Story in todays Daily Telegraph - Amazon in talks to start drone deliveries in the UK - Telegraph (http://www.telegraph.co.uk/finance/newsbysector/mediatechnologyandtelecoms/digital-media/11496237/Amazon-in-talks-to-start-drone-deliveries-in-the-UK.html)

I've been keeping an open (if sceptical) mind about all the Amazon drone stories, but even if only half true, that story, and the quotes by the UK transport minister, give them some new credibility.

Paul Misener, Amazon's vice president for global public policy, said in evidence to US senators that while the FAA was considering its requests to test, “we innovated so rapidly that the [drone] approved last week by the FAA has become obsolete. We don’t test it anymore. We’ve moved on to more advanced designs that we already are testing abroad."

Mr Goodwill said the UK is very interested in how the country can be at the forefront of drone technology and development.

"We're working with Amazon," he said. "And Government is working on the whole issue of drones. We're meeting with the British Airline Pilots Association and we're both keen to innovate."

It's not even April the First and I really can't see anything in these stories apart from generating publicity for Amazon.

As someone who has limited experience of flying a small drone, less than a year, I'm aware of so many things that could, and would, go wrong with a package delivery system such as the one that is proposed.

To allow a commercial retailer to bypass all the rules and regulations regarding safe operation whether that be keeping the craft within Line of Site or Extended Line of Site, not overflying busy roads or railway lines, checking that the landing zone is clear of people is simply not a realistic proposition.

All of the restrictions that are applied to commercial operators flying UAV's would need to be transferred to a service like this. And what back up would there be if something were to go wrong?

Lipo batteries? We all know how safe these are.

It's not even April the First and I really can't see anything in these stories apart from generating publicity for Amazon.
I think it's less what is being said, but who is saying it - Robert Goodwill is the UK minister directly responsible for aviation policy. ( https://www.gov.uk/government/people/robert-goodwill ) What motivation could he conceivably have for making a statement simply to give Amazon publicity?

Isn't it exactly the opposite? If he makes statements about matters that do turn out to have no substance, it's a big knock to his credibility.

And on the other side the quote is from "Paul Misener, Amazon's vice president for global public policy". The initial comments could well have been interpreted as a speculative wish list, and yes, the main effect to just get publicity. But now - if there isn't any substance behind the stories - it could be counter-productive. Eventually lead to an attitude of "oh god, not that again!" if no real progress can be shown.

As someone who has limited experience of flying a small drone, less than a year, I'm aware of so many things that could, and would, go wrong with a package delivery system such as the one that is proposed.

To allow a commercial retailer to bypass all the rules and regulations regarding safe operation whether that be ..........
I know what you mean...... and I did say I was keeping "an open (if sceptical) mind".

But don't underestimate what technological change can achieve if it's thought the returns justify it. The Wright brothers made their flight in 1903 - what do you think the man in the street would have said then if you were able to go back and describe the air war in 1943, just 40 years later?

Closer to home I remember well the first cellular phones in the late 1980's. It was inconceivable then that within not many years they'd become so ubiquitous, and able to do what they do now. And if I'd been told then that it wouldn't be long before every schoolchild would have one....?

Yes, with drones currently on commercial sale, there are lot's that can go wrong - but what's being talked about are systems way in advance of that. And it's not a question of "bypassing all the rules and regulations" - rather writing a new rule book from scratch for a future technology. Which is why I see a joint public statement from a senior Amazon figure and the UK minister responsible for the country's aviation policy as so significant.

If you do accept that there is any substance at all behind the reports, then the real question has to be "when?" I don't for one second expect it to be next year, and don't expect all deliveries to switch to drones overnight anyway. But I'm beginning to see some initial usages as increasingly likely.

I think it's less what is being said, but who is saying it - Robert Goodwill is the UK minister directly responsible for aviation policy. ( https://www.gov.uk/government/people/robert-goodwill ) What motivation could he conceivably have for making a statement simply to give Amazon publicity?

Isn't it exactly the opposite? If he makes statements about matters that do turn out to have no substance, it's a big knock to his credibility.

..................... If you do accept that there is any substance at all behind the reports, then the real question has to be "when?" I don't for one second expect it to be next year, and don't expect all deliveries to switch to drones overnight anyway. But I'm beginning to see some initial usages as increasingly likely.

