None of those is going to be a good idea using in a stadium, above a crowd, with buildings in the background (less than 100 yards away), etc. which are the big threat areas.
Drones that can carry a couple pounds or more of C4 (which is a lot) can fly literally 10 ft above roof tops and crowds at 100+ mph while zig zagging through buildings and alleyways - fully autonomously.
This isn’t theory, there are live demonstrations on YouTube.
Zigzagging through a forest fully autonomously (with no GPS) - there are videos on YouTube of a Swiss company doing similar. [https://youtu.be/m89bNn6RFoQ]
Folks have been using those and similar (cheaper) drones to drop Grenades on tanks in Ukraine.
As the autonomous flight is harder with smaller drones (less computing payload available, worse sensors), the only reason you’re not seeing both of these parts together in urban environment is the implication (for national security). It is there.
Here's the video of that https://www.youtube.com/watch?v=Lr7L2t-svJQ&t=16s - that's below 5mph and carrying an external payload of 0 grams, and the reason it's in a forest is because you can't fit them all in an alleyway and you need multiple in order for them to work.
>> Here is a YouTube video of a demonstration with a hexcopter
At 22mph, the claim was 100mph+
>> And here is a dude flying a DJI Inspire
At around 20 mph max but mostly hovering and for 8 minutes at that before running out of juice
>> Folks have been using those
No. People like yourself can easily dream this concept. It stands to reason, why wouldn't it be possible? The truth is that battery technology is no where near good enough yet.
It's the first idea tried in Syria, it's been tried thousands of times at this point and yet we have more documented cases of the aggressors killing or injuring themselves than of the 3 documented cases where this actually worked in practice - all 3 of which i think everyone can agree from the footage are pure fluke. Zero aiming was possible, it was just release and hope for the best in all 3 cases.
What people are actually using them for is for recon and aiming of artillery. They're effective in these roles and that's why they're used for this rather than strapping heavy explosives to them.
>> autonomous flight is harder with smaller drones (less computing payload available
That's not true. Compute isn't a restriction in small drone autonomy. A basic F4 is wayyyyy overkill for flight stability and control. Waypoint navigation is even lower power requirements since it amounts to a simple loop not much more complex than:
1. Get current GPS coordinates
2. Get current compass heading
3. Send directional input via serial peripheral to F3 or F4 flight controller
4. Repeat
The flight controller is doing far more processing than this - gyro input & smoothing for each of 3 axes, accelerometer (same), GPS location at approx 10Hz, optionally barometric input (rare and usually not necessary). These inputs are often processed through kalmann filter before finally calculating the torque command to issue to each ESC's microcontroller. This only happens at up to a few hundred Hz so you can see compute power is not a restriction.
In larger aircraft it's harder because you can't beat physics, in a small drone i can change the amount of torque applied to a propeller 4+ times per rotation in toy off the shelf drone tech today. With larger craft that have non-electric motors this is simply not possible. You can't instruct a mechanical engine to regulate its power output with the same fidelity of an electric motor.
>> worse sensors
The only required input sensor is a gyro - in practice we use accelerometers and GPS in autonomous drones but there are ways of avoiding the need for these (e.g. visual inputs / optical flow as a correction for drift). MEMS gyros are cheap and effective, what better sensors do you need?
Your reasoning is uninformed / dreamed up based on watching too many movies and not being familiar with the actual discipline.
Better algorithms are a restriction. The science / theory part of implementation was achieved by all projects fairly quickly (ardupilot, clean/betaflight, etc. etc.) - but the art part, the how it feels and performs in practice, is an unsolved problem under continual development.
>> drop Grenades
Too heavy given current battery technology and lack of aiming. These are not effective weapons which is why they have limited success. More people have been killed in war zones in the past decade by garden spades than toy / photo drones with explosives.
TL;DR until batteries technology moves forward by a few generations, the concept of a small drone as an offensive weapon will not be better than that offered by something like the switchblade drone. Which when you cut through the excitment, is less valuable than conventional weaponry in most fight scenarios.
Compute is not a limiting factor in a small drone for basic flight operations, it is a limiting factor for real-time optical guidance (aka in GPS denied areas and real-time object/target recognition). Which is exactly what I was referring to.
