DARPA is developing self-guiding bullets that can change course to better hit moving targets


Normal Tuesday night for Shia LaBeouf.

1 Like



Have they improved on this prior art? https://www.youtube.com/watch?v=kbRZKrvAZ7U


Isn’t it striking how some ‘futuristic’ props age extremely well, continuing to look like they’ve been time travelling for decades after their design, and some date themselves in the blink of an eye?

Despite smart bullets still being pretty cutting edge, those circuits just scream OLD in every detail.


The rocket propellant still looks believable though, no?

1 Like

That aspect hasn’t changed much(at least not visibly, fancy-but-nanoscale formulation details may vary), so it looks like it could have been manufactured tomorrow. Just the circuitry.

Are they making these bullets for Space Invaders?


Frankly, the snide saying “You could run; but you’d just die tired.” was creepy enough without the help…

1 Like

I hope they call it the “peace keeping” bullet!

1 Like

That said, what they will use as flight control surfaces and navigation sensors? Dead reckoning? Trajectory tracking from the gun and a radio or laser link to the projectile? Will the projectile be spin-stabilized and one point of control with angle of rotation sensing, or will it have active stabilization on both axes? What actuators, piezoelectric perhaps? So cool toy, so many questions!

The electronics looks old on the cutouts. Can be made way smaller, I can imagine power feeding from a supercap charged inductively just before shot, or a thermal battery lit like a tracer at the time of shooting. But for prototypes the not-so-small may be an advantage regarding to the developers who actually have to handle the circuitry. Or, more likely, it is an artist rendition as the actual appearance is likely to be classified.

1 Like

They already have bullets, and the bullets have the advantage of being cheap. The only people who would benefit from developing extra-expensive bullets are…

…oh. Never mind.


Serpentine! Serpenti…

1 Like

Came looking for Valiant, left satisfied.

This graphic explained a lot for me. In case embedding doesn’t work or whatever, the bullet follows a laser point down range. After you fire, you keep the rifle trained on the moving target so that its built-in laser can “paint” the target for the bullet. The bullet has fins and can adjust its trajectory up to 30 times per second. So it seems the main thing, as usual, is follow-through. You keep the rifle on target after the shot, and that’s the key.


Bullet makers? I am not following…

At any rate, super neat tech, but it will probably have limited uses if perfected. Right now I think it is for the .50 cal which is a huge bullet. Your average infantry won’t use it unless they make it smaller and cheaper.

Tracking Point scopes are another new thing and dang promising for increasing accuracy. Some really futuristic, Shadowrun like stuff.

1 Like

I’m curious if DARPA finally incorporated the Predator algorithm for this project: http://www.tested.com/tech/2109-object-tracking-algorithm-learns-from-mistakes-video/

So looks like a conventional laser guidance.

The size of the bullet and the power requirements limit both the sensors and the processing power, so the guidance platform will have to be a bit rudimentary. (Although a 10 inch long, .50 round has quite a lot of space inside.)

The laser beam is likely to be modulated, using some sort of code so the bullet knows it is its laser. (This will also make the vehicle crew aware that something is happening, as there are electrooptical devices to tell you that you are being targeted by something just like this. It is an array of photodiodes, generally, with a high-pass filter to let only the rangefinding and guidance laser signals through.)

If the bullet is spin-stabilized, there will be a sine signal from the optical sensor (tuned to filter only the laser wavelength, to suppress background, and then looking for the modulation of the beam), with amplitude corresponding to the error, and phase corresponding to the relative angle of the error vector to the projectile spin angle. Similar principle as tracking radars, or spinning-sensor heat-seekers. I guess this will be the easiest way, as this can be done with a single sensor and maybe a single fin.

If it is not spin-stabilized, it could not use any of the existing sniper guns and will have to be delivered as bullet-weapon combination. So my guess again goes to spin stabilization.

There may be countermeasures available, along the line of laser jammers (perhaps receive the guidance signal and rebroadcast it with amplitude modulation corresponding to the bullet rotation speed, introducing error into the guidance), or near-infrared smoke generators to absorb the beam.

1 Like