1 code implementation • 1 May 2024 • Luca Furieri, Clara Lucía Galimberti, Giancarlo Ferrari-Trecate
The growing scale and complexity of safety-critical control systems underscore the need to evolve current control architectures aiming for the unparalleled performances achievable through state-of-the-art optimization and machine learning algorithms.
1 code implementation • 14 Mar 2024 • Andrea Martin, Luca Furieri
The emerging paradigm of learning to optimize (L2O) automates the discovery of algorithms with optimized performance leveraging learning models and data - yet, it lacks a theoretical framework to analyze convergence and robustness of the learned algorithms.
1 code implementation • 23 Nov 2023 • Leonardo Massai, Danilo Saccani, Luca Furieri, Giancarlo Ferrari-Trecate
Further, we can embed prior knowledge about the interconnection topology and stability properties of the system directly into the large-scale distributed operator we design.
1 code implementation • 3 Nov 2023 • Daniele Martinelli, Andrea Martin, Giancarlo Ferrari-Trecate, Luca Furieri
In this work, we focus on the design of optimal controllers that must comply with an information structure.
no code implementations • 26 Jun 2023 • Andrea Martin, Luca Furieri, Florian Dörfler, John Lygeros, Giancarlo Ferrari-Trecate
Towards bridging classical optimal control and online learning, regret minimization has recently been proposed as a control design criterion.
1 code implementation • 28 Apr 2023 • Andrea Martin, Luca Furieri, Florian Dörfler, John Lygeros, Giancarlo Ferrari-Trecate
Specifically, we focus on the problem of designing a feedback controller that minimizes the loss relative to a clairvoyant optimal policy that has foreknowledge of both the system dynamics and the exogenous disturbances.
1 code implementation • 6 Apr 2023 • Daniele Martinelli, Clara Lucía Galimberti, Ian R. Manchester, Luca Furieri, Giancarlo Ferrari-Trecate
We validate the properties of NodeRENs, including the possibility of handling irregularly sampled data, in a case study in nonlinear system identification.
1 code implementation • 14 Nov 2022 • Andrea Martin, Luca Furieri, Florian Dörfler, John Lygeros, Giancarlo Ferrari-Trecate
We consider control of dynamical systems through the lens of competitive analysis.
1 code implementation • 22 Mar 2022 • Luca Furieri, Clara Lucía Galimberti, Giancarlo Ferrari-Trecate
We address the problem of designing stabilizing control policies for nonlinear systems in discrete-time, while minimizing an arbitrary cost function.
1 code implementation • 1 Mar 2022 • Andrea Martin, Luca Furieri, Florian Dörfler, John Lygeros, Giancarlo Ferrari-Trecate
As we move towards safety-critical cyber-physical systems that operate in non-stationary and uncertain environments, it becomes crucial to close the gap between classical optimal control algorithms and adaptive learning-based methods.
1 code implementation • 16 Dec 2021 • Luca Furieri, Clara Lucía Galimberti, Muhammad Zakwan, Giancarlo Ferrari-Trecate
A main challenge of NN controllers is that they are not dependable during and after training, that is, the closed-loop system may be unstable, and the training may fail due to vanishing and exploding gradients.
no code implementations • NeurIPS Workshop DLDE 2021 • Clara Galimberti, Luca Furieri, Liang Xu, Giancarlo Ferrari-Trecate
Deep Neural Networks (DNNs) training can be difficult due to vanishing or exploding gradients during weight optimization through backpropagation.
3 code implementations • 27 May 2021 • Clara Lucía Galimberti, Luca Furieri, Liang Xu, Giancarlo Ferrari-Trecate
Deep Neural Networks (DNNs) training can be difficult due to vanishing and exploding gradients during weight optimization through backpropagation.
no code implementations • 21 May 2021 • Luca Furieri, Baiwei Guo, Andrea Martin, Giancarlo Ferrari-Trecate
As we transition towards the deployment of data-driven controllers for black-box cyberphysical systems, complying with hard safety constraints becomes a primary concern.
no code implementations • 26 Feb 2021 • Luca Furieri, Baiwei Guo, Andrea Martin, Giancarlo Ferrari-Trecate
Recent work in data-driven control has revived behavioral theory to perform a variety of complex control tasks, by directly plugging libraries of past input-output trajectories into optimal control problems.