Why a right-half-plane zero feels wrong
A normal system moves toward the direction you ask it to go. A system with a right-half-plane zero can do the opposite first.
Model: H(s) = ωn²(1 ∓ s/z) / (s² + 2ζωn s + ωn²), with − for RHP and + for LHP. DC gain is one. The plot spans eight seconds and may not have settled at slow settings. The pole-zero plane rescales to keep all roots visible.
You ask for more. The output gives you less.
Apply a positive step command. With a right-half-plane zero, the response can initially move negative before turning around toward its final value. That wrong-way motion is an inverse response.
The poles can stay the same while the transient changes dramatically.
Move only the zero from the left half plane to the right half plane. The denominator and DC gain stay the same, but the numerator introduces a component with the wrong initial sign.
Minimum-phase behavior. It usually reinforces the direction of response.
Non-minimum-phase behavior. It adds phase lag and can create inverse response.
The zero becomes a bandwidth limit.
A low-frequency RHP zero places a fundamental restriction on how aggressively the loop can be controlled. The plant is initially trying to move the wrong way while the controller is demanding speed.
The boost converter can make this physical.
In CCM, increasing duty cycle initially keeps the switch on longer, temporarily reducing energy delivered to the output. Only later does the increased inductor current raise the output voltage. That energy-transfer mechanism creates the familiar RHP-zero behavior.