Chinese UCAV/CCA/Loyal Wingman (sensor, A2A and A2G) thread

[Blitzo]

Why is it such a stretch to think that China has CCAs? They’ve been working on drone data linking and AI assistance in aerial combat for years. Watch for satellite images from Dingxin and other airbases in the North West. One of these days you’ll see many UAVs you are not familiar with.

A better question is why do some people still perceive physical evidence as something which is important for relatively low yield things like UAVs or CCAs.

CCAs/UCAVs are one of those things that basically everyone is working on, and given PLA opsec (as well as the seemingly lower amount of interest that CCAs/UCAVs have for people/photographers as well as greater ease of hiding their development), it should be not a great surprise that we have little firm evidence of them.


I would almost say that projects disclosed at Zhuhai like FH-97/A are a red herring because, it creates a false belief or expectation like those represent the pinnacle or highest capability/profile CCA/UCAV that the PRC MIC is developing, whereas in reality it is probably better seen as just a random, unimportant sample of the totality of their efforts.


My advice for people is just to assume that the PLA is developing all of the below, with likely multiple models/airframes per category:
- high end A2A CCA
- medium end A2A CCA
- high end A2G UCAV/CCA
- medium end A2G UCAV/CCA
- high end ISR/AEW/EW UAV/CCA
- medium end ISR/AEW/EW UAV/CCA
- attritible low end CCA
- attritible low end sensor/EW UAV


Some of the projects that we know about (GJ-11, WZ-9/Divine Eagle etc) may be an airframe for one of the above categories, but they should not be viewed as the total sum of their efforts.

 

[Matrixdet]

They are probably not CCA's then, given the slow speed and zero-stealth regardless of the hype. However, they can be formidable autonomous ship-hunters and even more useful SAM/AAM baits if deployed in sufficient numbers. They are large enough to carry enough firepower to sink ships and smart enough to find those ships on their own, so everyone of them needs to taken seriously. And if deployed in large enough numbers concurrently, they will be effective in eroding the fighting capacity of their opponents. So these drones will probably be deployed like shock troops in that they will go in the first wave to weaken the opposition before the manned-unmanned teams.

The unmanned J5 and J6 as missile decoys is a good idea. I saw this concept a long time ago.

 

[tphuang]

https://twitter.com/i/web/status/1926225202012172697

CH-7 project getting publicity on CCTV. Being designed as a high speed reconnaissance aircraft to provide targeting data to fighter aircraft from behind. Now, I tend to think CH-7 is probably not stealthy enough for what PLAAF needs, but who knows. It's interesting that they are advertising this idea.

It makes a lot of sense with what we thought of for CCAs where it can operate in the front and also conduct EW. While J-36 sit behind them and launch missiles.

 

[Tomboy]

https://twitter.com/i/web/status/1926225202012172697

CH-7 project getting publicity on CCTV. Being designed as a high speed reconnaissance aircraft to provide targeting data to fighter aircraft from behind. Now, I tend to think CH-7 is probably not stealthy enough for what PLAAF needs, but who knows. It's interesting that they are advertising this idea.

It makes a lot of sense with what we thought of for CCAs where it can operate in the front and also conduct EW. While J-36 sit behind them and launch missiles.

I always thought CH series is mostly for export and if PLAAF is going to induct it it's gonna have a proper name like WZ-XX or something. I do agree that CH-7 probably isn't stealthy enough for PLAAF frontline duties or ISR over disputed airspace especially after I saw the paper saying that a beaked design is nearly a order of magnitude less stealthy than a pure flying wing. I would've expected a RQ-180 class drone for PLAAF service IMO.

 

[tankphobia]

IMO the lack of public CCA from major militaries is due to cost. If J-8s are still in service there's obviously a cost consideration, i.e there are more suitable pilots than aircraft. if capable CCAs approach manned aircraft in cost then there's little point making an CCA that is less capable than a manned aircraft if pilot is not the constraint. You don't need a CCA to be a SAM bait, the US have been using target drones as decoys for ages, i.e TALD/MALD.

 

[dingyibvs]

They are probably not CCA's then, given the slow speed and zero-stealth regardless of the hype. However, they can be formidable autonomous ship-hunters and even more useful SAM/AAM baits if deployed in sufficient numbers. They are large enough to carry enough firepower to sink ships and smart enough to find those ships on their own, so everyone of them needs to taken seriously. And if deployed in large enough numbers concurrently, they will be effective in eroding the fighting capacity of their opponents. So these drones will probably be deployed like shock troops in that they will go in the first wave to weaken the opposition before the manned-unmanned teams.

To do that you don't really need many of them. You fly out a few, and then for every one you mix in say 10 decoys.

But that's easily countered as well. You just need one stealth aircraft carrying a bunch of small cheap missiles to hunt the whole bunch of them down. Could even be an actual CCA to do that.

