9
$\begingroup$

I've seen many comments that pitchover and gravity turn are "open loop" and that powered explicit guidance (PEG) leading up to orbital insertion are "closed loop."

And this has me wondering what exactly is meant by open and closed loop.

Because engine nozzle actuators are normally controlled with feedback loops that may themselves receive tracking commands from guidance feedback loops.

This would be the case in PEG flight, where:

  1. Guidance is continuously using navigation data to revise its prediction of the rocket's state vector at orbital insertion to then update the required thrust vector to get there, and
  2. The lower-level thrust-vectoring controllers are continuously monitoring the rocket's deviation from the commanded thrust vector in order to then calculate the required engine nozzle actuator deflections to correct the deviation.

So as you can see, there are two feedback loops, and it's unclear which of them people are referring to when they the rocket is flying "open loop."

For sure I know that in PEG flight both guidance and thrust-vector control must be closed loop.

What about pitchover and gravity turn?

My original guess here was that guidance was open loop (just a preloaded "trajectory" that would give the rocket's attitude as a function of time), while the lower-level thrust vector controller remained closed loop (taking in the real-time deviation of the rocket from the preloaded attitude profile in order to calculate and issue corrective commands to the engine nozzle actuators).

But it's also possible that instead of preloading the rocket's attitude profile, they would preload the engine nozzle actuator commands required to follow that path. In that case, there would be no guidance loop (as no attitude profile would have been loaded and none would be calculated during pitchover/gravity turn), and the lower-level thrust vector controller would be open loop (as it would be issuing preloaded engine nozzle actuator commands instead of calculating those on the fly from the deviations from some desired attitude profile).

So I'm wondering which of these is meant when it's said that a rocket's flying "open loop"?

$\endgroup$
11
$\begingroup$

Your first guess was correct.

But it's also possible that instead of preloading the rocket's attitude profile, they would preload the engine nozzle actuator commands required to follow that path.

This wouldn't work, because there are a number of factors that cause trajectory deviation that can't be predicted. For example, engines can slightly over- or under-perform their rated thrust or gimbal rate specifications, wind can change direction, the density of the atmosphere varies slightly over time, and particularly for multi-engine stages, an engine could fail at any point in the ascent. Therefore, the attitude control is generally closed-loop even when the guidance is open-loop.

The major advantage of open-loop guidance in the early phase of flight is that it makes it easy to constrain the launcher's angle of attack while it's in dense atmosphere. If the attitude profile changes only slowly while it continues to accelerate, the AoA will remain close to zero at all times, which is the classic gravity turn. This minimizes cross-sectional drag and lateral stress on the rocket structure.

To quote Description and Performance of the Saturn Launch Vehicle's Navigation, Guidance, and Control System

During the first stage flight, the vehicle traverses the high aerodynamic pressure region. Structural loads from aerodynamic forces are kept as small as possible by controlling the vehicle with a minimum angle of attack; therefore, in this first stage flight, a standard tilt pro­gram is used and guidance corrections are not intro­duced before the early part of the second stage flight.

I believe "standard tilt program" means a pitch-versus-time program drives the attitude control system.

For attitude stabilization, attitude control signals from the inertial platform stabilized by three orthogonally-arranged single-degree-of-freedom gyros and angular rate signals derived from rate gyroscopes are used; in specific cases, control accelerometers can generate additional attitude control commands.

The Saturn V is a good object of study; a lot of information about it is available on NASA's web pages due to its historic significance, although modern practice may be different.

$\endgroup$
3
  • $\begingroup$ Oh nice, so I've been simulating it right all along, ha ha. Now about that preloaded gravity turn guidance profile... Do they track the angle of attack during launch in order to correct any deviation from zero? Or do they simulate with a mathematical constraint that AoA be near zero then load the simulated attitude profile onto the open-loop guidance system for launch (meaning you no longer need to monitor your angle of attack, you just need to trust your simulated attitude profile keeps it near zero)? $\endgroup$
    – user39728
    Mar 22 at 4:53
  • 1
    $\begingroup$ I think it must vary by launcher quite a bit. I believe Saturn V just used a pitch-versus-time program to control the rocket's attitude, and let the AoA take care of itself. This document on the Saturn V's GNC system may be of some use -- it's probably easier to find detailed info of this sort for the Saturn V than any other launcher, due to its historic significance. $\endgroup$ Mar 22 at 6:47
  • 1
    $\begingroup$ Shuttle worked like this except the independent variable in the table was velocity, not time. Shuttle's alpha and beta had to be within a predefined envelope (a stack of planes indexed by velocity). $\endgroup$ Mar 22 at 21:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy