I am currently working on Model Predictive Control. I somehow want to bound the region in which the optimal solution of the optimization problem will lie in. I though that this could maybe be done based on the previous solution: Under some assumptions, the optimal trajectory will lie within a certain region around the shifted previous trajectory.
Maybe some results on the Lipschitz continuity w.r.t the initial condition or some sensitivity bounds could be helpful.
Unfortunately, I was not succesful in my literature review yet. Are there any well known results in that direction? Ideally, I am looking for results in a scenario where there is also some noise/ model mismatch.
Many thanks for any help in advance.
