Exact Thrust-Reversal Limits of Bidirectional Propellers under Bounded Motor Inputs

What the math gives to ML

The paper identifies an actuator singularity: bidirectional-propeller thrust is quadratic in signed rotor speed, so the inverse thrust map has an unbounded derivative at zero thrust. This structure can transfer to neural controllers by making the thrust head explicitly aware of rotor-speed and motor-input regularity rather than optimizing thrust error alone. A practical adaptation is to parameterize learned reversals through a smooth latent speed coordinate or penalize the motor input implied by the predicted thrust trajectory, reducing current spikes and tracking failures during learned aerial interaction.