with it, a ground device its just a whim of the criminally inclined to trash it – so for my money even stupidier. As there are always criminals, and kids too dumb/ naive doing stupid stuff on a dare, and I can’t see how it would ever not be the case.

        Give it a commonly deployed way to call for an elevator – isn’t there a wireless IR? standard for those with disability to summon one to them? If there isn’t there really should be, then at least in any areas that have built with accessibility in mind the drone will be able to find a route.. Still not convinced it is at all a good idea though, why not just mandate a postal drop off point must exist within a set radius so everything is at the end collected and the deliveries only need to run to a selection of areas. I don’t think anybody would be too put out having to walk a few mins to pick up their mail.

    2. In my rural village neighbourhood there are 3-4 amazon deliveries a day, watching the drivers ie them dropping at most houses in the street over a week.
      But all have small sets of stairs and usually gates to deal with.
      Maybe we could start building air-drop receptors.
      Driver I spoke to said each van drops about 120 parcels a day.

    3. this business model would actually be very useful in my neighborhood. We probably have 4 or 5 amazon trucks that drive through daily, each dropping 10 + packages at a time. if they could park at the entrance to my subdivision and have the robot army swarm out, drive to the exist of the subdivision and pick them up, it would be much faster and probably save on gas.
      The heavy restrictions in my HOA would make for simple obstacles to overcome.

    4. Here in the U.K., Amazon are delivering to our road probably every day, and often one of our immediate neighbours is getting a delivery also that I notice. So I’d think this would work most places.
      Whether it’s cheaper to employ software and hardware devs to build these things rather than freelance drivers, it’s hard to say. But government legislation against “gig” work like delivery will push Amazon towards automation.

  2. Public relations BS. The enormously variable delivery sites with steps , ramps, gates…. etc make this entirely impractical. I think these things are stunts pulled by large companies to keep stock prices up. Gets in the news . Makes em seem like they are innovating.

    1. Absolutely. I was a fool to believe the flying drone delivery they started blathering about ten years ago. That’s right, ten years ago. They can’t even get the box sizes correct or stamp out counterfeit products. C’mon.

    2. Very geography dependant. The majority – but certainly not all – of homes in the U.K. have flat, gate-free access to the front door. But some roads all the houses have gates or steps. So they can just send humans to those roads and robots to the easy ones.

    1. I would take a hard pass on that option. It seems like a fairly trivial matter for a thief to follow the delivery drone into my garage and have free access to everything I don’t have nailed down.

      I have thought about building a “secured” parcel receptacle near my front door and giving Amazon access to that via the same system they use to enter other people’s homes.

      Generally, I’d rather buy from a brick and mortar store but the trend is not in my favor, especially since Fry’s Electronics shut down.

      1. one solution exists – brick and mortar drop-off centers. Not sure about US, but these are increasingly popular in EU. Either a fully automated rack with multiple compartments or local shops doing this for secondary income.

  3. Like others have said, steps, ramps, gates, etc make it impractical.

    But the wheeled drone like is shown here could carry a package and an aerial drone. The van stops once in the neighborhood, lets slip the drones of war, I mean, the wheeled drones, which go as close as they can to the door, then uses the quadcopter to finish the delivery.

    The quadcopter would only have to have as much energy to lift the package, fly 100 feet, land it, and fly back. The wheeled drone can carry a bigger battery pack.

    Of course, in my neighborhood, the sidewalks are so uneven the drone in the picture wouldn’t be able to go down the sidewalks.

    1. Might as well leave out the wheeled robot entirely and just use the truck as an aircraft carrier. Flight deck, elevators, a way to autonomously replace a battery and load a package for delivery…

  4. Considering that a number of my deliveries from Amazon have dumped the package without climbing steps “Eh, close enough”, not sure how a drone leaving it wherever it wasn’t too challenging is much different.

  5. I prefer having packages dropped at a reasonably convenient place such as a nearby supermarket, over some delivery van that rings the doorbell at some inconvenient day or time.

    It just might happen that the delivery guy is at my door when I’m in the supermarket.

    There are also other things to consider.
    According to youtube it is common in the US that packages get abandoned on the doorstep, and theft of those packages is apparently a thing. This does not happen in the Netherlands. You either have to accept a package personally (and sign for it), or a package may be delivered at your neighbours if you’re not at home and you receive a postcard in your mailbox.

  6. Honestly I think they should pursue increasing Amazon Lockers. Many people especially in cities don’t mind stopping at their neighborhood bodega or their grocery store they’re going to anyways to pick up their amazon packages – especially with porch pirating continuing to be a thing. I would think it would be much easier for neighborhood deliveries to be loaded into lockers, as it would be just one stop per neighborhood (efficient & quick), and since the lockers are known interfaces at fixed locations, delivering to them and loading them would be much easier to automate. With going to each house, I’m picturing a scenario like a wheeled droned spending 20 mins trying to figure out a weird front gate latch, and holding up the whole delivery plan.

    1. What is the concept of “stopping at” somewhere?! Everyone has worked from hone for the last 18 months. If I only picked up Amazon deliveries from a locker, I’d have got maybe 4 deliveries in the last year.

  7. I don’t remember the city, but there was a recent article about wheeled drones being approved to use the bike lanes in that city. Obviously, those who approved this gave no thought to the riders who will get killed when they have to swerve into the traffic lane to avoid a delivery drone that is taking up the bike lane. Or maybe they did think about it, but the bribe was large enough to ease their conscience.

    1. If cyclists react to something in the way by swerving into traffic they’ve killed themselves. There are thousands of other things that could be in front of them and they must be dealing with already, so we can probably conclude that delivery drones aren’t that big of a change.

      1. While I don’t disagree if the drones don’t act at all predictably and have no tells as to where they are going next that will get someone killed – the cyclist can’t react when no warning is given – you can read in the posture, sound, the way a car is driven the intent of its driver, and hopefully it uses its signals too, same thing with any pedestrian etc..

        These are pretty tiny boxes without any of that – being quite short they are going to be harder to spot anyway as the threats to cyclists are not usually unpredictable moving crap at ankle to knee high so the focus rightly wouldn’t be as intense for that area. You come round a corner you are looking at what is ahead in the path not for something small and boxy rejoining from somebodies drive, or even just pootling along normally but hidden for that moment they would have been in the checking the road up ahead zone by something else…

        Dog’s sort of count, but at least they have posture etc as a give away, and should be under the control of their owner, so you will have spotted the dog walker even if you didn’t see the dog yet – same thing with small children, can’t see them for the parked cars etc but by the actions of the adult you know they are there somewhere… Point being there are tall so very visible and easy to read markers that let you know something might be there – If the drones are just going to use cycle lanes and roads they should be taller, quite a bit taller – so they cross that threshold into the area drivers and cyclists really have to put most of their attention into, and should always be visible enough.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.