Efficient Automatic Tuning for Data-driven Model Predictive Control via Meta-Learning

TL;DR

This paper introduces Portfolio, a meta-learning approach that enhances AutoMPC's efficiency and stability in data-driven model predictive control tuning by warmstarting Bayesian Optimization with prior configurations.

Contribution

It proposes a novel meta-learning method called Portfolio that improves AutoMPC's tuning process by leveraging prior knowledge to warmstart Bayesian Optimization, reducing computational cost and increasing stability.

Findings

Portfolio outperforms pure Bayesian Optimization in experiments.

It achieves better solutions within limited computational resources.

Effective on multiple nonlinear control benchmarks and a real underwater robot dataset.

Abstract

AutoMPC is a Python package that automates and optimizes data-driven model predictive control. However, it can be computationally expensive and unstable when exploring large search spaces using pure Bayesian Optimization (BO). To address these issues, this paper proposes to employ a meta-learning approach called Portfolio that improves AutoMPC's efficiency and stability by warmstarting BO. Portfolio optimizes initial designs for BO using a diverse set of configurations from previous tasks and stabilizes the tuning process by fixing initial configurations instead of selecting them randomly. Experimental results demonstrate that Portfolio outperforms the pure BO in finding desirable solutions for AutoMPC within limited computational resources on 11 nonlinear control simulation benchmarks and 1 physical underwater soft robot dataset.