IEEE Control Systems Letters最新文献_第3页

A Trade-Off Between Time and Energy of Stochastic Higher-Order Networks on Multi-Directed Hypergraphs 多向超图上随机高阶网络的时间与能量权衡

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-12 DOI: 10.1109/LCSYS.2026.3652578

Sasa Chen;Dan Liu;Binrui Wang;Jun Mao;Xiaohang Li

引用次数: 0

Stochastic Coordinate Descent via Regret Minimization 基于遗憾最小化的随机坐标下降

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-12 DOI: 10.1109/LCSYS.2026.3653302

Yankai Lin;Xiaopuwen Wang

{"title":"Stochastic Coordinate Descent via Regret Minimization","authors":"Yankai Lin;Xiaopuwen Wang","doi":"10.1109/LCSYS.2026.3653302","DOIUrl":"https://doi.org/10.1109/LCSYS.2026.3653302","url":null,"abstract":"In this letter, we investigate the performance of stochastic coordinate descent algorithms for convex optimization problems from the novel perspective of regret minimization. Specifically, we consider a stochastic coordinate selection rule that might vary over time, whereas existing results commonly focus on stochastic selection rules that remain fixed. Moreover, unlike most existing results, which assume that the expectation of algorithm updates follows the exact negative gradient direction, our framework is applicable to any stochastic gradient descent (SGD) like algorithms with estimated gradients that satisfy our main assumption, thus making our approach more general than existing results. We show that the updates of the stochastic algorithm to solve a standard convex optimization problem also serve as an online algorithm for an online convex optimization problem that has time-varying cost functions at each time step. We prove that, under relatively weak assumptions, the online algorithm achieves sublinear regret against the solution of the problem, therefore ensuring the average performance of the algorithm within a finite time window. Finally, we verify our theoretical results by solving a linear regression problem using the proposed algorithm.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"9 ","pages":"3155-3160"},"PeriodicalIF":2.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Closing the Loop Inside Neural Networks: Causality-Guided Layer Adaptation for Fault Recovery Control 神经网络内部闭环：故障恢复控制的因果导层自适应

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-12 DOI: 10.1109/LCSYS.2026.3652898

Mahdi Taheri;Soon-Jo Chung;Fred Y. Hadaegh

{"title":"Closing the Loop Inside Neural Networks: Causality-Guided Layer Adaptation for Fault Recovery Control","authors":"Mahdi Taheri;Soon-Jo Chung;Fred Y. Hadaegh","doi":"10.1109/LCSYS.2026.3652898","DOIUrl":"https://doi.org/10.1109/LCSYS.2026.3652898","url":null,"abstract":"This letter studies the problem of real-time fault recovery control for nonlinear control-affine systems subject to actuator loss of effectiveness faults and external disturbances. We develop a two-stage framework that combines causal inference with selective online adaptation to achieve an effective learning-based recovery control method. In the offline phase, we introduce a causal layer attribution technique based on the average causal effect (ACE) to evaluate the relative importance of each layer in a pretrained deep neural network (DNN) controller compensating for faults. This provides a principled approach to select the most causally influential layer for fault recovery control in the sense of ACE, and goes beyond the widely used last-layer adaptation approach. In the online phase, we deploy a Lyapunov-based gradient update to adapt only the ACE-selected layer to circumvent the need for full-network or last-layer only updates. The proposed adaptive controller guarantees uniform ultimate boundedness (UUB) with exponential convergence of the closed-loop system in the presence of actuator faults and external disturbances. Compared to conventional adaptive DNN controllers with full-network adaptation, our methodology has a reduced computational overhead in the online phase. To demonstrate the effectiveness of our proposed methodology, a case study is provided on a 3-axis attitude control system of a spacecraft with four reaction wheels.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"9 ","pages":"3137-3142"},"PeriodicalIF":2.0,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Encoding High-Level Knowledge in Offline Multi-Agent Reinforcement Learning Using Reward Machines 基于奖励机的离线多智能体强化学习高级知识编码

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-06 DOI: 10.1109/LCSYS.2026.3651656

Shayan Meshkat Alsadat;Zhe Xu

引用次数: 0

Cooperative H∞ Consensus Control With Non-Convex Constraints and Parameter Uncertainties 具有非凸约束和参数不确定性的协同H∞一致控制

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-01 Epub Date: 2026-03-26 DOI: 10.1109/LCSYS.2026.3677911

Hongmei Zhang;Xue Li

引用次数: 0

Spiking Control of a Solenoid Valve for High-Precision Pressure Regulation in Soft Robotics 软机器人高精度压力调节电磁阀的峰值控制

