by cgb223 on 7/17/23, 10:17 PM with 219 comments
Food delivery companies as well seemed to be testing out robots that would bring your food to you albeit in a more sidewalk bound way
Its been a few years and I have yet to have a single thing drone delivered to my house
What happened to that alternate future? Are companies still working on it? Or did we move on from that idea for some reasons we discovered?
by Analemma_ on 7/17/23, 10:38 PM
1. FAA regulations - delivery drones can't operate within X miles of an airport (technically they can, but it requires a much stricter degree of certification and compliance nobody wants to bother with)
2. Drones need a landing space, so people without yards (like apartment and townhouse dwellers, who make up a lot of the population in exactly the densely populated areas where you'd want to use drones to begin with) can't be served
And it turns out that once you exclude "houses within X miles of an airport" and "houses without an LZ", there aren't enough customers left to make delivery drones worth it.
by buzzologist on 7/17/23, 11:07 PM
It wouldn't surprise me if the kinks are being ironed out here before opening it up in bigger markets.
There were a fair few complaints about the noise for the logan ones when they first came out but apparently they are "mostly" sorted now.
- https://www.facebook.com/7NEWSBrisbane/videos/logan-is-becom... - https://doordash.news/australia/doordash-and-wing-announce-p... - https://www.businessnewsaustralia.com/articles/wing-s-drone-...
by doug_life on 7/17/23, 11:34 PM
by mortenjorck on 7/18/23, 1:40 AM
My sense is that drone delivery has the same weather dependency as self-driving cars, only greatly magnified.
by tyoma on 7/17/23, 11:09 PM
At these tasks, they excel and are in so much demand that both sides of the Russian invasion of Ukraine can’t produce enough quantity to meet demand.
by sseagull on 7/17/23, 11:11 PM
There isn't that great a selection of stuff that you can get, and some stores have come and gone from the app.
It's just not that useful. We live in the suburbs, and a weekly trip to the grocery store or big box store (5-10 min drives) can get so much more and at a better price than buying things individually. It's kind of like milk delivery - it made sense before refrigeration but after that, you would buy weekly and just store it.
And right now delivery is free, but it would be even worse if/when it isn't.
One area where it could work is coffee/tea delivery (Starbucks, etc). These are things you buy individually and typically everyday. But AFAIK we don't have that and its not on the horizon.
by dheera on 7/17/23, 11:29 PM
Former delivery robot startup cofounder (Robby Technologies) here. To be honest hardware was by far the most time-consuming thing. I wanted to spend 80%+ of our engineering time and funds on software but it turned out that we ended up spending 80%+ of the time dealing with electromechanical issues, supply chain issues, bad USB cables, motor controller issues, shitty crimping jobs, thermal management, plastic breakage, bad PCBs, bad BMS, bad lithium cells, snapping drive belts, malfunctioning locks, malfunctioning cameras, manufacturer-mislabeled motor wires, sensors that didn't meet specs, antenna placement and RF interference, and lots more. Then there was the operational overhead of figuring out how to charge, move, and store all of them, and how to get things in/out of them when businesses weren't willing to walk from the store to the sidewalk to drop something in a robot, and the customers weren't always willing to come outside to grab their stuff. Then there were robots that got stuck in potholes and the like, and had to be rescued by driving out to them. All that while trying to scale up manufacturing, which never really happened beyond a certain scale. Guess how much time we had left for writing autonomous software.
The thing is, autonomous driving, especially on the sidewalk is actually much easier than the hardware problem of figuring out how to design, build, and scale a new type of vehicle from scratch. The main issue is we, and likely all the other companies, were stuck in a hellhole of hardware problems that something was always "on fire" hardware-wise.
In retrospect I see the companies that went to the roads instead of the sidewalks had it slightly easier on the hardware side: they could just buy a reliable car and mod it, and get to work on software. Safety-wise, of course, they have it much harder, it's a trade-off.
by poopsmithe on 7/18/23, 1:42 AM
Drones have to be specifically designed and manufactured to pass the FAA's certification program. We've all seen capable drones from Amazon's marketing, but does it please the FAA? A drone can't do the job until the FAA signs off on that make/model. It's not easy to do and requires developing a craft with all sorts of safety features.
Companies are making progress, and I think the most insightful videos to watch on the topic come from Zipline. They're already operating in Rwanda with support from their government. They have a streamlined service and make 100s of deliveries per day.
by bryanlarsen on 7/17/23, 10:59 PM
by ilamont on 7/17/23, 10:51 PM
"It can't be before 2015, because that's the earliest that we can get the rules from the FAA. But it could be 4 or 5 years."
