{"id":"arxiv/2603.16808","url":"https://arxiv.org/abs/2603.16808","tags":["math.OC","eess.SY"],"score":0,"title":"Exponential stability of data-driven nonlinear MPC based on input/output models","domain":"math.OC","source":"arxiv","authors":["Lea Bold","Irene Schimperna","Karl Worthmann","Johannes Köhler"],"reviews":0,"subject":"math.OC","summary":"We consider nonlinear model predictive control (MPC) schemes without stabilizing terminal conditions, where the model used in the optimization step is generated 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 by numerical simulations.","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-17","reactions":{}}