Search

Microgrids (MG) are exposed to voltage quality deterioration due to the presence of voltage unbalance. To deal with this problem, existing solutions based on Distributed Generation (DG) units interfaced by power electronics offer two type of strategies for voltage unbalance compensation depending on whether the compensation is performed at one remote node or at multiple local nodes. The first type is limited to a single node, and the second type is limited to apply at the DG units output (locally). This paper presents a multi nodal control scheme where DGs can compensate for voltage unbalance at multiple remote nodes of the MG, thus overcoming both state-of-the-art strategies limitations. In particular, negative-sequence voltage is eliminated at as many remote nodes as the number of available DG's. A systematic approach for the multiple-input/multiple-output (MIMO) nature of the problem is presented covering three aspects. First, a square MIMO control strategy is established and a feasibility test is derived to assess whether the problem can be solved. Second, the cross-coupling interaction between the multiple controllers is minimized by optimally selecting which DGs will contribute to mitigate the remote unbalances. Third, stability and transient dynamics are analyzed. Laboratory experimental results corroborate the control performance., This work was supported by I+D+i PID2021-122835OB-C21 research project, financed by MCIN/AEI/10.13039/501000011033 and FEDER “Una manera de hacer Europa., Peer Reviewed, Postprint (author's final draft)

In this paper, we propose two new algorithms for maximum-likelihood estimation (MLE) of high dimensional sparse covariance matrices. Unlike most of the state-of-the-art methods, which either use regularization techniques or penalize the likelihood to impose sparsity, we solve the MLE problem based on an estimated covariance graph. More specifically, we propose a two-stage procedure: in the first stage, we determine the sparsity pattern of the target covariance matrix (in other words the marginal independence in the covariance graph under a Gaussian graphical model) using the multiple hypothesis testing method of false discovery rate (FDR), and in the second stage we use either a block coordinate descent approach to estimate the non-zero values or a proximal distance approach that penalizes the distance between the estimated covariance graph and the target covariance matrix. Doing so gives rise to two different methods, each with its own advantage: the coordinate descent approach does not require tuning of any hyper-parameters, whereas the proximal distance approach is computationally fast but requires a careful tuning of the penalty parameter. Both methods are effective even in cases where the number of observed samples is less than the dimension of the data. For performance evaluation, we test the proposed methods on both simulated and real-world data and show that they provide more accurate estimates of the sparse covariance matrix than the state-of-the-art methods.

In this article, we consider the periodic event-triggered consensus problem for single-integrator multiagent systems. In existing approaches, consensus is typically achieved by trigger schemes enforcing a monotone decrease of a Lyapunov function. Such trigger schemes may, however, result in more transmissions than are actually required to ensure consensus or meet certain performance specifications. This is because a monotone decrease may be a rather conservative condition for a given Lyapunov function in an event-triggered setting. To overcome the conservativity, we propose a novel window-based trigger scheme, which allows us to leverage existing trigger schemes from the literature to derive less conservative ones. This is achieved by taking the past system behavior into account and allowing a temporary increase of the Lyapunov function as long as a decreasing tendency is still guaranteed. For that, information from previous time steps within certain time windows is considered. We provide an explicit bound on the evolution of the Lyapunov function that is the same for the window-based scheme and the (monotone) original trigger schemes. To illustrate the benefits of the window-based scheme, we validate its efficacy through a comprehensive simulation study and demonstrate that the scheme typically reduces the average update rate of the underlying communication structure in comparison to the original trigger schemes., QC 20240502

