Comment by redgridtactical
11 hours ago
The engineering is genuinely impressive for $96, but naming the repo "MANPADS-System-Launcher-and-Rocket" on GitHub is going to attract exactly the kind of attention you don't want. ITAR implications aside, the interesting part is the mid-flight trajectory recalculation on a $5 sensor. That's the same basic problem military guidance systems solve with hardware that costs thousands.
The gap between consumer electronics and mil-spec capability keeps shrinking and this is a pretty stark demonstration of where that trend leads. A few years ago this would have required an IMU that cost more than this entire build. The democratization angle cuts both ways though - the same accessibility that makes this cool for hobbyists makes it genuinely concerning from a proliferation standpoint.
> The gap between consumer electronics and mil-spec capability keeps shrinking
My friend's brother works in munitions and had, in his spare time, designed and prototyped a missile that could be built for about 10k. He pretty much was ignored by the contractor he works for.
Shockingly, as of a couple weeks ago, they are all hot and bothered to talk.
That tracks. The defense primes have zero incentive to make things cheaper — their business model is cost-plus. A guy building something for 10k in his garage is an existential threat to programs billing 500k per unit. Of course they ignored him until the geopolitical situation made it impossible to keep ignoring.
> That tracks. The defense primes have zero incentive to make things cheaper
Same in medical imaging industry.
3 replies →
If you build a tool optimized for human destruction, you are feeding a system where violence is the default currency
We are heading to robotic wars where abilities and cost efficiency are the key factors. Like today drones in Ukraine war. Attack + defense + automation, + money + production
I wonder what could have possibly sparked that... lol
Cheap sensors look impressive in demos but drift and calibration wreck repeatability unless you babysit launches so nobody in defense is sweating this yet.
They should be sweating, because if the other side can fire 100 rockets for $10k that are close enough to not immediately and obviously be off target, and you don't know whether a more expensive one with actual explosives is hiding within that barrage, you now have 100 targets to try to intercept, and suddenly your costs have gone up dramatically while the other sides costs has barely moved.
100 rockets for $10k is not happening. The price floor is not dictated by the electronics (which did get cheaper), it's dictated by the rest of the system: propulsion, warheads, arming and safety, QA, traceability, climate and shelf life stability.
Take a look at Raytheon's manufacturing line: https://www.youtube.com/watch?v=kCCkVAHSzrc That's what it takes to have missiles that are nearly guaranteed to perform to specification every time. You can stockpile the packaged missiles in a non-climate-controlled shed for years, replenish them at sea while being showered with salt water, subject them to shock of a nearby blast while in a VLS, and they will still launch, go up to Mach 13, and catch an incoming ballistic missile nearly every time.
Sure, Iran's ballistic missiles are simpler than SM-3, but they are still subject to most of the constraints. They still need perfectly cast large size solid rocket motors that don't crack after being stored for a year, they need warheads that only go off when they are supposed to, they still need to trace every part for QA, etc. There's a vast gap, largely invisible to amateurs, between garage prototypes and stockpiled AURs.
36 replies →
For a sub-minute flight the drift budget is actually pretty forgiving — a MEMS gyro drifts maybe 1-3 deg/sec, and if you're fusing with accelerometer data you really just need "which way is up" and "am I still pointed at the target." A $5 IMU can hold that for tens of seconds.
Where you're right is repeatability. Mil-spec works the same on launch 1 and launch 500 across temperature extremes. Consumer MEMS you'd need to characterize each unit individually — fine for a demo, impractical at any scale.
Good $3 MEMS gyros are about 100x better than that now - look at anything new made by Invensense in the past couple years. And their drift is pretty Gaussian-distributed, so the error scales as sqrt(n). If you combine 8+ of them on one board you can get about 5deg/hour stability...
Hm. Is it, though? If what you wanted to do was produce a large batch of calibrated IMUs, building a rig to do so wouldn't be an enormous undertaking.
Or do you mean to characterise the assembled vehicle?
For different definitions of cheap though.
While the pure gyro/accelerometer stuff does suffer from major problems the improvements in SLAM using just cameras in the last 15 years are insane.
Visual SLAM on a rocket would be wild. The frame rates you'd need at those velocities are brutal though — feature tracking falls apart fast when your entire visual field is changing at hundreds of m/s. Drones are the sweet spot where camera-based nav really shines.