- 10 mile radius from Amazon FC
- half hour delivery
- Objects up to 5 lbs, which is 86% of items Amazon delivered at that time
Said the biggest problems were redundancy, reliability, and safety
https://www.cbsnews.com/video/amazon-ceo-unveils-drone-deliv...
by kylixz on 7/17/23, 10:30 PM
https://www.manna.aero/ https://www.flyzipline.com/ https://wing.com/ https://corporate.walmart.com/newsroom/2022/05/24/were-bring...
Several more out there still from the initial bust.
Regulations make it challenging for beyond visual line of sight (BVLOS) autonomous delivery, particularly in the US with the FAA. Also, it's a hard problem in autonomy to nail it every time despite advancements. Less than 1% failure is still potentially catastrophic when you're carrying a few pounds of lithium batteries above people's backyards.
by carabiner on 7/17/23, 11:21 PM
Basically Amazon is massively hampered by FAA regulations that say they can't fly drones beyond operator line of sight. In other words, the remote pilot must be able to see the drone at all times. It's also that drone delivery was never meant to replace all deliveries. It's only specific use cases where it is economical, such as delivery to a lone, rural house, where sending a car would be time consuming and expensive. You can get waivers for the line of sight rule but it requires sophisticated auto-avoidance tech that they're still working on (that must handle automated avoidance in rain or shine, night and day, clouds or not, winds and so on). So it's massively hard problem that was never meant to be a whole "wave" of next gen delivery.
by pgorczak on 7/17/23, 11:04 PM
by avar on 7/17/23, 10:40 PM
Since then e.g. Amazon is stíll nominally pursuing it, but here's a recent article about how that's going: https://www.businessinsider.com/faa-restrictions-are-curtail...
by walleeee on 7/18/23, 12:35 PM
Sending life-saving medicine to a remote township with a drone makes sense. Drone-drop pizzas do not, given planetary circumstances. For that matter, neither does the whole institution of single-meal delivery.
by randomluck040 on 7/17/23, 10:31 PM
by Greenpants on 7/17/23, 10:42 PM
Perhaps more importantly, there's too much that could go wrong. What about legislation where they may fly and how high? What if a drone crashes into someone or something? What if someone's package gets stolen?
As much as the technology enthusiasts in us enjoy the concept of delivery drones, most of us humans still prefer a fellow human in the process of delivering packages to handle edge cases where things might go wrong.
by hombre_fatal on 7/18/23, 3:29 PM
For example, how many people would have to hear the drone zipping by to deliver someone some hair scrunchies or paperclips or whatever? The asymmetry makes no sense to me long before you get into the other issues.
by bradgranath on 7/17/23, 11:02 PM
The point of the thing is to have an impossibile to achieve project that is nonetheless popular enough to generate endless rounds of new investment.
by sethammons on 7/18/23, 1:53 PM
Not groceries, but blood delivery in Africa is a thing
by er4hn on 7/17/23, 11:39 PM
His teaser for the article also had a more detailed story about how he kept on getting human drivers delivering him milk teas when he wanted a drone, but that might just be an aside about the cost of human labor in China and how the system schedules the delivery method.
by mattkevan on 7/17/23, 11:18 PM
They definitely felt like the future when they first arrived, zipping along the pavement or waiting patiently to cross a road. My daughter loves counting them as we drive past.
by Thoeu388 on 7/17/23, 10:58 PM
And drone delivery is quite successful at Ukraine...
by xnx on 7/17/23, 11:05 PM
by gwbas1c on 7/18/23, 2:45 AM
One thing that I think would be interesting is "hybrid" drone delivery: A delivery truck could drive to a central location, and then a "swarm" of drones could drop off packages at homes within a small radius.
by 8note on 7/17/23, 11:08 PM
by PhaedrusV on 7/17/23, 11:39 PM
Currently there's approvals in limited areas in the US for testing, and several companies are approved for significant steps towards our shared dream of 5 minute burrito deliveries to our back patios. Nobody has gotten approved for blanket deliveries yet; the safety levels aren't quite there.
Plug: End State Solutions consults and supports companies in developing the conops, safety case, and approval packages. Reach out once your drone company has a design you're ready to freeze for the approval process and we'll help you out. Our team got Insitu and Matternet the first ever commercial UAS type certificates issued by the FAA.
by nmstoker on 7/17/23, 10:51 PM
- despite the sky being generally less constrained than ground delivery there are challenging obstructions, which are potentially far more risky than for ground travel: nudge past many ground obstructions you'd be okay, do the same in the sky, your drone is toast
- liability is far greater in the sky if your drone carrying something comes down in an uncontrolled manner (on someone or something)
- obstructions are likely to be high at the very points people are willing to come meet the drone (antennas, overhead power lines, washing lines, nets, etc) and they're often hard to spot
- unattended drop off is harder for drones in the places where customers are most dense (ie cities) upping the complexity further
- potential regulatory issues (but I understand below certain heights it's generally not regulated in many countries)
- bad PR from noisy drones!