Suitable representations of dynamical systems can simplify their analysis and control. On this line of thought, this article aims to answer the following question: Can a transformation of the generalized coordinates under which the actuators directly perform work on a subset of the configuration variables be found? We not only show that the answer to this question is yes but also provide necessary and sufficient conditions. More specifically, we look for a representation of the configuration space such that the right-hand side of the dynamics in Euler-Lagrange form becomes [\boldsymbol{I}\; \boldsymbol{O}]{T}\boldsymbol{u}, being \boldsymbol{u} the system input. We identify a class of systems, called collocated, for which this problem is solvable. Under mild conditions on the input matrix, a simple test is presented to verify whether a system is collocated or not. By exploiting power invariance, we provide necessary and sufficient conditions that a change of coordinates decouples the input channels if and only if the dynamics is collocated. In addition, we use the collocated form to derive novel controllers for damped underactuated mechanical systems. To demonstrate the theoretical findings, we consider several Lagrangian systems with a focus on continuum soft robots., Learning & Autonomous Control

Reconfigurable intelligent surfaces (RISs) allow controlling the propagation environment in wireless networks by tuning multiple reflecting elements. RISs have been traditionally realized through single connected architectures, mathematically characterized by a diagonal scattering matrix. Recently, beyond diagonal RISs (BD-RISs) have been proposed as a novel branch of RISs whose scattering matrix is not limited to be diagonal, which creates new benefits and opportunities for RISs. Efficient BD-RIS architectures have been realized based on group and fully connected reconfigurable impedance networks. However, a closed-form solution for the global optimal scattering matrix of these architectures is not yet available. In this paper, we provide such a closed-form solution proving that the theoretical performance upper bounds can be exactly achieved for any channel realization. We first consider the received signal power maximization in single-user single-input single-output (SISO) systems aided by a BD-RIS working in reflective or transmissive mode. Then, we extend our solution to single-user multiple-input multiple-output (MIMO) and multi-user multiple-input single-output (MISO) systems. We show that our algorithm is less complex than the iterative optimization algorithms employed in the previous literature. The complexity of our algorithm grows linearly (resp. cubically) with the number of RIS elements in the case of group (resp. fully) connected architectures.

Although eigenvectors belong to the core of linear algebra, relatively few closed-form expressions exist, which we bundle and discuss here. A particular goal is their interpretation for graph-related matrices, such as the adjacency matrix of an undirected, possibly weighted graph., Network Architectures and Services

This letter deals with the problem of state estimation for a class of systems involving linear dynamics with multiple quadratic output measurements. We propose a systematic approach to immerse the original system into a linear time-varying (LTV) system of a higher dimension. The methodology extends the original system by incorporating a minimum number of auxiliary states, ensuring that the resulting extended system exhibits both linear dynamics and linear output. Consequently, any Kalman-type observer can showcase global state estimation, provided the system is uniformly observable., QC 20240213

This paper proposes a novel image inpainting method to remove undesired objects in an image. Conventionally, missing regions are filled in using similar textures in an image as exemplars. However, unnatural textures are often generated due to the paucity of available samples. In this study, we take into account symmetric transformation of texture patterns to increase exemplars. To generate plausible textures in missing regions with variously transformed patterns, we employ two approaches: (1) we use spatial coherence of texture patterns when searching for similar patterns, and (2) we define a new degree of confidence of exemplars for determining pixel values. The effectiveness of the proposed method is demonstrated by comparing results by three methods., ICPR 2012 : 21th International Conference on Pattern Recognition , Nov 11-15, 2012 , Tsukuba, Japan

This letter proposes a generalised extended nested array with multiple subarrays (GENAMS) array via the maximum inter-element spacing (IES) constraint principle. Based on the IES set patterns of the two-sides extended nested array and the flexible extended nested array with multiple subarrays type-2, a generalised IES set pattern is derived. The generalised IES set is used to design the GENAMS array, which consists of six uniform linear arrays (ULAs). Compared to other arrays, the GENAMS have improved degrees of freedom (DOFs) and relatively less mutual coupling. Also, the achievable DOFs and the weight functions of the GENAMS array are derived and analysed in detail. Finally, numerical examples are used to show the merits of the GENAMS array.

