-
作者：Hudoba de Badyn, Mathias; Mesbahi, Mehran
作者单位：University of Washington; University of Washington Seattle; Swiss Federal Institutes of Technology Domain; ETH Zurich
摘要：We examine the H-2 norm of matrix-weighted leader-follower consensus on series-parallel networks. By using an extension of electrical network theory on matrix-valued resistances, voltages, and currents, we show that the computation of the H-2 norm can be performed efficiently by decomposing the network into atomic elements and composition rules. Finally, we examine the problem of efficiently adapting the matrix-valued edge weights to optimize the H-2 norm of the network.
-
作者：Chi, Ronghu; Hui, Yu; Huang, Biao; Hou, Zhongsheng
作者单位：Qingdao University of Science & Technology; University of Alberta; Qingdao University
摘要：The design and analysis of active disturbance rejection control (ADRC) are considered for a globally Lipschitz nonlinear discrete-time system, which is nonaffine to control inputs. A local dynamic linearization is proposed to transfer the original nonaffined nonlinear system into a nonlinear system affine to control input such that the open problem of selecting a control gain in the feedback control law of ADRC is solved. Furthermore, both a parameter updating law and an adaptive extended stat...
-
作者：Sharf, Miel; Jain, Anoop; Zelazo, Daniel
作者单位：Royal Institute of Technology; Indian Institute of Technology System (IIT System); Indian Institute of Technology (IIT) - Jodhpur; Technion Israel Institute of Technology
摘要：Equilibrium-independent passive-short (EIPS) systems are a class of systems that satisfy a passivity-like dissipation inequality with respect to any forced equilibria with nonpositive passivity indices. This article presents a geometric approach for finding a passivizing transformation for such systems, relying on their steady-state input-output relation and the notion of projective quadratic inequalities (PQIs). We show that PQIs arise naturally from passivity-shortage characteristics of an E...
-
作者：Gravell, Benjamin; Esfahani, Peyman Mohajerin; Summers, Tyler
作者单位：University of Texas System; University of Texas Dallas; Delft University of Technology
摘要：The linear quadratic regulator (LQR) problem has reemerged as an important theoretical benchmark for reinforcement learning-based control of complex dynamical systems with continuous state and action spaces. In contrast with nearly all recent work in this area, we consider multiplicative noise models, which are increasingly relevant because they explicitly incorporate inherent uncertainty and variation in the system dynamics and thereby improve robustness properties of the controller. Robustne...
-
作者：Malekipirbazari, Milad; Cavus, Ozlem
作者单位：Ihsan Dogramaci Bilkent University
摘要：In classical multiarmed bandit problem, the aim is to find a policy maximizing the expected total reward, implicitly assuming that the decision-maker is risk-neutral. On the other hand, the decision-makers are risk-averse in some real-life applications. In this article, we design a new setting based on the concept of dynamic risk measures where the aim is to find a policy with the best risk-adjusted total discounted outcome. We provide a theoretical analysis of multiarmed bandit problem with r...
-
作者：Subramanian, Venkat Ram; Lamperski, Andrew; Salapaka, Murti, V
作者单位：University of Minnesota System; University of Minnesota Twin Cities
摘要：Complex networked systems can be modeled as graphs with nodes representing the agents and links describing the dynamic coupling between them. Previous work on network identification has shown that the network structure of linear time-invariant (LTI) systems can be reconstructed from the joint power spectrum of the data streams. These results assumed that data are perfectly measured. However, real-world data are subject to many corruptions, such as inaccurate time-stamps, noise, and data loss. ...
-
作者：Li, Jingqi; Chen, Ximing; Pequito, Sergio; Pappas, George J.; Preciado, Victor M.
作者单位：University of California System; University of California Berkeley; University of Pennsylvania; Delft University of Technology
摘要：In this article, we study the target controllability problem of networked dynamical systems,in which we are tasked to steer a subset of network nodes toward a desired objective. More specifically, we derive necessary and sufficient conditions for the structural target controllability of linear time-invariant (LTI) systems with symmetric state matrices, such as those representing undirected dynamical networks with unknown link weights. To achieve our goal, we first characterize the generic rank...
-
作者：Li, Yingying; Qu, Guannan; Li, Na
作者单位：Harvard University; California Institute of Technology
摘要：This article considers online optimization with a finite prediction window of cost functions and additional switching costs on the decisions. We study the fundamental limits of dynamic regret of any online algorithm for both the with-prediction and the no-prediction cases. Besides, we propose two gradient-based online algorithms: receding horizon gradient descent (RHGD) and receding horizon accelerated gradient (RHAG); and provide their regret upper bounds. RHAG's regret upper bound is close t...
-
作者：Liu, Kun-Zhi; Wang, Xue-Fang; Teel, Andrew R.; Sun, Xi-Ming; Liu, Jun
作者单位：Dalian University of Technology; University of California System; University of California Santa Barbara; University of Waterloo
摘要：In this article, stability problems are investigated for hybrid systems with memory, which are developed to model hybrid systems affected by time delays. A nested Matrosov functional theorem is proposed to guarantee uniform asymptotic stability for time-varying hybrid systems with memory. Specifically, with a weak Lyapunov functional whose derivative in the flow set and whose difference in the jump set are negative semidefinite, if there exist some Matrosov functionals that satisfy nested cond...
-
作者：Zhu, Shiyong; Lu, Jianquan; Lou, Yijun; Liu, Yang
作者单位：Southeast University - China; Chengdu University; Linyi University; Hong Kong Polytechnic University; Zhejiang Normal University
摘要：This article considers asymptotic stability and stabilization of Markovian jump Boolean networks (MJBNs) with stochastic state-dependent perturbation. By defining an augmented random variable as the product of the canonical form of switching signal and state variable, asymptotic stability of an MJBN with perturbation is converted into the set stability of a Markov chain (MC). Then, the concept of induced equations is proposed for an MC, and the corresponding criterion is subsequently derived f...