With the UK Election just round the corner Robert Goodwills tenure as Parliamentary Under Secretary of State for Transport will not continue for much longer. And in the run up to an election any 'positive' publicity would need to be viewed with the same scepticism as his views on Europe.

But seriously any aerial package delivery service would need to be developed and tested to the same rigorous safety regime as the driverless cars now being trialled in the UK. ( with a live driver as backup )

I've just sent in the final draft of my Operations Manual to achieve the CAA BNUC-S permission to work with my DJI Inspire 1 quadcopter and is all about safety, risk assessment, procedure. Very strict rules govern professional UAV flying in the UK. The most basic of these are about proximity to roads, railways, crowds of people and most importantly piloting the craft with a clear line of site.

Another safety factor is the limitation of operating in favourable weather conditions. High wind speed, rain, snow or thunderstorms don't mix well with UAV operation

I could see the future use of 'robot drones' to deliver essential supplies to off-shore structures such a oil rigs or remote island communities,or in the desert, or the Australian outback but cannot see how this could be achieved safely over highly built up urban areas. I've no doubt that the technology would enable a basic aerial delivery service but whether people either want or would allow such a service is another matter.

I've absolutely no doubt that Amazon is serious about developing this service, or that have the resource to do this:

http://g-ecx.images-amazon.com/images/G/01/rowland/AmazonPetitionforExemption_July92014.pdf

Their aspiration is that "seeing Amazon Prime Air will be as normal as seeing mail trucks on the road today, resulting in enormous benefits for consumers across the nation"

And as pointed out in the article you referred to in your original post the reason Amazon are trying to get backing from the UK government is that they have become frustrated with the FAA's reluctance to allow developmental testing.

So yes it will be interesting to see how this develops.

I've just sent in the final draft of my Operations Manual to achieve the CAA BNUC-S permission to work with my DJI Inspire 1 quadcopter and is all about safety, risk assessment, procedure. Very strict rules govern professional UAV flying in the UK. The most basic of these are about proximity to roads, railways, crowds of people and most importantly piloting the craft with a clear line of site.
Yes, but those rules apply for drones with existing technology. We can only make guesses as to what Amazon is doing - assuming the whole thing is genuine - but it brings up the whole concept of "machine intelligence". It's conceivable that enough "intelligence" may be able to be given to the drone that it will follow a pre-determined course safely and without manual help.

I find the concept easier to instinctively believe than driverless cars - though there's enough talk about the latter coming from serious quarters that I suspect it will come about eventually. If an autonomous drone falls out of the sky, then OK, not good, and a risk of causing harm. But if a driverless car, with 4 people on board gets it wrong, and (say) swerves in front of another vehicle there's an almost inevitability of harm!

There are some things which you can predict will never happen without a complete reunderstanding of basic laws of physics - cars will never get below a certain minimum size, for example - the limiting factor becomes the size of a human being. And the speed of light puts constraints on lots of other things. But autonomous drones are an engineering problem. There is no absolute reason why they can't be engineered, even if they sound fanciful at the moment. (As would talk in 1987 about browsing the internet - which itself barely existed then - on a tiny mobile phone.)
So yes it will be interesting to see how this develops.
Indeed. Personally, I find factors such as exactly where the delivery will be made to (if, for example, you live in a flat) and weather considerations more difficult to see how to overcome than the technical ones.

But that's not to say impossible. At the moment of ordering, it may tell you where you have to be at what time to receive the delivery - a point close to where you live and where the system is confident it can deliver to. Ideally your back garden, but possibly a communal garden for flat dwellers? Or tell you that a delivery won't be possible for x hours because of weather? And nothing happens until you click the button to say "yes".

And individual politicians may come and go, but it's the very fact of a joint announcement between Amazon and the current aviation minister I find so interesting. It has to give the story more credibility.

YThere is no absolute reason why they can't be engineered, even if they sound fanciful at the moment. (As would talk in 1987 about browsing the internet - which itself barely existed then - on a tiny mobile phone.)

Indeed. Personally, I find factors such as exactly where the delivery will be made to (if, for example, you live in a flat) and weather considerations more difficult to see how to overcome than the technical ones.

