Thread by @Jack_Watling

How to Keep a Drone Alive: a thread. While the loss rate of UAVs in Ukraine - on both sides - is very high, it doesn’t have to be this way. Making UAVs survivable, however, comes with some trade-offs and support requirements. 1/20

What is downing UAVs?
1) Jamming and hijacking of their control frequencies.
2) Denial of satellite navigation.
3) Saturation of their electronics.
4) Shoot downs.
2/20

Most UAV losses are DJIs, commercial UAVs without any hardening, that has built in defeat mechanisms because the company wants to export and sell them to civilians, and therefore need to be controllable by law enforcement. 3/20

First rule, therefore, is don’t use openly available firmware and software to run your UAV that the enemy can exploit. Also, patch the system regularly so that when the enemy captures the UAV they are chasing a moving target. 4/20

To defeat jamming you could use a frequency hopping software defined radio. That is expensive though. Instead, you might have the UAV revert to existing orders. To follow those orders the UAV needs to know where it is at all times. 5/20

First way to do that is satellite navigation. But this can be spoofed or jammed. So then you need inertial navigation. For that to work though the UAV needs to know its satnav is being interfered with, otherwise its ‘return to base’ might end up being ‘land next to enemy’. 6/20

For this reason Iranian UAVs (for example) tend to have 4 or 5 antenna arrays that compare the signal from Glonas, Galileo, GPS and Beidou to see if they give wildly different readings. They also compare directional signals to see if they get conflicting signals. 7/20

If these UAVs do get conflicting signals they revert to inertial navigation, calculating its speed and altitude from its last known position and adding other data, from sensors to confirm its position. The more sensors, the more accurate, but also the more expensive. 8/20

Against inertial navigation the enemy might decide to just overload the electronics. If you know what chips are in a UAV you can literally saturate specific parts of the electronics causing them to break. 9/20

The way around this problem is to shield the electronics. Hardening is fairly simple, but it can be heavy and bulky. Here is the shield around the ‘brain’ of a Shahed-136 for example. Weight means a bigger UAV needing a more powerful motor. 10/20

Bigger UAVs of course have a larger radar cross section - assuming we aren’t using radar absorbing materials because then we are in a whole different world as regards cost - and are therefore more vulnerable to being detected and shot down. 11/20

Avoiding being shot down comes down to route planning and pilot skill. Using terrain to mask on the approach, popping up, and sitting high over infantry to avoid being detected all help. Of course, this requires an intelligence picture of where the enemy is. 12/20

Route planning can also defeat jamming. If you have a ‘cut’ of the electromagnetic spectrum, you can use it like a map to plot a route around areas of intense interference or along seams between jammers where onboard sensors will best resist their effects. 13/20

While good pilots can massively improve the survivability of UAS, things like reading an electromagnetic survey is a skill. You therefore need a training pipeline for pilots. That takes time and money. This has to be factored into the cost of the capability 14/20

You also can’t plan off an electromagnetic survey unless you have the ability to generate these in real time. That probably requires a NEO satellite constellation, software to stitch their returns together, the infrastructure to integrate it into route planning software. 15/20

And this is really what I am getting at in this thread. If you go cheap you will lose a lot of UAVs. The costs of protective measures go up steeply. For example, hardening the electronics on the Shahed-136 (and some other changes) saw its price jump from $20,000 to $40,000. 16/20

Decent inertial navigation and the sensors for it are great. But what you end up with is the Orlan-10, which has a price point of $80,000-$120,000 depending on payload. The cost and complexity of manufacture then starts to shape what you can do with it. 17/20

I expect Ukrainian UAV survivability to increase over time as they get more mil-spec UAVs and their training pipeline distributes more skilled pilots. But so long as every infantry platoon wants UAVs in overwatch there will be a demand for cheap and attributable platforms. 18/20

The interesting question for those countries seeking to adapt their own forces to integrate UAVs is where they see the price point and the use cases, and where they accept trade-offs in design. 19/20

My final observation is that UAV and Counter-UAV activity is very cat and mouse, so you’d be foolish to go all in on one system. There is real merit in having multiple tiers and being able to iteratively develop them in response to the evolving threat picture 20/20