Learning-Based Optimal Control with Performance Guarantees for Unknown Systems with Latent States

Abstract

As control engineering methods are applied to increasingly complex systems, data-driven approaches for system identification appear as a promising alternative to physics-based modeling. While the Bayesian approaches prevalent for safety-critical applications usually rely on the availability of state measurements, the states of a complex system are often not directly measurable. It may then be necessary to jointly estimate the dynamics and the latent state, making the quantification of uncertainties and the design of controllers with formal performance guarantees considerably more challenging. This paper proposes a novel method for the computation of an optimal input trajectory for unknown nonlinear systems with latent states based on a combination of particle Markov chain Monte Carlo methods and scenario theory. Probabilistic performance guarantees are derived for the resulting input trajectory, and an approach to validate the performance of arbitrary control laws is presented. The effectiveness of the proposed method is demonstrated in a numerical simulation.

Accepted version submitted to the 2024 European Control Conference (ECC)

References in corpus (6)

• Stable Gaussian Process based Tracking Control of Euler-Lagrange Systems
• Hilbert Space Methods for Reduced-Rank Gaussian Process Regression
• Online learning-based Model Predictive Control with Gaussian Process Models and Stability Guarantees
• Bayesian Inference and Learning in Gaussian Process State-Space Models with Particle MCMC
• Ancestor Sampling for Particle Gibbs
• Distributed Bayesian Online Learning for Cooperative Manipulation