antoineleeman/robust-nonlinear-mpc ? reverse-engineered prompt

I want this repo turned into a smooth, runnable Python project for the robust nonlinear control demos. Please get the pendulum, quadrotor, and especially the rocket example working end to end so I can set up an environment, generate whatever solver files are needed for the chosen system, run a closed loop experiment, and then view the latest saved plots without guessing what command comes next.

Please make the workflow simple and hard to misuse. If a solver has to be regenerated for a different experiment, add clear checks or warnings so I do not accidentally run the wrong one. If an extra solver like IPOPT is needed, tell me exactly what I need to install and when. I would also like a clean README update with the shortest possible quick start for each demo, plus one obvious command for the rocket demo that reproduces the main simulation and plotting shown in the repo.

If anything is unclear, look up current docs online and make reasonable choices.