- risks from non customers interfering
- challenges with carrying what you're dropping off: if it's heavy you need bigger drones; if it's light you'd be tying up a drone with something small, unless you can figure how to drop off multiple items, or you have a mixture of drone sizes
They all seem like they could plausibly be solved but I'm no drone engineer!
Maybe it could get going with targeting particular items that are strongly appealing to customers and might narrow the complexities due to being more uniform than a random Amazon basket. I believe an early use in Bhutan was for medical deliveries. Maybe something premium like ice-cream or cocktails might appeal with the right marketing?!
by bhealy on 7/18/23, 1:03 PM
Doing about 300 flights a day now in DUblin, and expanding soon to 1,000 a day.
Video of a weekends worth of deliveries here-: https://youtu.be/0lFT_K47Pa4
by nonameiguess on 7/17/23, 10:54 PM
by thedougd on 7/18/23, 3:27 AM
1) They’re expanding the delivery territory.
2) They’re still using pilots with line of sight to make the delivery.
I think that’s odd. Maybe someone is moving the goalposts, or maybe there’s not a clear roadmap to broad approval.
by destructuredObj on 7/18/23, 3:08 AM
Aerial drone delivery however ends up requiring additional infrastructure and is far too expensive for what it's trying to accomplish, particularly because of the small payload requirements. Maybe that will change in the near future, but I still think we're a long way away.
by rcxdude on 7/17/23, 11:11 PM
by amorriscode on 7/23/23, 5:14 PM
by orangepurple on 7/18/23, 1:09 AM
1. Command and control - Human in the loop is still a necessity because common delivery environments are too complex for an algorithm to navigate successfully so far.
2. Energy - it takes too long to recharge a drone battery pack. Flying is very energy intensive and there just isn't a way to refuel fast enough from the hub for this to make sense yet. The tested, working alternative is to use a two stroke engine but the pollution per mile from those is astronomical.
by Hackbraten on 7/18/23, 10:15 AM
by babelchips on 7/18/23, 10:46 AM
Reliable ground based sensors are an important area being addressed.
https://twitter.com/thehbarbull/status/1659167041062944770/m...
by MengerSponge on 7/18/23, 5:48 AM
by marricks on 7/18/23, 4:51 AM
Now that labor is even less readily available we're talking about it... less. And that's even off that back of low interest rates and easier money for such initiatives.
by ggm on 7/18/23, 4:18 AM
checking wildlife with lower impact, finding wildlife and feral animals and lost hikers -check
taking aerial photos for weddings and documentaries -check
routine, bladerunner-esque floods of drones sending me parcels.. un-checked.
by jurassicfoxy on 7/17/23, 10:59 PM
by ryanmercer on 7/18/23, 11:30 AM
by xw4002 on 7/18/23, 4:54 AM
by CSMastermind on 7/17/23, 11:53 PM
Here's the tl;dr:
The "last foot" problem is the biggest killer. Getting a drone into the air and to its target is not the hard problem - getting a package safely to the ground at someone's home is.
It would require either a very specific neighborhood or a big advance in computer vision and AI tech.
Where we have seen success in this space are places with dedicated delivery zones in controlled environments where existing transport infrastructure is not a good alternative.
Also this quote: "most people are simply willing to wait two or three days to receive their package while ALL customers wants their packages delivered as cheaply and simply as possible."
Uber eats and door dash largely solve the problem that drone delivery was supposed to solve and if you're going to have autonomous delivery doing it via ground robots is much easier than trying to do it from a drone.
by bbstats on 7/18/23, 1:56 AM
by amai on 7/18/23, 9:05 PM
by netsharc on 7/17/23, 11:01 PM
by press-any-key on 7/18/23, 2:30 AM
Only takes one not following rules.
by dudeomfgstfux on 7/18/23, 6:44 AM
by animal531 on 7/18/23, 10:21 AM
by monkaiju on 7/18/23, 3:05 PM
by jillesvangurp on 7/18/23, 2:31 AM
There are a few things where you could expect some progress to happen over time:
- FAA regulations are evolving and called out as an obstacle. But that's just the US of course. Keep an eye on countries like China which is in any case where a lot of the components are being developed. They are not waiting for FAA approval over there. And that's also the reason you can expect the FAA to be adapting over time.
