Exponential stability of data-driven nonlinear MPC based on input/output models
This work addresses stability guarantees for data-driven nonlinear MPC, which is incremental as it extends existing MPC frameworks to input-output models without full state knowledge.
The paper tackles the problem of ensuring exponential stability in nonlinear model predictive control (MPC) using data-driven input-output models, establishing stability under conditions like long prediction horizons and proportional error bounds, with verification through kernel interpolation and a numerical example.
We consider nonlinear model predictive control (MPC) schemes using surrogate models in the optimization step based on input-output data only. We establish exponential stability for sufficiently long prediction horizons assuming exponential stabilizability and a proportional error bound. Moreover, we verify the imposed condition on the approximation using kernel interpolation and demonstrate the practical applicability to nonlinear systems with a numerical example.