Those examples yes are not showing 100mph. I’m afraid the video I’m referring to has been pulled, and the in-person demonstrations I’ve been asked to not talk about or publicize (now that I follow up with the parties involved).
The battery technology is more
than sufficient for what I’m describing. My M300RTK has no issues carrying 5+ lbs for 15 minutes or 1-2 lbs for 40 minutes, which is more than sufficient, and scale is more than adequate here.
You’re a couple years behind on this assessment. I literally have a 10 lb ‘hunter killer’ autonomous anti-drone drone from a company I’m not going to name sitting a few desks away from me right now.
>> I’m afraid the video I’m referring to has been pulled, and the in-person demonstrations I’ve been asked to not talk about or publicize
Interesting.
Let's take a second to take this from back of a fag packet physics.
To produce a 100mph drone is easy - lets take one of mine (there's nothing special about it) as a starting point and then lets see what we can change to achieve load carrying while retaining this speed.
Each motor & prop combo produces 1.5kg of thrust. They do this while drawing 40amps each from a 6s - which they can sustain for up to around 10 seconds before burning out MOSFETs on the ESCs (not to mention deforming the battery plug due to 160amps passing through a 60amp connector). This craft has a weight of just under 350 grams and when coupled with an 1800mah 6s, has an all up weight just under 650 grams.
We want to carry "several pounds" - lets make it easy, lets say 2lbs - and retain that 100mph speed.
Adding 10 grams prevents us hitting 100mph, we need to scale up our thrust. First problem - there's no other prop pitch or blade count for this motor size which will increase thrust while retaining prop speed, so we need to move up a size in motor. Since we need to carry 2lbs, you might think we only need each motor / prop combo to produce at an extra 0.5lbs but in practice we're going to need more than this because we need to sustain the aircraft vertically, whilst also providing enough thrust horizontally to reach the target speed, so in practice we will need more than an extra 8lbs of thrust (remember how 6kgs of thrust is needed to carry a 650g craft to 100mph - you don't just get to add 2lbs of thrust and call it done).
Problem - we need a motor / prop combo that produces above 2.5kg thrust each - so we're into 14 inch propellers and we need a motor heavy enough to take around 60amps at 6s (i'm being conservative to try and make the maths work out in your favour but i hope we can both agree that only 60amps is pure fantasy here).
That's a new problem, our weight went up and so we need even more thrust to compensate.
That's another new problem, we have more thrust but now we need a power supply capable of a higher discharge rate - that kind of battery entails even more weight.
But now there's a new problem, our larger props mean we're well away from the happy 300-400mph tip speed and now we're approaching the speed of sound - which of course means we can't fly.
But now there's a new problem, the battery that can supply that discharge does exist, the only problem is it weighs almost 2kg by itself.
And so we're back to you can't fly 2lbs payload at 100mph on today's battery tech.
Camera operators have been trying hard for years at this point to put a RED or some other expensive heavy camera / lens combo right next to race cars on track. There are two solutions for that today: 1 - fly the heavy lens and get that sweet image but do it at 20mph or 2 - keep up with the 100mph+ action but film it on a gopro or similar.
You're totally right, I have no idea what I'm talking about!
Some random babble:
I'm glad you're running 6S at least. I was worried you were running something like a 2S rig.
Is your your assertion is no one can build a quadcopter style drone that can go 100mph and carrying a 2lb payload?
Or that you can't add 2lbs to your 6 inch racing style quad and go 100mph without changing something design wise?
Because the first has a number of commercial products (albeit using hybrid powertrains due to longer desired range [https://soninhybrid.com/], [https://dronedj.com/2022/01/04/firefly-really-heavy-lift-dro...]). They'd have no issues doing the same thing over shorter distance, but folks wouldn't buy them - because for the commercial use cases, speed is only useful when it means covering more distance (and is therefore is really speed + distance). Tweaking some variables and changing some thinking would get you something from DJI that would have no issues, but it wouldn't sell, at least to the typical commercial or recreational market.
I would be VERY surprised if DJI didn't already have a healthy set of military contracts for doing exactly this, but if so, it isn't anything public.
The second yeah, should hopefully be obvious. It isn't THAT far off though, as you'll see below.
