Drone swarms. What's your thoughts on the tech and applications good and bad?

[large_ghostman]

I got an uncle who lives in Suffolk UK, i think its gone now but the USA had a air base there near mildenhall, they had the blackbird spy plane. My uncle was a roofer who worked at the base, in the grounds was a lorry container! Inside apparently a desk and it was classed as some south American country!!! A bit like an embassy is foreign soil.

Then you have the fact that recently the States murdered some UK citizens because our Gov said sure if you see them drone them and so they did. Then you also have two helicopters flying into a country without permission or notification and assassinating a guy.

I am sure the UK does similar things, but the point is your constitution is like everything else, it only applies when they say it applies. Look at a certain Island, i find it ironic that any country that sneers at others for poor human rights abuses has kept people locked up with no trial and no way to protest, sure its not on your soil so the constitution dosnt apply, sure they are bad people (maybe but without a trial and evidence how do you know?).

The point is dont think governments stay inside the law, they can bend and twist it with the best. And before anyone says well it was Bin Laden so it dosnt matter, the fact remains a country was invaded without knowledge or permission and a man was executed without trial. Not a single person blinked an eyelid, The UK has proven to be complicit with torture using methods we openly condemned before, during and after the events, yet we still did it.

But the post was about the tech, the assumption a swarm needs a central control is wrong. And as you point out if your bad enough your dead whatever you do.

Here a few years back a MI5 or MI6 guy was found dead, in a bathroom in a suitcase that was padlocked from the outside and he was tied and handcuffed, VERDICT it was an accident!!

https://en.wikipedia.org/wiki/Death_of_Gareth_Williams (found some info on it but wiki differs from from other stuff i seen on it).

 

[large_ghostman]

Ok i am digging out my old files. I will stick on a web page.

Mike question for you, how long did you get from the engine on petrol and what was/is the weight on the engine? I want to use this as a bench mark, cant find much that I wrote down on the fuel types but i dont think i need it. I have a almost new 10cc 2 stroke, i am going to see if i can get it going on petrol without a spark plug or messing with spark or timing systems. Its one the bigger engines I have tried this on, not all glow engines i have tried this on work.

Once i get some half useful information i will post the link up to the main page.

 

[tcmtech]

Some info worth a bit of ponderance regarding running glow engines on diesel fuel.

http://www.flyrc.com/make-your-own-diesel-conversion-head/

**broken link removed**

One head scratcher on this would be about using E85 Ethanol/gasoline mix as a way to reduce the viscosity and possible way to lower the compression ignition temperature issue.

What has me pondering on the idea of the E85 addition to improve the compression ignition issue is that I have used it for years to get old diesel to run on used oil with minimal fuel system changes. It works well at a ~ 10- 25% mix ratio (depends on the average weight of the oil base stock) to thin out used engine oil while also improving its compression ignition characteristics to the point that the old diesel would start and run on it like its normal diesel fuel so I am thinking it would also have a similar effect on the same characteristics for a tiny glow engines operation.

Just a ponderance being I don't have any such engine to experiment with at this point to see what sort of effect it may have.

 

[large_ghostman]

nah thats me not explaining, the Glow you run unmodified on petrol, i never got diesel run mate. the diesel comment was more about the fact those engines have double the fuel efficiency but cost alot, are not too easy to get and are normally too small.

unmodified glow engines on petrol work ok, but i gave up. no one in the model world i was a part was interested. Until you tube banned my old channel i had a video of it working with a .65 and a .75 glow engine, they still called voodoo and fidget spinner. I didnt have a use for it in the end so gave up, now move forward a few years and drones do have an application. So its something i might visit again, the fire service do use them now, but mine would have a few features useful for them.

It was Mikes comment about a swarm to look for people that kind of made me think about it again, the wind turbine application i wanted it for is redundant. The three Turbines got decommissioned because of the land being sold and a host of things no one ever explained to me. But in my spare time i would like to at least conclude this project, while i got it working except for having the engine actually in a drone powering it, i just feel with some modern tech i could take it further than i did before.

Just feels unfinished to me, although for its purpose it was finished, actually i finished it properly in our wood when i switched off the guidance by mistake , still find the odd bit now and then.