But that's not to say impossible. At the moment of ordering, it may tell you where you have to be at what time to receive the delivery - a point close to where you live and where the system is confident it can deliver to. Ideally your back garden, but possibly a communal garden for flat dwellers? Or tell you that a delivery won't be possible for x hours because of weather? And nothing happens until you click the button to say "yes"

David, you hit upon something that I hadn't thought of before. I keep thinking of ways Amazon could deliver to your dwelling, but a more feasible concept would be pre-defined safe routes for the UAS to fly, to designated drop off areas that are close to your dwelling. Those drop zones could be manned by Amazon personnel who would then assure delivery to the proper recipient. I am confident they will play out many of the what if scenarios and have contingency plans in place. For example, if a UAS goes down, a locator beacon, not unlike current aircraft possess, would allow timely retrieval and if proper safe corridor routes are in use, then the likely hood of damage or injury is greatly reduced.

But that's not to say impossible. At the moment of ordering, it may tell you where you have to be at what time to receive the delivery - a point close to where you live and where the system is confident it can deliver to. Ideally your back garden, but possibly a communal garden for flat dwellers? Or tell you that a delivery won't be possible for x hours because of weather? And nothing happens until you click the button to say "yes".

And individual politicians may come and go, but it's the very fact of a joint announcement between Amazon and the current aviation minister I find so interesting. It has to give the story more credibility.

Sure if the commercial imperative exists a service this could well become reality but I'd rather be looking up at birds flying through the air than buzzing Amazon drones.

But why not pizza delivery ?

or specialist deliveries to hard to get locations . . . . .

Drone carrying drugs and weapons crashes into prison in botched smuggling attempt - Mirror Online (http://www.mirror.co.uk/news/uk-news/drone-carrying-drugs-weapons-crashes-5378549)

David, you hit upon something that I hadn't thought of before. I keep thinking of ways Amazon could deliver to your dwelling, but a more feasible concept would be pre-defined safe routes for the UAS to fly, to designated drop off areas that are close to your dwelling. Those drop zones could be manned by Amazon personnel who would then assure delivery to the proper recipient.
I don't know about the States, but in the UK the idea of "Click & Collect" is becoming increasingly popular. (I think for Amazon as well as other companies.) Order something online, and pick it up at a local location, no having to wait at home to accept the delivery.

What I suggested above could just become an extension of that, if you don't have your own location such as a garden for drop off.

And they wouldn't even need to be dedicated Amazon personnel at the drop zone. I did a return to Amazon a month or so back and after printing the label I just dropped it off at the local supermarket.

But do you really need people at all? Let's say I suddenly realise I want something like a computer hard drive urgently. I live in a block of flats, so no garden for delivery, but there's a small park 200 yards down the road. I go online, put in the order, and am told to be in the park in 37 minutes time. I arrive, and a minute later a drone appears overhead and I get a text - "Are you ready to collect your package? Reply yes to this text if so. Access code 1357" I reply "yes" and watch the drone land, key "1357" into the drones keypad, and take out my hard drive - less than an hour after ordering - and watch the drone fly away.

Far fetched? Science fiction? I'm still inclined to say yes, and can still think of a thousand and one issues to be overcome to be practical - but possible in the future? I'm increasingly beginning to think "yes".

In the end it will be decided by economics. How much will such a drone delivery cost relative to a conventional delivery, and how popular would such a service be? And what premium will people be prepared to pay over current deliveries for the service?

But with advanced technology - machine intelligence - I can actually foresee future legislation PREFERRING automated and pre-programmed, non-line of sight flights, over simple manual flying. And delivery drones don't have any of the privacy issues surrounding current ones whose main job at the moment is for cameras.

In Amazon's letter to the FAA requesting permission to test these delivery drones they talk about Geofencing

'Geo‐fencing is a feature in a software program that uses GPS or radio frequency triangulation to define geographical boundaries. A geo‐fence is a virtual barrier – effectively an electronic box in which the sUAS will be confined'.

Now in this instance they are referring to the area they want to use to test their machines but presumably this will also become the means by which they set out their route. This is similar to setting up waypoints on a ground station for the purpose of aerial surveying or 3D mapping.

