arXiv subjects

Lorenzo Sforni

Publications and source records attributed to Lorenzo Sforni.

5 recordsLinked to original sources

arXiv2026

Stability-Certified On-Policy Data-Driven LQR via Recursive Learning and Policy Gradient

In this paper, we investigate a data-driven framework to solve Linear Quadratic Regulator (LQR) problems when the dynamics is unknown, with the additional challenge of providing stability certificates for the overall learning and control scheme. Specifically, in the proposed on-policy learning framework, the control input is applied to the actual (unknown) linear system while iteratively optimized. We propose a learning and control procedure, termed Relearn LQR, that combines a recursive least squares method with a direct policy search based on the gradient method. The resulting scheme is analyzed by modeling it as a feedback-interconnected nonlinear dynamical system. A Lyapunov-based approach, exploiting averaging and timescale separation theories for nonlinear systems, allows us to provide formal stability guarantees for the whole interconnected scheme. The effectiveness of the proposed strategy is corroborated by numerical simulations, where Relearn LQR is deployed on an aircraft control problem, with both static and drifting parameters.

eess.SY↗

arXiv2025

DATA-DRIVEN PRONTO: a Model-free Solution for Numerical Optimal Control

This article addresses the problem of data-driven numerical optimal control for unknown nonlinear systems. In our scenario, we suppose to have the possibility of performing multiple experiments (or simulations) on the system. Experiments are performed by relying on a data-driven tracking controller able to steer the system towards a desired reference. Our proposed DATA-DRIVEN PRONTO algorithm iteratively refines a tentative solution of the optimal control problem by computing an approximate descent direction via a local trajectory perturbation. At each iteration, multiple trajectories are gathered by perturbing the current trajectory with a suitable dither signal, and then used to obtain a data-driven, time-varying linearization. The exploration is guided by the tracking controller, so that perturbed trajectories are obtained in closed loop. We show local convergence of DATA-DRIVEN PRONTO to a ball about an isolated optimal solution, whose radius depends on the amplitude of the dither signal. We corroborate the theoretical results by applying it to an underactuated robot.

eess.SY↗

arXiv2024

Multi-Robot Target Monitoring and Encirclement via Triggered Distributed Feedback Optimization

We design a distributed feedback optimization strategy, embedded into a modular ROS 2 control architecture, which allows a team of heterogeneous robots to cooperatively monitor and encircle a target while patrolling points of interest. Relying on the aggregative feedback optimization framework, we handle multi-robot dynamics while minimizing a global performance index depending on both microscopic (e.g., the location of single robots) and macroscopic variables (e.g., the spatial distribution of the team). The proposed distributed policy allows the robots to cooperatively address the global problem by employing only local measurements and neighboring data exchanges. These exchanges are performed through an asynchronous communication protocol ruled by locally-verifiable triggering conditions. We formally prove that our strategy steers the robots to a set of configurations representing stationary points of the considered optimization problem. The effectiveness and scalability of the overall strategy are tested via Monte Carlo campaigns of realistic Webots ROS 2 virtual experiments. Finally, the applicability of our solution is shown with real experiments on ground and aerial robots.

cs.RO↗

arXiv2023

GoPRONTO: a Feedback-based Framework for Nonlinear Optimal Control

In this paper we We propose GoPRONTO, a first-order, feedback-based approach to solve nonlinear discrete-time optimal control problems. This method is a generalized first-order framework based on incorporating the original dynamics into a closed-loop system. By exploiting this feedback-based shooting, we are able to reinterpret the optimal control problem as the minimization of a cost function, depending on a state-input curve, whose gradient can be computed by resorting to a suitable costate equation. This convenient reformulation gives room for a collection of accelerated numerical optimal control schemes. To corroborate the theoretical results, numerical simulations on the optimal control of a train of inverted pendulum-on-cart systems are shown.

math.OC↗

arXiv2023

A Distributed Online Optimization Strategy for Cooperative Robotic Surveillance

In this paper, we propose a distributed algorithm to control a team of cooperating robots aiming to protect a target from a set of intruders. Specifically, we model the strategy of the defending team by means of an online optimization problem inspired by the emerging distributed aggregative framework. In particular, each defending robot determines its own position depending on (i) the relative position between an associated intruder and the target, (ii) its contribution to the barycenter of the team, and (iii) collisions to avoid with its teammates. We highlight that each agent is only aware of local, noisy measurements about the location of the associated intruder and the target. Thus, in each robot, our algorithm needs to (i) locally reconstruct global unavailable quantities and (ii) predict its current objective functions starting from the local measurements. The effectiveness of the proposed methodology is corroborated by simulations and experiments on a team of cooperating quadrotors.

cs.RO↗