From a piece in today's New York Times on the push for more and better military microdrones:
There are some 4,800 Ravens in operation in the Army, although plenty get lost. One American service member in Germany recalled how five soldiers and officers spent six hours tramping through a dark Bavarian forest — and then sent a helicopter — on a fruitless search for a Raven that failed to return home from a training exercise. The next month a Raven went AWOL again, this time because of a programming error that sent it south. “The initial call I got was that the Raven was going to Africa,” said the service member, who asked for anonymity because he was not authorized to discuss drone glitches.
Comments
Deleted my comment because it didn't make sense.
I'm sure we already have means of adding encrypted video feeds and control interfaces from existing MILSPEC hardware. The issues with existing hardware are:
1. Is it small enough for sUAV application, and
2. Is it designed to be expendable, and the encryption unbreakable if the UAVs fall into enemy hands (as we see in the picture accompanying the original post!).
Is there really an issue with using open source avionics, and just secure the data/control feeds?
@Ethan
"You can still use open source data link protocols while having them secure. Its just a matter of adding an encryption wrapper."
Prime attack method on encyption algorithms rely on having known unencrypted message, or algorithmic dependencies known in unencrypted message.
Using open, known and very frequenctly used protocols NOT DESIGNED to be wrapped up, is opening the door for such attacks. It also limits the domain of possible encoding keys since you know the subset of messages that coul;d be passed to encode stage.
I agree with both sides (and am in the military myself!). As far as the security issue, I would think that the main concern would be the datalinks (both the GCS transmitter if operating, and the video feed).
The other issue with open source is usually the TOS requires any work done on the back of an open source project is also required to be either non-commercial or open source as well.
@CA "Why does the military have to operate under different rules? Why couldn't they use open source?"
Maybe because if you assume that the work of the engineers is almost free, has to be donated etc or they shoudl earn not significantly more than UAV user or truck driver, you will fall progressively into communist rules which assumed nobody should be better than working class.
All this just to wake up 50 years later, that money-driven extravertics can deliver better, cheaper, faster (as long as they have competition in form of aggressiely communist bloc).
Military is not supporting such forms of 'knowledge should be given free' pressure, what fits well their original role of preserving capitalist point of view.
Such failures only and only occur due to human error, bad weather condition at some distant waypoint, Lipo underperformance, preventive maintainance of the airframe, sensors and electronics not done properly at regular times.
Once the firmware is well tested and hardware is ok then there is not way it would go bad.
@Bosak
"Actually there is even longer tradition to transfer from civilian sector to military:
-Even civilian GPS has its military version!"
Isn't it the other way around?
Comment
-
1
-
2
-
3
of 3 Next