Contraction theory for nonlinear stability analysis and learning-based control: A tutorial overview

H Tsukamoto, SJ Chung, JJE Slotine - Annual Reviews in Control, 2021 - Elsevier
Contraction theory is an analytical tool to study differential dynamics of a non-autonomous
(ie, time-varying) nonlinear system under a contraction metric defined with a uniformly …

Learning stable Koopman embeddings

F Fan, B Yi, D Rye, G Shi… - 2022 American Control …, 2022 - ieeexplore.ieee.org
In this paper, we present a new data-driven method for learning stable models of nonlinear
systems. Our model lifts the original state space to a higher-dimensional linear manifold …

Stable linear subspace identification: A machine learning approach

L Di Natale, M Zakwan, B Svetozarevic… - 2024 European …, 2024 - ieeexplore.ieee.org
Machine Learning (ML) and linear System Identification (SI) have been historically
developed independently. In this paper, we leverage well-established ML tools—especially …

Stability guarantees for continuous rl control

B Song, JJ Slotine, QC Pham - arxiv preprint arxiv:2209.07324, 2022 - arxiv.org
Lack of stability guarantees strongly limits the use of reinforcement learning (RL) in safety
critical robotic applications. Here we propose a control system architecture for continuous …

Robustness analysis and training of recurrent neural networks using dissipativity theory

P Pauli, J Berberich, F Allgöwer - at-Automatisierungstechnik, 2022 - degruyter.com
Neural networks are widely applied in control applications, yet providing safety guarantees
for neural networks is challenging due to their highly nonlinear nature. We provide a …

Neural-rendezvous: Learning-based robust guidance and control to encounter interstellar objects

H Tsukamoto, SJ Chung, B Donitz, M Ingham… - arxiv preprint arxiv …, 2022 - arxiv.org
Interstellar objects (ISOs), astronomical objects not gravitationally bound to the Sun, are
likely representatives of primitive materials invaluable in understanding exoplanetary star …

Uncertainty learning for lti systems with stability guarantees

F Ghanipoor, C Murguia, PM Esfahani… - 2024 European …, 2024 - ieeexplore.ieee.org
We present a framework for learning of modeling uncertainties in Linear Time Invariant (LTI)
systems to improve the predictive capacity of system models in the input-output sense. First …

Learning first principles systems knowledge from data: Stability and safety with applications to learning from demonstration

A Dani, I Salehi - Artificial Intelligence in Manufacturing, 2024 - Elsevier
The chapter presents two methods of constrained learning of dynamical system models from
data with applications to learning from demonstration (LfD) or imitation learning in the …

Matrix Measure Flows: A Novel Approach to Stable Plasticity in Neural Networks

L Kozachkov, JJ Slotine - arxiv preprint arxiv:2212.12639, 2022 - arxiv.org
This letter introduces the notion of a matrix measure flow as a tool for analyzing the stability
of neural networks with time-varying weights. Given a matrix flow--for example, one induced …

On the improvement of model-predictive controllers

L Féret, A Gepperth, S Lambeck - arxiv preprint arxiv:2308.15157, 2023 - arxiv.org
This article investigates synthetic model-predictive control (MPC) problems to demonstrate
that an increased precision of the internal prediction model (PM) automatially entails an …