2024

最新文章

• Probabilistic Constraint for Safety-Critical Reinforcement Learning AI翻译
  作者：Chen, Weiqin; Subramanian, Dharmashankar; Paternain, Santiago
  作者单位：International Business Machines (IBM); IBM USA
  摘要：In this article, we consider the problem of learning safe policies for probabilistic-constrained reinforcement learning (RL). Specifically, a safe policy or controller is one that, with high probability, maintains the trajectory of the agent in a given safe set. We establish a connection between this probabilistic-constrained setting and the cumulative-constrained formulation that is frequently explored in the existing literature. We provide theoretical bounds elucidating that the probabilisti...
• Privacy-Preserving Average Consensus Through Network Augmentation AI翻译
  作者：Ramos, Guilherme; Aguiar, Antonio Pedro; Kar, Soummya; Pequito, Sergio
  作者单位：Universidade de Lisboa; Instituto de Telecomunicacoes; Universidade do Porto; Carnegie Mellon University; Uppsala University
  摘要：Average consensus protocols play a central role in distributed systems and decision-making, such as distributed information fusion, distributed optimization, distributed estimation, and control. A key advantage of these protocols is that agents exchange and reveal their state information only to their neighbors. In its basic form, the goal of average consensus protocols is to compute an aggregate such as the average of network data; however, existing protocols could lead to leakage of individu...
• Noninterference Analysis of Bounded Petri Nets Using Basis Reachability Graph AI翻译
  作者：Ran, Ning; Nie, Jingyao; Meng, Aiwen; Seatzu, Carla
  作者单位：Hebei University; University of Cagliari
  摘要：In this article, we deal with two problems related to security and privacy of bounded Petri nets, namely, noninterference analysis and enforcement. A system could be monitored by different types of users, high-level and low-level users, who have access to different information even if both know the structure of the system. Low-level users can observe only the occurrence of a subset of events. On the contrary, high-level users can observe the occurrence of all the events affecting the system dy...
• Robust Variable Projection Algorithm for the Identification of Separable Nonlinear Models AI翻译
  作者：Chen, Guang-Yong; Su, Xiang-Xiang; Gan, Min; Guo, Wenzhong; Chen, C. L. Philip
  作者单位：Fuzhou University; South China University of Technology
  摘要：Robust nonlinear regression frequently arises in data analysis that is affected by outliers in various application fields such as system identification, signal processing, and machine learning. However, it is still quite challenge to design an efficient algorithm for such problems due to the nonlinearity and nonsmoothness. Previous researches usually ignore the underlying structure presenting in the such nonlinear regression models, where the variables can be partitioned into a linear part and...
• Solving Feynman-Kac Forward-Backward SDEs Using McKean-Markov Branched Sampling AI翻译
  作者：Hawkins, Kelsey P.; Pakniyat, Ali; Theodorou, Evangelos; Tsiotras, Panagiotis
  作者单位：University System of Georgia; Georgia Institute of Technology; University of Alabama System; University of Alabama Tuscaloosa; University System of Georgia; Georgia Institute of Technology
  摘要：We propose a new method for the numerical solution of the forward-backward stochastic differential equations (FBSDE) appearing in the Feynman-Kac representation of the value function in stochastic optimal control problems. Using Girsanov's change of probability measures, it is demonstrated how a McKean-Markov branched sampling method can be utilized for the forward integration pass, as long as the controlled drift term is appropriately compensated in the backward integration pass. Subsequently...
• A Low-Power Multi-High-Gain Observer Design for a Class of Nonlinear Systems AI翻译
  作者：Mousavi, S.; Guay, M.
  作者单位：Queens University - Canada
  摘要：In this article, a low-power multi-high-gain observer (low-power MHGO) is proposed, where multiple low-power observers of order 2n-2 are used to improve the transient response of HGOs, as well as reducing their sensitivity to high-frequency measurement noise. The MHGO methodology is applied to low-power observers to aggregate the advantages of the traditional HGO, low-power HGO, and MHGO. It is shown that there exists a combination of the unknown parameters for which the weighted estimated sta...
• Homogeneous Control Design Using Invariant Ellipsoid Method AI翻译
  作者：Wang, Siyuan; Duan, Haibin; Zheng, Gang; Ping, Xubin; Boutat, Driss; Polyakov, Andrey
  作者单位：Beihang University; Beihang University; Xidian University
  摘要：The invariant ellipsoid method is aimed at minimization of the smallest invariant and attractive set of a control system operating under bounded external disturbances and parametric uncertainties. This article extends this technique to a class of the so-called generalized homogeneous system. The generalized homogeneous optimal (in the sense of invariant ellipsoid) controller allows further improvement of the control system providing a faster convergence and smaller overshoots. Theoretical resu...
• Output Discernibility of Topological Variations in Linear Dynamical Networks AI翻译
  作者：Fan, Ziye; Wu, Xiaoqun; Mao, Bing; Lu, Jinhu
  作者单位：Wuhan University; Shenzhen University; Beihang University; Zhongguancun Laboratory
  摘要：The conditions under which topological variations in networked linear dynamical systems can be discerned from their outputs are investigated. The output-indiscernible space is completely characterized without conditions imposed on the topology matrix. It is demonstrated that a topological change can be output-indiscernible even if the original and the altered topology matrices share no common eigenvalues. Furthermore, the necessary and sufficient condition for output discernibility is proposed...
• Open-Loop Chance Constrained Stochastic Optimal Control via the One-Sided Vysochanskij-Petunin Inequality AI翻译
  作者：Pacula, Isabella; Oishi, Meeko
  作者单位：University of New Mexico
  摘要：While many techniques have been developed for chance constrained stochastic optimal control with Gaussian disturbance processes, far less is known about computationally efficient methods to handle non-Gaussian processes. In this article, we develop a method for solving chance constrained stochastic optimal control problems for linear time-invariant systems with general additive disturbances with finite moments and unimodal chance constraints. We propose an open-loop control scheme for multiveh...
• Networked Control for Nonlinear Plants Subject to Updating Disorder AI翻译
  作者：Yu, Hao; Chen, Tongwen
  作者单位：Beijing Institute of Technology; University of Alberta
  摘要：This article addresses a networked control problem for nonlinear plants that are subject to several network-induced issues simultaneously, including multiple independent communication channels, time-varying transmission intervals, sensor schedule protocols, large transmission delays, and updating disorder. A new indicator, the maximum number in updating disorder, is proposed to evaluate the intensity of updating disorder. Then, to describe the complicated relationship between transmission and ...