I'm just wondering how existing, legitimate, certified commercial UAV operators will interact with this proposed Prime Delivery service. One of the first pre-flight checks we are taught to carry out before we plan a professional assignment is to check whether any NOTAMS exist for the are we intend to fly in. Will Amazon be posting NOTAMs ( a notice filed with an aviation authority to alert aircraft pilots of potential hazards along a flight route or at a location that could affect the safety of the flight ) or maybe they would have a website showing their geofenced routes ?

Who would have the right of way up there?

But for me the biggest stumbling block to this idea is the lack of local distribution centres or Fulfilment centres as Amazon calls them. Just as an example my nearest one is in Slough, that's 44 miles away, which would equate to a round trip of well over a hundred miles by the time the Amazon Drone had bypassed Heathrow and Gatwick airport.

I'm sorry but you can probably tell I'm not a big fan of Amazon. They seem to want to chew the bone from both ends at once, pay minimum tax in any jurisdiction whilst at the same time employing people via agency to undertake robot style jobs on minimum hours contracts. So whilst I've a huge interest in the development of drone technology the thought of Amazon making a deal so they can use the UK as a testing ground for their aerial delivery system does not fill me with joy.

Well the story has leapt forward a bit . . . . . . . from my perspective I just hope that the correct decisions will be made and that people realise it's ok to say no to Amazon's plans to fill the sky with their flying shopping trolleys.

Amazon tests delivery drones at secret Canada site after US frustration | Technology | The Guardian (http://www.theguardian.com/technology/2015/mar/30/amazon-tests-drones-secret-site-canada-us-faa)

I'm just wondering how existing, legitimate, certified commercial UAV operators will interact with this proposed Prime Delivery service. One of the first pre-flight checks we are taught to carry out before we plan a professional assignment is to check whether any NOTAMS exist for the are we intend to fly in. Will Amazon be posting NOTAMs ( a notice filed with an aviation authority to alert aircraft pilots of potential hazards along a flight route or at a location that could affect the safety of the flight ) or maybe they would have a website showing their geofenced routes ?

Who would have the right of way up there?
I would suspect it would be similar to existing rules in VFR airspace, or for that matter as exists for boats at sea. So one of the challenges for Amazon is to develop anti-collision technology. But if they do - and let's say the rules are similar to those at sea - then existing drones must follow the same rules from manual input. So if your drone and an Amazon one are on a collision course, and the rules say you both have to turn right (say), then you're obliged to do so.

I appreciate that may be easier for an autonomous system (using on board cameras and it's own "intelligence) than one being flown from the ground - the more so the further away it is. So I'd further predict it won't be too long before it becomes a requirement for ALL drones being used for commercial purposes to have such "intelligence". It may well be that any drone without such a system can only be flown below 200ft - commercial flying being between 2-500ft with mandatory autonomy. It may also become a requirement for all such drones to "broadcast" locally their GPS co-ordinates to facilitate that, even "talk" to other drones in the vicinity.

Move to autonomous systems (actually TAKE AWAY the line of sight vision requirement - providing they have the appropriate technology) and whole new profitable fields open up. Power line observation? Search and Rescue? Existing drones are already showing limited worth now, autonomy would make them come of age.

This may not sit well with current drone users who currently enjoy near empty skies. (At least up to 500ft.)I'm sorry, but I don't think that can last. Any more than the earliest motorists could keep the roads to themselves for ever. But just as the rising profile of motor cars brought benefits (no need for a man with red flag! :-) ) as well as restrictions, it may be the same here.

And whilst it may be Amazon that the current news stories are about, I don't see them being alone if a workable technology comes about. One reason for developing such technology may be to be the main supplier to third parties - not simply for own use. As the Guardian article says: "Drone technology is seen by many tech companies and aeronautics experts as the next frontier for innovation, with billions of dollars potentially in the balance." Those billions of dollars won't simply come from delivering Amazon parcels.