This article investigates the design of online stealthy attacks with the aim of moving the system's state to the desired target. Different from the design of offline attacks, which is only based on the system's model, to design the online attack, the attacker also estimates the system's state with the intercepted data at each instant and computes the optimal attack accordingly. To ensure stealthiness, the Kullback-Leibler divergence between the innovations with and without attacks at each instant should be smaller than a threshold. We show that the attacker should solve a convex optimization problem at each instant to compute the mean and covariance of the attack. The feasibility of the attack policy is also discussed. Furthermore, for the strictly stealthy case with zero threshold, the analytical expression of the unique optimal attack is given. Finally, a numerical example of the longitudinal flight control system is adopted to illustrate the effectiveness of the proposed attack.

Fractional-order dynamical networks are increasingly being used to model and describe processes demonstrating long-term memory or complex interlaced dependencies among the spatial and temporal components of a wide variety of dynamical networks. Notable examples include networked control systems or neurophysiological networks which are created using electroencephalographic (EEG) or blood-oxygen-level-dependent data. As a result, the estimation of the states of fractional-order dynamical networks poses an important problem. To this effect, this article addresses the problem of minimum-energy state estimation for discrete-time fractional-order dynamical networks, where the state and output equations are affected by an additive noise that is considered to be deterministic, bounded, and unknown. Specifically, we derive the corresponding estimator and show that the resulting estimation error is exponentially input-to-state stable with respect to the disturbances and to a signal that is decreasing with the increase of the accuracy of the adopted approximation model. An illustrative example shows the effectiveness of the proposed method on real-world neurophysiological networks. Our results may significantly contribute to the development of novel neurotechnologies, particularly in the development of state estimation paradigms for neural signals such as EEG, which are often noisy signals known to be affected by artifacts not having any particular stochastic characterization.

In this article, we consider distributed nonconvex optimization over an undirected connected network. Each agent can only access to its own local nonconvex cost function and all agents collaborate to minimize the sum of these functions by using local information exchange. We first propose a modified alternating direction method of multipliers (ADMM) algorithm. We show that the proposed algorithm converges to a stationary point with the sublinear rate O(1/T) if each local cost function is smooth and the algorithm parameters are chosen appropriately. We also show that the proposed algorithm linearly converges to a global optimum under an additional condition that the global cost function satisfies the Polyak-Łojasiewicz condition, which is weaker than the commonly used conditions for showing linear convergence rates including strong convexity. We then propose a distributed linearized ADMM (L-ADMM) algorithm, derived from the modified ADMM algorithm, by linearizing the local cost function at each iteration. We show that the L-ADMM algorithm has the same convergence properties as the modified ADMM algorithm under the same conditions. Numerical simulations are included to verify the correctness and efficiency of the proposed algorithms., QC 20230412

This letter proposes a generalised extended nested array with multiple subarrays (GENAMS) array via the maximum inter-element spacing (IES) constraint principle. Based on the IES set patterns of the two-sides extended nested array and the flexible extended nested array with multiple subarrays type-2, a generalised IES set pattern is derived. The generalised IES set is used to design the GENAMS array, which consists of six uniform linear arrays (ULAs). Compared to other arrays, the GENAMS have improved degrees of freedom (DOFs) and relatively less mutual coupling. Also, the achievable DOFs and the weight functions of the GENAMS array are derived and analysed in detail. Finally, numerical examples are used to show the merits of the GENAMS array.