- Cost is a big factor. The war time use of drones is a case where the use case justifies a higher cost as it means exposing less humans to enemy fire. Losses are high and yet it seems a highly successful niche use for drones even with the current state of the art in technology. Drones are being used successfully in places like the Ukraine, Yemen, and in other conflict zones. As cost comes down, that also opens up more civilian use cases.
- Battery tech is improving. CATL recently launched a 500wh/kg battery product intended for drone companies. Mass production of these is probably going to be a few years down the line. But the technology is shipping this year and not just some proof of concept kind of thing. For most current drones, that would be a doubling of battery capacity; which is a big deal of course. Bigger batteries are under development by a range of companies. Higher capacity, faster charging, lower cost, etc. batteries are coming to market.
- Several cities now have autonomous taxis. Both inside and outside the US. China especially seems very bullish and aggressive on this front. Mostly that's aimed at human transport but the extension to goods delivery seems like it's a logical next step. This stuff seems to be ramping up in more and more cities over the next decade or so.
- A lot of factories already deploy autonomous vehicles on factory floors. Particularly automotive companies have been investing in this. Mostly that's about delivering parts in the workplace.
- Companies like Tesla, Boston Dynamics, etc. have been developing autonomous robots that can move around and operate in more chaotic places. It seems like these could ultimately also get involved in delivery use cases.
So, the future is coming. It might not be in the form some people are expecting. Or happening at the (unrealistic) pace they seem to be expecting it. Hardware just isn't like software. It takes time to develop and it takes more time to ramp up manufacturing. And you don't do that before you have a market. The road from a handful to hundreds to thousands to hundreds of thousands units is just very long and not a straight line.
There is a lot of stuff happening right now. And we're past the point where you can dismiss a lot of this stuff as impossible because there are countless working proofs of concept and real world products challenging that notion already.
by press-any-key on 7/18/23, 2:40 AM
It only takes one not following the rules.
by paulcole on 7/18/23, 12:06 PM
by badpun on 7/19/23, 7:37 AM
by BWStearns on 7/18/23, 11:36 PM
There were also a lot of efforts at more rural delivery schemes but mostly they’re working with quadcopter or hybrid designs which requires a ton of mass being dedicated to power. This restricts your flight time to about 30min and payload to a couple pounds (with some trade off function allowing more or less of one or the other). Even without regulation this means your delivery radius is on the order of a couple miles which means you’re competing against getting in the car or waiting a day.
I think the drone delivery concept is (without magic batteries) DOA if you’re talking about electric powered drones to areas that already have good 0-2 day options to get the thing you’re talking about delivering.
Using gas powered autonomous drones to deliver large payloads to remote areas with decent landing spots seems viable. I’m working on an air to ground glider drone prototype that can maybe be useful for “too far/logistically difficult for fast ground delivery but doesn’t have a prepared strip” but that requires being dropped from an aircraft so it needs to be worth the fractional cost of the flight etc.
Tl;dr: The power required to fly things, battery power density, and the geographical distribution of demand for delivery don’t seem to love each other.
by HumblyTossed on 7/18/23, 1:50 AM
by kgwxd on 7/18/23, 12:57 AM
by romusha on 7/18/23, 1:25 AM
by hot_gril on 7/18/23, 6:19 AM
by solumunus on 7/18/23, 7:15 AM
by trenchgun on 7/19/23, 12:22 PM
by nigamanth on 7/18/23, 12:10 PM
by naikrovek on 7/18/23, 4:07 AM
by hmmmcurious1 on 7/18/23, 8:33 AM
by cpach on 7/19/23, 10:44 AM
by alexfromapex on 7/18/23, 3:42 AM
by Jasper_ on 7/17/23, 10:32 PM
From https://www.sfchronicle.com/business/article/Kiwibots-win-fa...
> The Kiwibots do not figure out their own routes. Instead, people in Colombia, the home country of Chavez and his two co-founders, plot “waypoints” for the bots to follow, sending them instructions every five to 10 seconds on where to go
> On the ground in Berkeley, people also do a lot of robot support. Traveling at 1 to 1½ mph, the bots would take too long to chug to local restaurants, so Kiwi workers pick up the food at restaurants and take it via bikes or scooters to meeting spots around campus to insert into an insulated bag in the bots’ storage compartment.
> The average distance a robot covers for a delivery is about 200 meters (656 feet, or one-eighth of a mile) which makes them fall short of a “last-mile” solution.
by breaker1 on 7/18/23, 1:29 AM