Learning-based Model Predictive Control for Safe Exploration
This addresses safety concerns for real-world control systems, offering a method to enable safe exploration in dynamic environments, though it builds on existing ideas with incremental improvements.
The paper tackles the problem of providing safety guarantees for learning-based control in safety-critical applications by introducing a learning-based model predictive control scheme with provable high-probability safety guarantees, using Gaussian process priors and terminal set constraints to ensure safe trajectories and exploration.
Learning-based methods have been successful in solving complex control tasks without significant prior knowledge about the system. However, these methods typically do not provide any safety guarantees, which prevents their use in safety-critical, real-world applications. In this paper, we present a learning-based model predictive control scheme that can provide provable high-probability safety guarantees. To this end, we exploit regularity assumptions on the dynamics in terms of a Gaussian process prior to construct provably accurate confidence intervals on predicted trajectories. Unlike previous approaches, we do not assume that model uncertainties are independent. Based on these predictions, we guarantee that trajectories satisfy safety constraints. Moreover, we use a terminal set constraint to recursively guarantee the existence of safe control actions at every iteration. In our experiments, we show that the resulting algorithm can be used to safely and efficiently explore and learn about dynamic systems.