arXiv subjects

Fethi Bencherki

Publications and source records attributed to Fethi Bencherki.

8 recordsLinked to original sources

arXiv2026

Optimal input design via Frank-Wolfe

We study optimal input design over a finite horizon for linear dynamical systems. The goal is to minimize a weighted inverse-covariance (information) criterion subject to an energy budget. The set of covariances achievable by causal policies is convex but lacks a tractable explicit description, ruling out projection-based methods. We show that Frank--Wolfe applies naturally: each linear minimization subproblem is a budget-constrained finite-horizon linear quadratic (LQ) problem, solvable by a Riccati recursion and one-dimensional bisection over a Lagrange multiplier. Using smoothness of the objective over the feasible set, we establish an $\mathcal{O}(1/M)$ convergence rate for the objective value, while strong convexity yields an $\mathcal{O}(1/\sqrt{M})$ rate for the iterates. We further extend the framework to input design for system identification with unknown dynamics and adaptive online LQR, and illustrate the approach numerically.

math.OC↗

arXiv2026

Minimax optimal dual control of positive systems: an exact solution for scalar input-sign uncertainty

While recent advances in minimax dual control have led to exact solutions for uncertain general linear time-invariant systems as well as (sub)optimal dual controllers, corresponding results for linear positive systems are still lacking. This paper aims to fill this gap and thereby pave the way toward scalable dual control algorithms. We study the general minimax optimal dual control problem for positive linear systems with unknown dynamics and reformulate it as a standard zero-sum dynamic game. By allowing randomized control inputs, we solve the corresponding Bellman equation exactly for the scalar case with sign uncertainty in the input. This yields an implicit dual control policy that is optimal both in terms of cost and $\ell_1$-gain. The optimal dual policy uses exploration in a specific region of the hyperstate space to conduct optimal probing. Outside this exploration regime, the controller reduces to a deterministic certainty equivalence policy, indicating that sufficient information has been obtained to identify the correct input direction. In addition, these results allow us to analyze fundamental limitations of minimax dual control for positive systems and provide a foundation for more general dual control problems for positive systems for future work.

math.OC↗

arXiv2026

Minimax adaptive control for finite sets of positive linear systems

We present a minimax adaptive control framework for discrete-time positive linear systems with parametric uncertainty and adversarial disturbances. The uncertainty in the system dynamics is assumed to lie in a finite set of possible plants. We formulate the problem as a dynamic game between the controller, which minimizes the cost, and an adversary, which selects both the disturbances and the plant dynamics to maximize the cost. An equivalent reformulation of the original game transforms the problem into a standard minimax two-player zero-sum dynamic game. This enables the problem to be addressed via minimax dynamic programming. We provide an explicit solution to the Bellman inequality, yielding stabilizing, positivity preserving policies without requiring an initially stabilizing controller. The resulting controller enjoys robustness guarantees in the form of bounded $\ell_1$-gain from disturbances to errors. Once the uncertain parameters have been sufficiently estimated, the controller behaves like a standard $\mathcal H_\infty$-type optimal controller for positive linear systems. The theoretical findings are supported by numerical experiments illustrating the resulting adaptive controller in action.

math.OC↗

arXiv2024

Adaptive Control of Positive Systems with Application to Learning SSP

An adaptive controller is proposed and analyzed for the class of infinite-horizon optimal control problems in positive linear systems presented in (Ohlin et al., 2024b). This controller is derived from the solution of a "data-driven algebraic equation" constructed using the model-free Bellman equation from Q-learning. The equation is driven by data correlation matrices that do not scale with the number of data points, enabling efficient online implementation. Consequently, a sufficient condition guaranteeing stability and robustness to unmodeled dynamics is established. The derived results also provide a quantitative characterization of the interplay between excitation level and robustness to unmodeled dynamics. The class of optimal control problems considered here is equivalent to Stochastic Shortest Path (SSP) problems, allowing for a performance comparison between the proposed adaptive policy and model-free algorithms for learning the stochastic shortest path, as demonstrated in the numerical experiment.

