Comparing Open Loop Methods

Method	What's commanded	Consistent across battery voltage?	Knows the "correct" value?	Typical FRC use
Percent Output	A fraction of whatever voltage is currently available	No	No	Driver-controlled joystick input
Voltage Control	An exact number of volts	Yes	No	Anything needing repeatable behavior run to run
Feedforward	A predicted voltage, computed from a desired velocity and acceleration	Yes	Approximately, if the model is accurate	Flywheels, drivetrain velocity control, arms and elevators

Each row trades away a little more guesswork than the one before it, but none of them ever check whether the result actually matched what was wanted; that's the one thing every open loop method, no matter how good its prediction, structurally cannot do.

When Open Loop Is Enough

Open loop control isn't a stepping stone to something better in every case, it's the right permanent choice whenever a mechanism's behavior is predictable enough that a good model beats the cost and complexity of measuring and correcting. A drivetrain under direct human control barely needs anything more than percent output, since the driver is the feedback loop, watching the robot and adjusting the joystick in real time.

Where It Falls Short

Open loop struggles the moment a mechanism's behavior stops being predictable: a chain that's stretched slightly, a game piece adding unmodeled weight to an arm, a wheel slipping on carpet, a battery sagging further than expected. None of that is visible to an open loop system, since nothing is ever measured after the command goes out. Fixing that requires exactly what open loop control leaves out: a sensor, and a way to react to what it reports, which is the whole subject of Closed Loop Control.