 

[nsaspook]

if you have military drones snooping around, you've got problems that won't be solved by jamming the drones.

Right, then it's time to go full-metal jacket on those drones.

Some of the toys you can own in Oregon.
https://machineguncentral.com/Default.aspx

 

[large_ghostman]

You tube might have deleted the channel now, i couldnt find the video i was after. But in some places there its now a kind of sport. The one i remember had several jeeps with HUGE machines welded on the back, lined up with groups of guys with all kinds of guns. The idea is people fly the drones back and forth along this mud bank and the people with the guns try and hit the drone!!

I think its 3 times they have to go in the firing line, really really funny to watch, you can hardly see these little drones, but you do see the earth erupt back and forth. Then when the runs are complete you get to see whats left if anything. No idea what some the guns are, but to me they look like what you want with you to take down a full size plane!!

Surprisingly hard to hit the little drones, some the videos has 20-30 people all shooting at the same time at a single drone! I have absolutely no idea how anything could get past a wall of fire like that. Its a shame you tube seem to be taking down any channel that isnt making serious ad money.... I mean any channel that is family friendly and have a cat in

Found one, this isnt the really funny one, but does show the kind of fire power they were throwing at it, the one i really wanted to find had loads of these all firing at once at the drone.

These ones are special rounds designed specifically to take down drones. They appear ok as long as the drone isnt moving and is tied to a pole in the ground..... Well most drones are anyway arnt they.

 

[large_ghostman]

Interesting to watch a military drone swarm, its supposed to be legit but honestly i got a feeling this is more fidget spinner than drone.

 

[Pommie]

Jason,
The engine I was looking at was the OS GF40.
It produces 3.75HP (2.8kW) burning 1.2kg of fuel per hour and weighs 1.5kg with silencer and ignition supply.
The electric motor I matched it with is the **broken link removed**.
It has kV = 172 so at 8600 revs it will produce 50V and weighs 1.3kg

So hopefully you can build a generator that weighs around 4kg and produces a continuous 2kW.

2kW can keep 30kg aloft very easily. A 4kg generator with 14kg of other flying stuff (frame, motors, electronics etc.) would leave 12kg for fuel = 10 hours flight time.

As I said though, the tricky part is the 3 phase to DC converter. It has to be power factor corrected so the load looks purely resistive.

Mike.

 

[large_ghostman]

OS engines have come down a huge amount!!! Thats silly money compared to a while back. Cant argue with those numbers.... 30kg with modern gear is more than your going to need, cant get my head around 3.75HP from something that size. Looking out the window and can just about see our two Shetland ponies in the paddock.

The question i keep asking myself is, what is actually stopping Amazon from making any real progress? Surely it cant be basic physics now stopping them, but if it was a health and safety issue then they are very silent on it. BTW i take it this is real petrol and not the stuff you can almost drink? We only have the real petrol in these parts, i think the 80/20 stuff would likely get drunk around this area

 

[Pommie]

Amazon will not go down the smelly, noisy petrol route as it doesn't fit their image. They want clean battery powered units. Personally, I can't see drone home deliveries anytime soon.

However, mapping fires, finding people, patrolling beach areas for sharks/rips etc and lots of farm uses come to mind.

The fact that the energy density of petrol (46Mj/kg) is almost 100 times more than Li-Ion (0.5Mj/kg) is testimony to the vastly different flight times possible.
(I know engines only 40% efficient but that's still 40 times more power)

Mike.

 

[tcmtech]

cant get my head around 3.75HP from something that size. Looking out the window and can just about see our two Shetland ponies in the paddock.

I run into people who have trouble with power/energy density comparisons quite often. My brother had that issue when I got my tractor backhoe put together.

Sure, my tractor backhoe that can only move ~2 - 2 1/2 cubic feet of dirt in one scoop and only go down ~ 9 feet compared to the big multi cubic yard comercial excavators we rent every now and then for our bigger projects. To which his comment was it would be near useless and not worth the time and money I put into setting it up because a scoop shovel was about the same in his view.

Ah, but can you put 50+ HP and 4000+ pounds of downforce behind hand powered scoop shovel and do it for hours on end? I think not! so, guess who does the second most digging behind myself with it now?