math.OC↗

arXiv2024

Achieving consensus in networks of increasingly stubborn voters

We study opinion evolution in networks of stubborn agents discussing a sequence of issues, modeled through the so called concatenated Friedkin-Johnsen (FJ) model. It is concatenated in the sense that agents' opinions evolve for each issue, and the final opinion is then taken as a starting point for the next issue. We consider the scenario where agents {also take a vote at the end of each issue} and propose a feedback mechanism from the result (based on the median voter) to the agents' stubbornness. Specifically, agents become increasingly stubborn during issue $s+1$ the more they disagree with the vote at the end of issue $s$. We analyze {this model} for a number of special cases and provide sufficient conditions for convergence to consensus stated in terms of permissible initial opinion and stubbornness. In the opposite scenario, where agents become less stubborn when disagreeing with the vote result, we prove that consensus is achieved{, and we demonstrate the faster convergence of opinions compared to constant stubbornness.

math.OC↗

arXiv2021

Observer-based switched-linear system identification

In this paper, we present a methodology to identify discrete-time state-space switched linear systems (SLSs) from input-output measurements. Continuous-state is not assumed to be measured. The key step is a deadbeat observer based transformation to a switched auto-regressive with exogenous input (SARX) model. This transformation reduces the state-space identification problem to a SARX model estimation problem. Overfitting issues are tackled. The switch and parameter identifiability and the persistence of excitation conditions on the inputs are discussed in detail. The discrete-states are identified in the observer domain by solving a non-convex sparse optimization problem. A clustering algorithm reveals the discrete-states under mild assumptions on the system structure and the dwell times. The switching sequence is estimated from the input-output data by the multi-variable output error state space (MOESP) algorithm and a variant modified from it. A convex relaxation of the sparse optimization problem yields the block basis pursuit denoising (BBPDN) algorithm. Theoretical findings are supported by means of a detailed numerical example. In this example, the proposed methodology is also compared to another identification scheme in hybrid systems literature.

eess.SY↗

arXiv2021

Basis transform in switched linear system state-space models from input-output data

This paper addresses the problem of basis correction in the context of LSS identification from input-output data. It is often the case that identification algorithms for the LSSs from input-output data operate locally. The individually identified local submodel estimates reside in distinct state bases, which mandates performing a basis correction that facilitates their coherent patching for the ultimate goal of performing output predictions for arbitrary inputs and switching sequences. We formulate a persistence of excitation condition for the inputs and the switching sequences that guarantee the presented approach's success. These conditions are mild in nature, which proves the practicality of the devised algorithm. We supplement the theoretical findings with an elaborating numerical simulation example.

eess.SY↗

arXiv2021

Realization of multi-input/multi-output switched linear systems from Markov parameters

This paper presents a four-stage algorithm for the realization of multi-input/multi-output (MIMO) switched linear systems (SLSs) from Markov parameters. In the first stage, a linear time-varying (LTV) realization that is topologically equivalent to the true SLS is derived from the Markov parameters assuming that the submodels have a common MacMillan degree and a mild condition on their dwell times holds. In the second stage, zero sets of LTV Hankel matrices where the realized system has a linear time-invariant (LTI) pulse response matching that of the original SLS are exploited to extract the submodels, up to arbitrary similarity transformations, by a clustering algorithm using a statistics that is invariant to similarity transformations. Recovery is shown to be complete if the dwell times are sufficiently long and some mild identifiability conditions are met. In the third stage, the switching sequence is estimated by three schemes. The first scheme is based on forward/backward corrections and works on the short segments. The second scheme matches Markov parameter estimates to the true parameters for LTV systems and works on the medium-to-long segments. The third scheme also matches Markov parameters, but for LTI systems only and works on the very short segments. In the fourth stage, the submodels estimated in Stage 2 are brought to a common basis by applying a novel basis transformation method which is necessary before performing output predictions to given inputs. A numerical example illustrates the properties of the realization algorithm. A key role in this algorithm is played by time-dependent switching sequences that partition the state-space according to time, unlike many other works in the literature in which partitioning is state and/or input dependent.

eess.SY↗