But for me the biggest stumbling block to this idea is the lack of local distribution centres or Fulfilment centres as Amazon calls them. Just as an example my nearest one is in Slough, ...........
I can't argue with that - but in the end it comes down to economics. As I said before:
In the end it will be decided by economics. How much will such a drone delivery cost relative to a conventional delivery, and how popular would such a service be? And what premium will people be prepared to pay over current deliveries for the service?
Technically, distance from depot may be one of the easiest problems to solve - just build more depots! But is that economic? Realistically, there may be ways round it, maybe having satellite depots with an in stock range of the most popular items, and the ability to run fast shuttles for the wider range. I don't know the answer, but I'm sure they must have thought about it.

To add to David's excellent suppositions, I can see the adoption in the future of current aviation technology that full sized aircraft use. In fact, in the FAA's NPRM, they discussed whether current technology would allow for a transponder aboard sUAS. They concluded that it's not feasible at present. But that one piece of gear would allow for TCAS to be implemented. An sUAS could easily maneuver away from a full aircraft or another UAS.

For non pilots, TCAS is Traffic Collision and Avoidance System and relies on the transponder that all aircraft are required to have if they operate in controlled airspace. It's the device that makes your plane's tail number/flight number and altitude to be displayed on a controller's secondary radar screen. But in the last few decades, those same transponders have allowed the TCAS system to warn pilots of an impending collision. Even if the UAS couldn't transmit, but only receive a much stronger beacon from a full sized aircraft, they could 'get out of the way' and it wouldn't really be necessary for the flight crew to know the UAS was even there. The receiver could be made quite small, it's transmitting a signal that requires much more power and thus, larger, heavier circuitry.

-gb-

I can see the adoption in the future of current aviation technology that full sized aircraft use. In fact, in the FAA's NPRM, they discussed whether current technology would allow for a transponder aboard sUAS.
I think in the future there will be drones and drones - not one size fits all. What you say may certainly be required for the larger ones - which may be allowed to fly higher than 500ft and quite fast - but I think most people here are interested in smaller versions, more along the type Amazon seem to be proposing.

And for something restricted to 50mph (say) and lower than 500ft, a standard transponder may be overkill. What about something advertising it's presence via broadcasting it's GPS co-ordinates, track and speed periodically? At 50mph max, it wouldn't need anything like the range (hence power) of that required for an airliner at 500mph. Another point may be that it is required to broadcast it's ID, something I can see as inevitable after the recent drone incidents in Paris.

That may seem onerous to current users, but it needn't be seen that way. At the moment the authorities are simply nervous of drones - arguably with good reason - sensible legislation, via the right technology, could alleviate a lot of that fear and be just what's needed to make general drone usage generally acceptable.

I think in the future there will be drones and drones - not one size fits all. What you say may certainly be required for the larger ones - which may be allowed to fly higher than 500ft and quite fast - but I think most people here are interested in smaller versions, more along the type Amazon seem to be proposing.

And for something restricted to 50mph (say) and lower than 500ft, a standard transponder may be overkill. What about something advertising it's presence via broadcasting it's GPS co-ordinates, track and speed periodically? At 50mph max, it wouldn't need anything like the range (hence power) of that required for an airliner at 500mph. Another point may be that it is required to broadcast it's ID, something I can see as inevitable after the recent drone incidents in Paris.

That may seem onerous to current users, but it needn't be seen that way. At the moment the authorities are simply nervous of drones - arguably with good reason - sensible legislation, via the right technology, could alleviate a lot of that fear and be just what's needed to make general drone usage generally acceptable.

I tend to agree it might be overkill for the sUAS (under 55lbs.) being considered in the NPRM. Broadcasting your GPS coordinates is great, but that means all other aircraft need to have something to receive and process that info. Lower than 500 feet is certainly less congested airspace, but helicopter pilots are still affected by some of their ops (ie air ambulance). They already have transponders, so to me, if it were feasible, it would be quicker and easier to add transponders to sUAS that work with the existing system. Being small, that's why I mentioned perhaps receive only for the UAS that can then take the necessary evasive action.

It's good in a way that companies with deep pockets like Amazon are interested and are doing research. Their discoveries will hopefully trickle down to smaller craft as technology marches forward. Heck, I remember back in the early 90's when smaller aircraft were transitioning from 2d Loran C navigation to GPS. Initially, the GPS system wasn't reliable enough to be used for precision instrument approaches. Then GPS approaches were developed, but the FAA mandated a GPS receiver had to have RAIM, (receiver autonomous integrity monitoring) to be used for precision approaches. Now, we have GPS guided small UAS craft like it was no big deal. So I have hope that the technical hurdles of all of this will be worked out as we go forward.

