Pick-to-Learn Calibration of an MPC Policy for an Origin-to-Destination Flight Problem
Machine Learning
Summary
The authors show how the Pick-to-Learn method can help tune a Model Predictive Control (MPC) system that guides an aircraft through uncertain crosswinds while avoiding risky areas with poor connectivity. They start with many example wind scenarios and use the method to reduce them to just a few key situations that still capture important risks. Their tuned MPC policy successfully avoids those low-connectivity zones across all tested cases and has a mathematically backed small chance of failure when facing new wind conditions. This demonstrates a way to safely control complex systems under uncertainty using data.
Authors
Marco C. Campi, Simone Garatti
Abstract
This paper illustrates the Pick-to-Learn methodology applied to the calibration of a Model Predictive Control policy. While developed around a specific example, the presentation is meant to highlight a methodology of broad applicability. The example concerns an aircraft traveling from an origin point to a destination point in the presence of uncertain crosswinds and a low-connectivity zone that should be avoided. The MPC policy is parameterized by two hyperparameters, which are selected from data by the P2L procedure. Starting from a dataset of 400 wind realizations, also called scenarios, P2L identifies a final compression set containing only two informative scenarios. The resulting MPC policy avoids the low-connectivity zone on all available scenarios and, according to the P2L theory, satisfies a probabilistic risk bound of $4.8\%$ at confidence level $1-10^{-5}$, where the risk is the probability of entering the low-connectivity zone in a future flight under a new wind realization not included in the sample.