In this article, we treat extended balancing for continuous-time linear time-invariant systems. We take a dissipativity perspective, thus, resulting in a characterization in terms of linear matrix inequalities. This perspective is useful for determining a priori error bounds. In addition, we address the problem of structure-preserving model reduction of the subclass of port-Hamiltonian systems. We establish sufficient conditions to ensure that the reduced-order model preserves a port-Hamiltonian structure. Moreover, we show that the use of extended Gramians can be exploited to get a small error bound and, possibly, to preserve a physical interpretation for the reduced-order model. We illustrate the results with a large-scale mechanical system example. Furthermore, we show how to interpret a reduced-order model of an electrical circuit again as a lower dimensional electrical circuit., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Learning & Autonomous Control

We propose estimating the scale parameter (mean of the eigenvalues) of the scatter matrix of an unspecified elliptically symmetric distribution using weights obtained by solving Tyler's M-estimator of the scatter matrix. The proposed Tyler's weights based estimate (TWE) of scale is then used to construct an affine equivariant Tyler's M-estimator as a weighted sample covariance matrix using normalized Tyler's weights. We then develop a unified framework for estimating the unknown tail parameter of the elliptical distribution (such as the degrees of freedom (d.o.f.) ? of the multivariate t (MVT) distribution). Using the proposed TWE of scale, a new robust estimate of the d.o.f. parameter of MVT distribution is proposed with excellent performance in heavy-tailed scenarios, outperforming other competing methods. R-package is available that implements the proposed method.

Fractional-order dynamical networks are increasingly being used to model and describe processes demonstrating long-term memory or complex interlaced dependencies among the spatial and temporal components of a wide variety of dynamical networks. Notable examples include networked control systems or neurophysiological networks which are created using electroencephalographic (EEG) or blood-oxygen-level-dependent data. As a result, the estimation of the states of fractional-order dynamical networks poses an important problem. To this effect, this article addresses the problem of minimum-energy state estimation for discrete-time fractional-order dynamical networks, where the state and output equations are affected by an additive noise that is considered to be deterministic, bounded, and unknown. Specifically, we derive the corresponding estimator and show that the resulting estimation error is exponentially input-to-state stable with respect to the disturbances and to a signal that is decreasing with the increase of the accuracy of the adopted approximation model. An illustrative example shows the effectiveness of the proposed method on real-world neurophysiological networks. Our results may significantly contribute to the development of novel neurotechnologies, particularly in the development of state estimation paradigms for neural signals such as EEG, which are often noisy signals known to be affected by artifacts not having any particular stochastic characterization., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team Sergio Pequito

This paper proposes a robust cooperative distributed secondary control strategy for DC microgrids, with the main focus on reducing communication burdens. To this end, we adopt a sparsity-promoting consensus-based distributed secondary control framework for converter-interfaced DC microgrids consisting of multiple distributed generation (DG) units. The proposed control strategy relies on less (sparse) information exchange over a communication network and does not require exchanging voltage signals or their estimated values amongst neighbouring DG units. Rigorous Lyapunov-based stability certificates for DC microgrids are derived. The stability conditions can be verified without the knowledge of distribution lines connecting different DG units. The effectiveness of the proposed secondary control approach is verified by experimental results and a comparative simulation case study on a multiple-DG DC microgrid.

This brief discusses data-driven design techniques for batch-to-batch optimization problems and proposes a new input-mapping-based online uncertainty compensation method for optimization-based iterative learning control (ILC) with limited memory. Since process uncertainties are generally inevitable, we collect historical data that incorporates past inputs and outputs to provide more insight into plant uncertain dynamics, resulting in a more accurate optimization model for optimal ILC solution. Instead of learning from a single step, our proposed method maintains a memory of the latest executed steps and updates the ILC solution using a linear combination of the memory and quadratic programming. A rigorous theoretical analysis shows that such a design provides robust benefits as well as asymptotic and monotonic stability properties under mild conditions. Finally, we demonstrate our design through an illustrative numerical example. IEEE

In this article, we study the optimal iterative learning control (ILC) for constrained systems with bounded uncertainties via a novel conic input mapping (CIM) design methodology. Due to the limited understanding of the process of interest, modeling uncertainties are generally inevitable, significantly reducing the convergence rate of the control systems. However, huge amounts of measured process data interacting with model uncertainties can easily be collected. Incorporating these data into the optimal controller design could unlock new opportunities to reduce the error of the current trail optimization. Based on several existing optimal ILC methods, we incorporate the online process data into the optimal and robust optimal ILC design, respectively. Our methodology, called CIM, utilizes the process data for the first time by applying the convex cone theory and maps the data into the design of control inputs. CIM-based optimal ILC and robust optimal ILC methods are developed for uncertain systems to achieve better control performance and a faster convergence rate. Next, rigorous theoretical analyses for the two methods have been presented, respectively. Finally, two illustrative numerical examples are provided to validate our methods with improved performance.