Thanks for a great discussion, David.

-gb-

The message of the story is the U.S. is over-regulating and killing prospective business. Maybe Amazon deliveries will never come to pass on a large scale, but who knows what innovation would come to pass. Surely, when it does, the patent won't be in the U.S.

You can say thanks to the FAA for that.

For non pilots, TCAS is Traffic Collision and Avoidance System and relies on the transponder that all aircraft are required to have if they operate in controlled airspace. It's the device that makes your plane's tail number/flight number and altitude to be displayed on a controller's secondary radar screen. But in the last few decades, those same transponders have allowed the TCAS system to warn pilots of an impending collision. Even if the UAS couldn't transmit, but only receive a much stronger beacon from a full sized aircraft, they could 'get out of the way' and it wouldn't really be necessary for the flight crew to know the UAS was even there. The receiver could be made quite small, it's transmitting a signal that requires much more power and thus, larger, heavier circuitry.

-gb-

That's not exactly correct. Transponders don't broadcast the tail number. Also most full size planes don't have TCAS its only required on aircraft above 33,000 pounds and/or more than 30 passenger seats. So even if drones had TCAS that wouldn't be particularly helpful. Also, requiring TCAS on drones would probably set the technology back a decade. Could you imagine how many false alarms resulting from erroneous operation there would be?

Most of the litigation over the past year had to do with defining the FAA's authority and although its authority goes from an inch or so above the ground up, the floor of the NAS is mostly around 500' So if sUAS stays below 500' and airplanes above, there is very little reason for "see and avoid" technology. Also be aware that there's a big difference between "see and avoid" and "collision avoidance."

The FAA could establish victor airways between cities, at specific altitudes where delivery drones can fly for higher quantity payload delivery. For local delivery the sUAS rules would apply. Whatever they decide, with the recent NPRM the FAA took a step in the right direction and have established a framework that should allow for development to start.

That's not exactly correct. Transponders don't broadcast the tail number. Also most full size planes don't have TCAS its only required on aircraft above 33,000 pounds and/or more than 30 passenger seats. So even if drones had TCAS that wouldn't be particularly helpful. Also, requiring TCAS on drones would probably set the technology back a decade. Could you imagine how many false alarms resulting from erroneous operation there would be?

Most of the litigation over the past year had to do with defining the FAA's authority and although its authority goes from an inch or so above the ground up, the floor of the NAS is mostly around 500' So if sUAS stays below 500' and airplanes above, there is very little reason for "see and avoid" technology. Also be aware that there's a big difference between "see and avoid" and "collision avoidance."

The FAA could establish victor airways between cities, at specific altitudes where delivery drones can fly for higher quantity payload delivery. For local delivery the sUAS rules would apply. Whatever they decide, with the recent NPRM the FAA took a step in the right direction and have established a framework that should allow for development to start.

Correct. I didn't state that very clearly. It's your transponder's assigned squawk code that the controller tags with your aircraft's identity. But it's still the transponder that makes the feat possible on the controller's screen. Maybe that's not a practical solution for UAS operating below 500 agl. I'm just imagining ways to alleviate the fears expressed by pilots of manned aircraft. See this example.

RAW VIDEO: Drone near Chopper 7, March 16, 2015 | www.kirotv.com (http://www.kirotv.com/videos/news/raw-video-drone-near-chopper-7-march-16-2015/vDLGPq/#cmComments)

And BTW, the operator of the UAS was visited by the FAA and cleared. He had video proof of his own that he was nowhere near the news choppers and he yielded the airspace to them when they arrived on scene.

When I flew model airplanes in the early 1990's, there was no perceived threat to privacy or other aviation because there were no onboard cameras or telemetry. I had to keep the model in sight and close enough to not lose control signals. Now, with just several hundred dollars, anyone can fly a model so sophisticated that they can take it way up into the NAS and have a first person view with telemetry overlays like in the example above. I only say this because it still boggles my mind how far we have advanced the art in just a few decades.