Constraint tightening in the Python interface

Hello,

I am new to Acados and want to implement a tube MPC. Unfortunately, I am still very inexperienced with Acados and there are a lot of different constraint types offered here. Which constraint type is suitable for the necessary constraint-tightening and how am I able to update this tube-tightening in every iteration?

I want to implement this first on the base of the 2 given examples of the Python interface:

<Acados_root>/examples/acados_python/getting_started/minimal_example_ocp.py
<Acados_root>/examples/acados_python/race_cars/main.py

Thank you in advance for the help!
Marco

Hi Marco,

I can recommend this repository as an example of MPC with constraint tightening: GitHub - FreyJo/zoro-NMPC-2021: Zero-Order Robust Nonlinear Model Predictive Control with Ellipsoidal Uncertainty Sets
More precisely here: zoro-NMPC-2021/run_tailored_robust_control.py at main · FreyJo/zoro-NMPC-2021 · GitHub

Basically, it is only possible to do it by interacting with the solver in between SQP iterations.

Cheers!

1 Like

Hey, thank you very much for your reply! I cloned the repo and will look into it! It sounds very promising :slight_smile: Thank you for your help!

Cheers
Marco