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-01 Epub Date: 2026-03-09 DOI: 10.1109/LCSYS.2026.3672133

Andrés J. Serrano-Balbontín;Inés Tejado;Blas M. Vinagre;Sumeet S. Aphale;Andres San-Millan

{"title":"Spiking Control of a Solenoid Valve for High-Precision Pressure Regulation in Soft Robotics","authors":"Andrés J. Serrano-Balbontín;Inés Tejado;Blas M. Vinagre;Sumeet S. Aphale;Andres San-Millan","doi":"10.1109/LCSYS.2026.3672133","DOIUrl":"https://doi.org/10.1109/LCSYS.2026.3672133","url":null,"abstract":"Solenoid valves are widely used for pressure regulation in soft pneumatic robots, yet their intrinsic electromechanical nonlinearities, including dead-zone, saturation, and pressure-dependent dynamics, pose significant challenges for control design. Conventional pulse-modulated strategies, such as pulse-width modulation (PWM), often aggravate these effects by ignoring the transient behaviour associated with valve switching. This work introduces a spiking-based control strategy inspired by neuromorphic principles and specifically tailored for pneumatic pressure regulation in soft actuators. The proposed method uses a neuron-like modulator in cascade with a conventional controller to effectively shape the valve input, intrinsically compensate dead-zone and saturation, and improve the linearity of the pressure dynamics through parameter tuning. A complete experimental validation is conducted on a soft pneumatic actuator (SPA) driven by a solenoid valve, benchmarking the proposed approach against standard pulse-modulation techniques. The results show substantial improvements in both static and dynamic linearity, reduced nonlinear distortions, and enhanced consistency in closed-loop pressure control across operating points.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"10 ","pages":"127-132"},"PeriodicalIF":2.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11426894","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147557647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Robust Direct Data-Driven Ellipsoidal Predictive Control Scheme 一种鲁棒直接数据驱动椭球体预测控制方案

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-01 Epub Date: 2026-03-18 DOI: 10.1109/LCSYS.2026.3675270

Francesco Giannini;Fabrizio Ranieri

{"title":"A Robust Direct Data-Driven Ellipsoidal Predictive Control Scheme","authors":"Francesco Giannini;Fabrizio Ranieri","doi":"10.1109/LCSYS.2026.3675270","DOIUrl":"https://doi.org/10.1109/LCSYS.2026.3675270","url":null,"abstract":"This paper presents a Direct Data-Driven Control strategy for unknown linear systems subject to process noise. The idea is to customize a set-theoretic receding horizon controller within a data driven setting by properly converting Model-based conditions in terms of data series on the available input and state signals. The main results here are 1) a set of Linear Matrix Inequalities conditions that, by using the available data, allows the computation of the inner ellipsoidal approximation of the predecessor set of a given ellipsoidal set; and 2) a Direct Data-Driven Control scheme - named Robust Ellipsoidal Set-Theoretic Data-Enabled Predictive Control - in charge of controlling the plant by means of the previously computed sets. The resulting control strategy guarantees satisfaction of state and input constraints. Moreover, (a) in absence of process noise, the proposed solution is proven to coincide with the Model-based one; whereas (b) in the presence of noise, the proposed solution computes an inner approximation of the controllability sets able to maintain the required guarantees. Numerical simulations show that the proposed control scheme solves the problem of the constrained regulation to an equilibrium of an unknown plant by means of an informative trajectory without identification of the underlined dynamics, despite any realization of the process noise.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"10 ","pages":"163-168"},"PeriodicalIF":2.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147557782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Decentralized Stubborn Observer Redesign With Exact Outlier Cancellation 具有精确离群值对消的分散顽固观测器重新设计

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-01 Epub Date: 2026-06-02 DOI: 10.1109/LCSYS.2026.3699396

Beatrice Zambotti;Vincent Andrieu;Daniele Astolfi;Laurent Bako;Luca Zaccarian

引用次数: 0

High-Order CBF-Compatible Semantics and Hierarchical Synthesis for Bounded-Time STL 有界时间STL的高阶cbf兼容语义和层次综合

IF 2

IEEE Control Systems Letters Pub Date : 2026-01-01 Epub Date: 2026-06-05 DOI: 10.1109/LCSYS.2026.3700689

Chao Tang;Penghong Lu;Zhenpeng Lao;Gang Chen

引用次数: 0

Curvature-Aware Expected Free Energy as an Acquisition Function for Bayesian Optimization 曲率感知的期望自由能作为贝叶斯优化的获取函数