Choose Wisely: Data-Enabled Predictive Control for Nonlinear Systems Using Online Data Selection

Joshua Näf, Keith Moffat, Jaap Eising, Florian Dörfler

Abstract: This paper proposes Select-Data-Enabled Predictive Control (Select-DeePC), a new method for controlling nonlinear systems using output-feedback for which data are available but an explicit model is not. At each timestep, Select-DeePC employs only the most relevant data to implicitly linearize the dynamics in ``trajectory space.'' Then, taking user-defined output constraints into account, it makes control decisions using a convex optimization. This optimal control is applied in a receding-horizon manner. As the online data-selection is the core of Select-DeePC, we propose and verify both norm-based and manifold-embedding-based selection methods. We evaluate Select-DeePC on three benchmark nonlinear system simulators -- rocket-landing, a robotic arm and cart-pole inverted pendulum swing-up -- comparing them with standard DeePC and Time-Windowed DeePC, and find that Select-DeePC outperforms both methods.