An Improved Data Augmentation Scheme for Model Predictive Control Policy Approximation
This work addresses a specific bottleneck in MPC policy approximation for control systems, offering an incremental improvement over prior sensitivity-based methods.
The paper tackles the challenge of efficiently generating training data for approximating Model Predictive Control (MPC) policies by proposing an improved data augmentation scheme based on predictor-corrector steps, which enforces user-defined accuracy and achieves error bounds independent of neighborhood size.
This paper considers the problem of data generation for MPC policy approximation. Learning an approximate MPC policy from expert demonstrations requires a large data set consisting of optimal state-action pairs, sampled across the feasible state space. Yet, the key challenge of efficiently generating the training samples has not been studied widely. Recently, a sensitivity-based data augmentation framework for MPC policy approximation was proposed, where the parametric sensitivities are exploited to cheaply generate several additional samples from a single offline MPC computation. The error due to augmenting the training data set with inexact samples was shown to increase with the size of the neighborhood around each sample used for data augmentation. Building upon this work, this letter paper presents an improved data augmentation scheme based on predictor-corrector steps that enforces a user-defined level of accuracy, and shows that the error bound of the augmented samples are independent of the size of the neighborhood used for data augmentation.