For the situation I'm mentioning, there are different tradeoffs, and I'm doing the best I can without causing problems re: confidential details:
You're getting confused on amperage, because peak amperage output is a matter of parallelizing batteries and design of the cell, and is NOT battery chemistry limitations (per-se). You can double your peak available amperage by changing the way your battery is wired. It doesn't add weight. It does halve capacity however, but that is range, not peak power.
Thermal management is one of the bigger issues in practice, and higher voltages tend to help a lot there (less resistance in wiring for a given amount of power delivered). It also results in less weight.
If you had a 40 amp peak 6S pack (which is only ~ 800 watts peak), you'd get the same per-pack equivalent draw with 3x in parallel (120amp total is 40 amp per pack) or 2.6kw. But frankly, 12S (or higher voltage) is better in every way but expense/parts availability if you're handy with electronics. A 12S 40A pack delivers 3x the actual power for the same wiring, assuming components are insulated/rated for the voltage.
And yeah, I know 'peak amps' numbers on most batteries are a lie, first thing I do when I do a build is instrument power flow, including amperage under load.
ESCs/motor controllers would likely be your primary sourcing difficulty, but you can get COTS 120 Amp individual motor ESCs [https://www.getfpv.com/electronics/electronic-speed-controll...] inexpensively if you're running 6S, and not too crazy expensive if you're running 12S like that particular one is designed for. That one also already has BLHeli_32, but most of them can be re-flashed, and then every COTS flight controller can deal with them pretty easily. That particular ESC is rated for controlling 32kw PER MOTOR, by the way. You could use a 4 in 1, but as you note, meh on total power rating. They are compact. They also have thermal management issues. Individual ESCs can get out into the airstream easier, and if you fry one it's less work to replace.
Motors with the desired design criteria would be next, sourcing difficulty wise, but that would depend on specific propeller/fan design. There are a LOT of options, and winding your own actually isn't THAT hard (or re-winding a COTS one for different characteristics).
Most likely anyone who is doing what I describe at a state actor level is going to be doing their own ESC design, and ESC + flight control firmware anyway, and the electronics to make a compact ESC for arbitrary amperage/voltages are very well understood. They're used in everything from elevator motors, to milling machines, to handtools. If it's someone doing a military style contract, they'll of course do their own battery packs, nominally to maximize their performance envelope for whatever mission they're trying to get the gov't to cough up cash for, but also because COTS will make them look less awesome.
But it's easy to rewire COTS packs to do pretty much whatever you want tradeoff wise, re: voltage, amperage, etc. Lithium batteries are amazing for their ability to handle very high peak power draw, and in the scenario that I'm describing, the bottleneck would likely be thermals, not chemistry or weight, and considering the time window being discussed, there are lightweight and interesting ways of handling that.
100 mph == 1.6 miles per minute after all, and it's highly unlikely the drone would be anything but hot gas and rapidly disassembled small parts after that amount of time.
You're getting confused on props and tip speed, because you aren't thinking of, or aren't aware of, ducted fans which specifically solve this problem by giving you much higher thrust at the cost of a little (or a lot, depending on the frame design) extra weight. A little less than double, depending on conditions.
Prop pitch and design would look weird, comparatively, depending on the mission, from what you're used to seeing, and depending on the specific numbers, elements like blade count, leading and trailing edge, chord length, etc. tend to differ.
High agility has different airflow and stall characteristics than maximum speed and/or maximum weight and propellers designed for one tend to be not great for the other.
That said, there are a number of aerospace engineers quite familiar with the problem, and it's well solved at this point, up to ~ 160ish mph. And by well solved, I mean Embry-Riddle has it in their basic curricula, and there are a LOT of graduates rattling around. They're pretty smart too, in my experience, and are itching for problems like this to solve. Especially if you are ok with things exploding until they get it right.
After that it requires more complex solutions (usually involving a degree of direct jet propulsion) and some amount of fire.
Alright, I'm done. Anything more and I'm going to get in trouble.
Drones that can carry a couple pounds or more of C4 (which is a lot) can fly literally 10 ft above roof tops and crowds at 100+ mph while zig zagging through buildings and alleyways - fully autonomously.
This isn’t theory, there are live demonstrations on YouTube.