 

[tcmtech]

One of my questions with fuel powered drone concept is why not just use a fueled engine for each rotor rather than do a multi step process to convert the fuel energy into fling motion?

What's the downside to using servo throttle control of and engine each prop VS full electric motor drive fro each?

 

[Pommie]

I'm guessing that fuel engine response isn't quick enough to get the fine level of control a drone needs.

Mike.

 

[large_ghostman]

Amazon will not go down the smelly, noisy petrol route as it doesn't fit their image. They want clean battery powered units. Personally, I can't see drone home deliveries anytime soon.

However, mapping fires, finding people, patrolling beach areas for sharks/rips etc and lots of farm uses come to mind.

The fact that the energy density of petrol (46Mj/kg) is almost 100 times more than Li-Ion (0.5Mj/kg) is testimony to the vastly different flight times possible.
(I know engines only 40% efficient but that's still 40 times more power)

Mike.

While the petrol engine you link too is ideal, i remember the plane for the turbines and the little diesel engine on it. No idea wht size you can get them in, i am aware that second hand it was pretty hard to get hold of even a small one. The one i was told to look out for was called FROG. Diesel engines tune correctly (model ones) are over twice as fuel efficient and have alot of low down torque. It might be an idea to see if I can get hold of a reasonable size one, likely will need one of those maplin self build gear boxes.

Looking at your figure mike, The rev range needed is within the capability of a decent sized Diesel, but i honestly have no idea if they even make them big enough to do it.

Amazon is just something that crops up now and again in conversation with someone i know, when they first brought up the idea they were talking 5 year time scales which have long passed. Considering the money they threw at it..... Agreed petrol is not what they would do, what i expected to see by now was progress on small scale hydrogen cells. Yet even that seems stationary compared to other energy areas.
So many reasons i would prefer diesel, or even glow unmodified. But i am aware i do have a prejudice against the model petrol engines.

But as above, I cant argue against those kinds of figures. I am trying to remember why petrol was a no go originally, I will have it all in a folder printed out somewhere. Nothing springs to mind, except maybe some safety rules. Until i find my original file i am guessing. For Aus the applications are never ending, my own interest was alot more niche, we dont get wild fires much bigger than a few acres! Hard to get lost in the Uk when 5 hours walking is going to get you somewhere .

But search and rescue in the hills etc, or even places like the coast line at say Torbay in summer, or Cornwall in the surfer beaches. Plenty of applications, weight wise depending on the application there is alot of scope with that kind of capacity. Time is the killer at the moment, coupled with a shove it all in box and forget where you put it mentality! I will order one of the motors, they are alot smaller than i using and better output. I am also told there are some new propellers out that offer significantly better lift.

Although so far these have been harder than unicorns to actually track one down! How come you abandoned the project your end?

 

[large_ghostman]

I'm guessing that fuel engine response isn't quick enough to get the fine level of control a drone needs.

Mike.

Spot on, the difference in response is the same as Usain Bolt V grandma with her legs in a sack and a zimmer frame, the motors changed speed constantly by very small amounts. It gives you incredible control especially in say wind or quick response situations like when its self flying.

The other less important reason is power loss, with electric you have a battery back up. Ok you can get a motor go down but that is extremely rare and on a hexacopter not that critical. But also weight is an issue and the amount of fuel needed. ESC tend to go, but mainly because most people buy them in, almost everything on the original one i did was hand rolled.

 

[Pommie]

Managed to find the fuel consumption figures for OS engines. The interesting comparison is the petrol powered 32cc 2 stroke -v- 32cc 4 stroke. Both produce the same power but the 2 stroke uses twice as much fuel.

Mike.
Edit, check out the glow engine consumption!!!

 

[large_ghostman]

Managed to find the fuel consumption figures for OS engines. The interesting comparison is the petrol powered 32cc 2 stroke -v- 32cc 4 stroke. Both produce the same power but the 2 stroke uses twice as much fuel.

Mike.
Edit, check out the glow engine consumption!!!

I was aware glow was poor, what surprises me is the current difference. I got a similar pdf somewhere from way back when 4 strokes were first out, if i cant find it on the computer i am sure its on the net somewhere.