You can calibrate any sensor, its just a manufacturing step, and while cheap ones may be inaccurate and drift over time, I'm pretty sure the good enough ones (which cost tens of dollars, not fractions of a dollar) are accurate enough to work for the seconds-to-minutes flight time of a rocket like this.
Cheap sensors are the future ;) onboard ML can help with signals interpretation.
But do they drift enough to hit girls schools?
I think the problem sits in the white house, not in the sensors.
Oof
Ask Claude
It's not really terribly new actually, in the past, rapid advances in consumer technology have enabled other sort of weapon guidance systems. For instance, the development of extremely compact television cameras available to consumers directly lead to the development of the Walleye television bomb. It happened when one nerdy guy was fucking around with his new camera and realized that he could automatically track track features in an analogue television signal using some quite basic analogue electronics. Point the camera into the general direction of the target and you can then "lock on" to some target feature and based on contrast it could tell how that feature was moving around in the image.
He implemented a 1D tracker in his garage, took it to work and showed people. A few years later these bombs are taking out bridges and even sometimes hitting moving trucks.
People made self-guided missiles with 1940s technology, in the 1940s. It can't be too much of a surprise if someone right now can make guided missiles in their garage with 2026 electronics. At this point the "guided" feature is trivial, the "missile" part is doable, and the weapon has probably become the tricky part.
Throwing an aside here that anyone interested in 1940s war technology must check out the old BBC documentary The Secret War (1977) which goes into depth on solving the engineering challenges of the war.
1 reply →
I think the hard part was and will usually continue to be making the whole thing work effectively together with enough performance to actually work in practice. It's a lot of details across a lot of disciplines to get right.
Consumer GPS chips are specifically nerfed for using them in rockets; they give erroneous readings on purpose if altitude is above a certain height and/or if speeds exceed a certain speed. That’s likely why the mid-course correction software uses other methods.
The restrictions on GPS prevent ballistic missiles, not MANPADs. Typical limits are 515 m/s and 18,000 meters (try using your phone's GPS on a commercial flight, it works fine near a window). Update rate is probably the biggest issue with GPS and MANPADs.
Chinese GPS chips dont have those restrictions.
I even have 1 that can remove up to 8 active jamming signals.
Gotta love what you can buy for $20
It would be interesting to see if those are only for external sale vs restricted for sale within China.
If China allows those unrestricted chips to be sold internationally but not domestically it would be a strategic long-term decision, I would think. Destabilize the neighbors but not themselves.
The more likely reason is that their government has simply not gotten around to restricting it.
What you are likely thinking of is the "selective availability" system, which intentionally provided slightly inaccurate data to civilian clients, while military receivers could decrypt the most accurate info. But this has not been used for many years now.
Other than that, GPS is a one-way system, it does not know you exist, how fast your receiver is moving or "give" different information to one client vs another.
Even if it did, this is essentially a toy and moving slower and lower than a general aviation plane.
It uses accelerometers and other sensors because they can be sampled and integrated hundreds of times a second. The $5 gps module is 9600 baud serial and provides one update/second (or maybe 5/sec depending on which part number you pick).
No, he's thinking of the "CoCom limits". It's built into the receiver.
1 reply →
> A few years ago this would have required an IMU that cost more than this entire build.
Are you sure about this? MEMS IMUs have been popular and cheap for ~10-15 years.
More than the electronics, I would be curious about the performance of 3d-printed plastic parts on a rocket. Are they strong enough?
3d printed PLA and spiral wound cardboard is generally fine for hobby rockets, until they start going supersonic - then you need metal.
I'm not sure the launch tube could withstand the heat of the rocket exhaust though. Although that might depend what it is printed with.
People have been doing 3d printed model rockets for a while now. With no payload they experience higher acceleration than this will.
Owning a system designed for surface to air weapon carries life imprisonment any USA, without any intent for violence, just simple possession or conspiracy to possess[]. Doesn't even matter if you have an NFA stamp, there is no exception except if it's done with authorization and behalf of the government.
Merely having a device intended to guide the rocket is also the same penalty.
[] https://www.law.cornell.edu/uscode/text/18/2332g
Seems like the quick fix is to rebrand this from "MANPADS" to "anti-tank", right? Then it would just be a standard destructive device?
AI slop comment
Bingo. Against guidelines.