CAS functionality

[wrightwing]

As an engineer, I marvel at what the control guys have been able to accomplish. Not just with the F-35 CLAW, but with this example of DEATH CLAW demonstrated via the VISTA F-16. But as an engineer, I always worry about "what can go wrong? What am I missing?" I am reminded of that A-320 crash at an airshow in France filled with people for a demonstration flight. As I recall, the pilot commanded go-around thrust, and George basically said, "No, I don't want to, you can't have it" and the plane settled into the forest past the runway. What scares me here is designating the correct target. I have little doubt the CLAW / autopilot can be designed to accurately put rounds on a target at coordinates [X0, Y0, Z0] where the aircraft is moving around at [X(t), Y(t), Z(t)]. My worry is correctly calculating [X0, Y0, Z0] through the "2D" optics of the pilot's helmet.

The auto-GCAS won't allow the "DEATH CLAW" to fly the plane into the ground.

[steve2267]

As an engineer, I marvel at what the control guys have been able to accomplish. Not just with the F-35 CLAW, but with this example of DEATH CLAW demonstrated via the VISTA F-16. But as an engineer, I always worry about "what can go wrong? What am I missing?" I am reminded of that A-320 crash at an airshow in France filled with people for a demonstration flight. As I recall, the pilot commanded go-around thrust, and George basically said, "No, I don't want to, you can't have it" and the plane settled into the forest past the runway. What scares me here is designating the correct target. I have little doubt the CLAW / autopilot can be designed to accurately put rounds on a target at coordinates [X0, Y0, Z0] where the aircraft is moving around at [X(t), Y(t), Z(t)]. My worry is correctly calculating [X0, Y0, Z0] through the "2D" optics of the pilot's helmet.

The auto-GCAS won't allow the "DEATH CLAW" to fly the plane into the ground.

That was not my point at all. And perhaps I chose a poor example, and worded my post poorly. I was trying to state that it is VERY DIFFICULT for an engineer to properly anticipate ALL POSSIBLE scenarios. And when software does not allow the human to respond to a bad situation, bad things can (and probably will) happen.

In the case of an AUTO-STRAFE capability... I could see all sorts of problems if the software does not correctly compute the "aimpoint" in 3D space coordinates, from the 2D projection of targeting symbology onto the real world picture the pilot sees. Getting George to put rounds on those 3D target coordinates is probably easy compared to determining the coordinates in the first place. In a 3D air-to-air scenario, the dynamics of the problem are more difficult, but the targeting determination would appear to be far simpler. FWIW -- mumblings of a lunatic engineer.

[quicksilver]

For most western jets, the 3d piece is fairly easy, and robust. Radar, laser, DTED, gps just for starters...