The figures they give are getting close to diesel figures, like car figures I would subtract a little. But even doing that these engines have come some way!! Really impressive, i was expecting that engine to be around £1,200. Well at least it explains why you can pick glow engines up for peanuts on ebay! How close were your figures in real world tests?

First time i seen 4 stroke get parity power wise, you got to admire the engineering thats gone into that.

 

[KeepItSimpleStupid]

Shooting down drones:

https://www.thefirearmblog.com/blog/2017/09/22/__trashed-37/

 

[large_ghostman]

That makes alot of assumptions. For a start while it was staged, by the time they noticed the drone they were already dead! Second why is there this continuing assumption that a drone swarm would be controlled by a constant signal? The drone only needs telling where to end up or patrol, what to look for and what do once its found the target.

Yes ideally you stream back video and information, but even 3 years the tech was there for autonomous flight. The other problem is many of these devices assume the frequency is above the gig range and mainly around the 2 gig range. mine used sub gig frequency mainly because it was what I managed to get hold, apart from the data it sent back it there wasnt any control signal sent to the drone. It was designed to self fly and make certain decisions itself.

if you watch the military swarm video you are dealing with a large number of drones, even if you got 1/4 on the ground you have the problem of a true mesh system, most will be taking off again the moment that thing isnt pointing at them. The most effective anti drone device i have seen work so far, is a raptor! These would be extremely hard to defend against, but most electronic means are not going to save your ass!

So in a war or aggressive situation your patrolling and spot 10 drones heading straight towards you, you point the device and as your about to pull the trigger....... some guy comes around the corner with a bow and arrow. The old Zigbee mesh network has been updated and i am aware there is another that is close to production, Silicon labs are playing the cards very close to the chest but the tiny amount of info i have, the next generation of mesh systems are something else!

I also dont understand the mindset of slave/master with drones, trues mesh networking makes this redundant. lets take a car and CAN network, ditch all but the basic bits and what you have is a number of independent systems that can communicate to each other. So very very basically lets take the brake system, it dosnt matter which system tells the brake to ease off or switch pressure the brake system excepts the message and does it.

Ok in a car it isnt actually like that, but the CAN system is designed that way. Its just a hierarchy of messages on a bus with a destination, true mesh networks take this further, each node can actively search for the fastest path for the most urgent messages.

There is a hotel in London that has the silicon labs Zigbee system, it controls everything in the hotel. A while back i visited the place with the UK sales Director to see the system working, while it was way above my head i got two feelings with the system. One was complete awe at how robust it was and how much of it you could hack out but the system still function, the other feeling was almost a creepy one. You see it working and watch various bits turned off, but the function being taken over by another system, you start to feel really pleased the thing dosnt have a brain as such and no one has yet decided to try and make it think for itself.

But in all honesty i doubt we are that far away now from what was pure science fiction maybe 10 years ago. There is a really really old film i saw on TV, War Games its called and its the normal computer in charge of nuclear weapons type film. I bet everyone watching that when it came out slept just fine after seeing it, watch it today and you got wonder just how far off we are from building a system like this if we really wanted too.

As an aside anyone been to a TED talk? I was at one in November, i have been trying to think of the guys name but its long gone... Anyway one of the speakers was a mesh network specialist, some of his predictions for the next 5-10 years made you sit and think.

 

[large_ghostman]

Just thought of another thing..

How many you old guys remember driving way way back without say car insurance or MOT or even Car Tax? Go back 15 years in time and ask yourself what you would think of a system that could identify you and check your criminal record, while you drone at 80mph down a road and within 10ms have checked your number plate and knows if you have insurance, if so with whom you have it. It knows how long you got on your MOT and when you paid your car tax, it does this from over 1/2 mile away with hundreds of other vehicles around at the same time. And just to add insult to injury it knows by how much you just broke the speed limit over a 1/2 mile section!

No idea how many people are aware of this but, Kings Cross station in London has a room with monitors in. Nothing special about that, but i only know of this room because i happened to be standing by a door when it suddenly opened and two guys came out walking fast. As they came out they closed the door behind them, but it was enough time to see a huge bank of monitors and little green box's darting all over the screens and turning red on peoples faces.

my assumption is this is in every main Train station, airport and other similar place. What is it? I have no idea but £1-£100 its top of the range facial recognition linked to men who would wear dark glasses if we had sun in the UK