IEEE Transactions on Automatic Control

最新文章

• Distributionally Robust Model Predictive Control: Closed-Loop Guarantees and Scalable Algorithms AI翻译
  作者：McAllister, Robert D.; Esfahani, Peyman Mohajerin
  作者单位：Delft University of Technology
  摘要：We establish a collection of closed-loop guarantees and propose a scalable optimization algorithm for distributionally robust model predictive control (DRMPC) applied to linear systems, convex constraints, and quadratic costs. Via standard assumptions for the terminal cost and constraint, we establish distributionally robust long-term and stagewise performance guarantees for the closed-loop system. We further demonstrate that a common choice of the terminal cost, i.e., via the discrete-algebra...
• Nonasymptotic and Robust Distributed Algebraic Estimation for Linear Time-Varying Systems AI翻译
  作者：Zhang, Yu-Qing; Liu, Da-Yan; Boutat, Driss; Wu, Ze-Hao
  作者单位：Universite de Orleans; Foshan University
  摘要：This article investigates the nonasymptotic and robust distributed algebraic state estimation for linear time-varying systems using a network of sensors in noisy environments. First, at each sensor node, the system state is transformed into a linear combination of a group of nodes' local observable states, allowing for distributed estimation by estimating a reduced-order state for each node. Second, without requiring initial conditions, the estimation scheme based on generalized modulating fun...
• Distributed Stochastic Projection-Free Algorithm for Constrained Optimization AI翻译
  作者：Jiang, Xia; Zeng, Xianlin; Xie, Lihua; Sun, Jian; Chen, Jie
  作者单位：Chinese University of Hong Kong; Beijing Institute of Technology; Nanyang Technological University
  摘要：This article proposes a distributed stochastic projection-free algorithm for large-scale constrained finite-sum optimization whose constraint set is complicated such that the projection onto the constraint set can be expensive. The global cost function is allocated to multiple agents, each of which computes its local stochastic gradients and communicates with its neighbors to solve the global problem. Stochastic gradient methods enable low computational complexity, while they are hard and slow...
• DGLG: A Novel Deep Generalized Legendre-Galerkin Approach to Optimal Filtering Problem AI翻译
  作者：Shi, Ji; Jiao, Xiaopei; Yau, Stephen S. -T.
  作者单位：Capital Normal University; University of Twente; Tsinghua University
  摘要：The optimal filtering problem for general nonlinear and continuous state-observation systems attracts lots of attention in the control theory. The essence of optimal filtering requires solving the Duncan-Mortensen-Zakai (DMZ) equation in a computationally feasible way. Under the pioneering work of Yau-Yau filtering, the DMZ equation is reduced to a pathwise computation of a forward Kolmogorov equation with time-varying initial conditions, which is very challenging. To overcome the computationa...
• A Fixed-Point Iteration Scheme for Sensitivity-Based Distributed Optimal Control AI翻译
  作者：von Esch, Maximilian Pierer; Volz, Andreas; Graichen, Knut
  作者单位：University of Erlangen Nuremberg
  摘要：This article presents a sensitivity-based algorithm for distributed optimal control problems (OCP) of multi-agent systems with nonlinear dynamics and state/input couplings, as they arise, for instance, in distributed model predictive control. The algorithm relies on first-order sensitivities to cooperatively solve the distributed OCP in parallel. The solutions to the resulting local OCPs are computed with a fixed-point scheme and communicated within one communication step per algorithm iterati...
• Distributed Solvers for Network Linear Equations With Scalarized Compression AI翻译
  作者：Wang, Lei; Ren, Zihao; Yuan, Deming; Shi, Guodong
  作者单位：Zhejiang University; Nanjing University of Science & Technology; University of Sydney
  摘要：Distributed computing is fundamental to multiagent systems, with solving distributed linear equations as a typical example. In this article, we study distributed solvers for network linear equations over a network with node-to-node communication messages compressed as scalar values. Our key idea lies in a dimension compression scheme that includes a dimension-compressing vector and a data unfolding step. The compression vector applies to individual node states as an inner product to generate a...
• Approximation by Simple Poles-Part II: System Level Synthesis Beyond Finite Impulse Response AI翻译
  作者：Fisher, Michael W.; Hug, Gabriela; Dorfler, Florian
  作者单位：University of Waterloo; Swiss Federal Institutes of Technology Domain; ETH Zurich
  摘要：In Part I, a novel Galerkin-type method for finite dimensional approximations of transfer functions in Hardy space was developed based on approximation by simple poles. In Part II, this approximation is applied to system level synthesis, a recent approach based on a clever reparameterization, to develop a new technique for optimal control design. To solve system level synthesis problems, prior work relies on finite impulse response approximations that lead to deadbeat control, and that can exp...
• Nonconvex Optimization Problems for Maximum Hands-Off Control AI翻译
  作者：Ikeda, Takuya
  作者单位：University of Kitakyushu
  摘要：Maximum hands-off control is the optimal solution to the L(0 )optimal control problem. While convex approximation is typically used to relax this problem, it does not necessarily result in maximum hands-off control. Therefore, this study introduces a nonconvex approximation method and a class of nonconvex optimal control problems that are always equivalent to the maximum hands-off control problem. A computation method based on difference of convex functions optimization is then derived and num...
• Expectation-Maximization-Based Remote State Estimation Under State-Dependent Packet Dropouts: From Maximum A Posteriori Perspective AI翻译
  作者：Nie, Yao; Wang, Zidong; Liu, Qinyuan
  作者单位：Tongji University; Tongji University; Shanghai Artificial Intelligence Laboratory; Tongji University; Brunel University
  摘要：This article is concerned with the problem of remote state estimation for linear discrete-time systems with packet dropouts. The packet dropouts are state dependent, which occur only when the real-time target state enters specific observation-denied regions. At the remote estimator, the batch and iterative expectation-maximization (EM) algorithms are proposed for computing the optimal estimation in the maximum a posteriori (MAP) sense. It is shown that the iterative estimator exhibits a struct...
• Convergence and Sample Complexity of Policy Gradient Methods for Stabilizing Linear Systems AI翻译
  作者：Zhao, Feiran; Fu, Xingyun; You, Keyou
  作者单位：Tsinghua University; Tsinghua University
  摘要：System stabilization via policy gradient (PG) methods has drawn increasing attention in both control and machine learning communities. In this article, we study their convergence and sample complexity for stabilizing linear time-invariant systems in terms of the number of system rollouts. Our analysis is built upon a discounted linear quadratic regulator (LQR) method which alternatively updates the policy and the discount factor of the LQR problem. First, we propose an explicit rule to adaptiv...