IMO currently it's hard to make CCAs work. The best bet is probably a platform that's designed to be optionally manned to begin with. It needs to be modular so that all components needed to support manned flight can be relatively easily swapped out (not necessarily in the field though). This allows the CCA to have the same speed, range, and stealth as a manned fighter while keeping parts commonality high and thus cost relatively low. It would then offer advantages not by being expendable, but by being able to carry more electronics or munitions (say via top weapons bay where the cockpit would be). It's still hard to make it work because besides the technological difficulties it's unclear if those advantages can be offset by not having pilots.

 

[tamsen_ikard]

IMO the lack of public CCA from major militaries is due to cost. If J-8s are still in service there's obviously a cost consideration, i.e there are more suitable pilots than aircraft. if capable CCAs approach manned aircraft in cost then there's little point making an CCA that is less capable than a manned aircraft if pilot is not the constraint. You don't need a CCA to be a SAM bait, the US have been using target drones as decoys for ages, i.e TALD/MALD.

Maybe that's why You have US CCA like Anduril fury which is supposed to be cheap, attritable and likely to be the decoy, scout for a data link missile shot.

 

[mond]

To do that you don't really need many of them. You fly out a few, and then for every one you mix in say 10 decoys.

But that's easily countered as well. You just need one stealth aircraft carrying a bunch of small cheap missiles to hunt the whole bunch of them down. Could even be an actual CCA to do that.

IMO currently it's hard to make CCAs work. The best bet is probably a platform that's designed to be optionally manned to begin with. It needs to be modular so that all components needed to support manned flight can be relatively easily swapped out (not necessarily in the field though). This allows the CCA to have the same speed, range, and stealth as a manned fighter while keeping parts commonality high and thus cost relatively low. It would then offer advantages not by being expendable, but by being able to carry more electronics or munitions (say via top weapons bay where the cockpit would be). It's still hard to make it work because besides the technological difficulties it's unclear if those advantages can be offset by not having pilots.

Being unmanned is already a massive advantage in that it circumvents the bottleneck of pilot training. Countries like the US and China face greater difficulty training pilots than producing planes.

Even if this was not the case, there are strategic benefits to unmanned platforms. As it stands, in peacetime, it takes much longer to train a pilot than it does to produce a fighter. In a wartime scenario, there is no way to shorten pilot training time without sacrificing quality, because learning is not a parallelizable process. On the other hand, manufacturing is highly parallelizable, so turnover can be shortened considerably. This means that unmanned systems can be ramped up in a matter of weeks, while pilot training would have a lag of a year or at least half a year if you sacrifice quality. Such an advantage would have significant impact.

Unmanned systems were not feasible in the past because algorithms and compute capacity were simply not sufficient. It is only in the last ten years where computing has advanced enough for autonomous or semi-autonomous agents to be able to reliably carry out complex missions, and all localized aboard a fighter jet.

So to conclude, it is not difficult to make CCA's work in the sense that they will provide a strategic advantage over purely manned systems. Of course, US programs have aggressively advertised some of the things you mention, like lower cost or attritability/expendibility. One has to consider that these points may be more aimed toward congressional approval than actual strategic necessity.

 

[by78]

Very nice screen caps of CH-7.

 

[dingyibvs]

Being unmanned is already a massive advantage in that it circumvents the bottleneck of pilot training. Countries like the US and China face greater difficulty training pilots than producing planes.

Even if this was not the case, there are strategic benefits to unmanned platforms. As it stands, in peacetime, it takes much longer to train a pilot than it does to produce a fighter. In a wartime scenario, there is no way to shorten pilot training time without sacrificing quality, because learning is not a parallelizable process. On the other hand, manufacturing is highly parallelizable, so turnover can be shortened considerably. This means that unmanned systems can be ramped up in a matter of weeks, while pilot training would have a lag of a year or at least half a year if you sacrifice quality. Such an advantage would have significant impact.

Unmanned systems were not feasible in the past because algorithms and compute capacity were simply not sufficient. It is only in the last ten years where computing has advanced enough for autonomous or semi-autonomous agents to be able to reliably carry out complex missions, and all localized aboard a fighter jet.

So to conclude, it is not difficult to make CCA's work in the sense that they will provide a strategic advantage over purely manned systems. Of course, US programs have aggressively advertised some of the things you mention, like lower cost or attritability/expendibility. One has to consider that these points may be more aimed toward congressional approval than actual strategic necessity.

I do agree with your line of thinking and I've made similar arguments in the past, but some others on the board don't seem to think finding enough pilots is gonna be an issue. I don't really know what it takes to train a pilot so I can't say for sure, but the way things are developing with the J-36 almost like an AWACS with a crew of 2, I'd think that becoming a pilot will become increasingly more difficult and more of a bottleneck than aircraft production.