Hacker News Reader

Comment by unabridged | Hacker News Reader

How useful are fighter jets? Shouldn't we be building swarms of unmanned drones instead?

Basically in any reasonably capable enemy airspace, you need to fly with emissions control - ie. no radio, no radar else passive detection systems will find, triangulate and guide a SAM on you.

If you're doing drones, this means programming fully autonomous AI to identify and kill targets - which no one quite wants to do just now. For the same reason, "swarms" are tricky since you need radio coordination. Drones have proven useful against significantly less technologically capable adverseries but don't forget the drone captured by Iran by what appears to be control signal spoofing

>Basically in any reasonably capable enemy airspace, you need to fly with emissions control - ie. no radio, no radar else passive detection systems will find, triangulate and guide a SAM on you.

A cruise missile on final approach don't need forward radars or datalink back to base, and the same goes for almost any kind if missile used since the 2nd world war. Just because remote controlled drones don't work under heavy ECM don't mean the military don't have robots that can. And some of the US's drone use a directed signal towards an satellite instead of omni directional RF signals for control.

The Problem here is that stealth don't add a whole lot of value to the observe first then shoot kind of missions where you really do want an human in the pilot seat, so multiplying the cost pr air hour by a order of magnitude to get some level of radar stealth is the kind of economics that causes mission failures in real conflicts where real life supply problems dominate.

For the F-35 with it's dependence on BVR missiles for defense going silent also means that if it can be spotted by a ground observer(which it likely can) it faces a high risk if the enemy have enough time to scramble an interdiction patrol into visual range, a tactic that was perfected by the RAF doing the battle of Britain(using radars that could probably see the F-35).

You can forget the ground observer. Stealth is the reason why it can fly high all the time to avoid visual/audio detection. Non-stealth planes like F-15E are forced into High-Low-Low-High altitude flight profile. where low level ingress and egress will e.g. drastically reduce range. F-15E going around threats both ingress and egress uses about 30% of its fuel to do so. F-22 would use around 6% in comparison. So basically stealth means that you can waste less fuel on threat avoidance.

Yes. I expect manned fighter jets will become obsolete within 20 years. A few thoughts on the subject:

1. Drones can handle much higher G-forces than manned fighter jets. This is a huge advantage.

2. AI technology is progressing very rapidly so there will soon be no benefit to having a human in the aircraft. Even today the US military probably has AI that can beat any human in a simulated dogfight.

3. Pilots are very expensive. Not just to train but also in terms of the political cost of having pilots killed or captured by the enemy.

4. Eliminating the pilot opens up a range of options. For example, instead of a $200 million fighter jet it may be more effective to build 10 drones for $20 million each, or even 100 drones for $2 million each. Maybe half the drones will have stealth capabilities and half will have AWCS-like communication and control features.

5. During WWII German and Russian tanks were roughly equivalent. But the Russians could build tanks at a far greater rate than the Germans, making the German loss almost inevitable. Similarly, in future the country that can produce many drones quickly and cheaply may dominate air warfare. In this regard China will likely have an advantage over the US.

6. The military-industrial complex is a huge bureaucracy in which many people have a financial and psychological incentive to continue development of manned fighter jets. For this reason I expect such development to continue well beyond its use-by date, probably until there is a real-world battle where a large, previously dominant air force is soundly beaten by a small power using new drone technology.

3. Pilots are very expensive. Not just to train but also in terms of the political cost of having pilots killed or captured by the enemy.

In terms of ethics, there is a need to ensure that war has a human cost or there is very little to stop countries doing it all the time. If the cost of war is purely financial, then I think perhaps that might be the beginning of the end to the world as we know it.

It's a multi-role fighter aircraft, not simply air superioity. It needs to be able to carry air to surface ordnance

We're not quite in the sci-fi future yet Drone swarms exist, but mostly do intelligence, surveillance, and reconnaissance in the field. The perdix drones that the us military does have are powered by propeller, so so they can't keep up with a fighter jet that can go Mach 1.6 or carry or current supply of air to surface missiles. I believe that the way they would attack a fighter aircraft is turn themselves into Kamikaze missiles.

They're still working the bugs out and trying to perfrct their autonomous behavior...which scares the shit out of me as I write this. This is the only application of AI that causes me fear and anxiety. Thankfully, they don't have any weapons platforms yet

Reply on news.ycombinator.com