This article involves the innovative development of a symmetric Gauss–Seidel ADMM algorithm to solve the H∞ guaranteed cost control problem. In the presence of parametric uncertainties, the H∞ guaranteed cost control problem generally leads to the large-scale optimization. This is due to the exponential growth of the number of the extreme systems involved with respect to the number of parametric uncertainties. In this work, through a variant of the Youla–Kucera parameterization, the stabilizing controllers are parameterized in a convex set; yielding the outcome that the H∞ guaranteed cost control problem is converted to a convex optimization problem. Based on an appropriate reformulation using the Schur complement, it then renders possible the use of the ADMM algorithm with symmetric Gauss–Seidel backward and forward sweeps. Significantly, this approach alleviates the often-times prohibitively heavy computational burden typical in many H∞ optimization problems while exhibiting good convergence guarantees, which is particularly essential for the related large-scale optimization procedures involved. With this approach, the desired robust stability is ensured, and the disturbance attenuation is maintained at the minimum level in the presence of parametric uncertainties. Rather importantly too, with the attained effectiveness, the methodology thus evidently possesses extensive applicability in various important controller synthesis problems, such as decentralized control, sparse control, and output feedback control problems.

Soft robots are intrinsically underactuated mechanical systems that operate under uncertainties and disturbances. In these conditions, this letter proposes two versions of PID-like control laws with a saturated integral action for the particularly challenging shape regulation task. The closed-loop system is asymptotically stabilized and matched constant disturbances are rejected using a very reduced amount of system information for control implementation. Stability is assessed on the underactuated dynamic model through the Invariant Set Theorem for two relevant classes of soft robots, i.e., elastically decoupled and elastically dominated soft robots. Extensive simulation results validate the proposed controllers., Learning & Autonomous Control

We study the uniform approximation of echo state networks (ESNs) with randomly generated internal weights. These models, in which only the readout weights are optimized during training, have made empirical success in learning dynamical systems. Recent results showed that ESNs with ReLU activation are universal. In this article, we give an alternative construction and prove that the universality holds for general activation functions. Specifically, our main result shows that, under certain condition on the activation function, there exists a sampling procedure for the internal weights so that the ESN can approximate any continuous casual time-invariant operators with high probability. In particular, for ReLU activation, we give explicit construction for these sampling procedures. We also quantify the approximation error of the constructed ReLU ESNs for sufficiently regular operators. IEEE

Many applications in signal processing require the estimation of mean and covariance matrices of multivariate complex-valued data. Often, the data are non-Gaussian and are corrupted by outliers or impulsive noise. To mitigate this, robust estimators are employed. However, existing robust estimation techniques employed in signal processing, such as M-estimators, provide limited robustness in the multivariate case. For this reason, this paper introduces the signal processing community to the highly robust class of multivariate estimators called multivariate S-estimators. The paper extends multivariate S-estimation theory to the complex-valued domain. The theoretical performances of S-estimators are explored and compared with M-estimators through the practical lens of the minimum variance distortionless response (MVDR) beamformer, and the empirical finite-sample performances of the estimators are explored through the practical lens of direction-of-arrival (DOA) estimation using the multiple signal classification (MUSIC) algorithm.

In this article, we study the optimal iterative learning control (ILC) for constrained systems with bounded uncertainties via a novel conic input mapping (CIM) design methodology. Due to the limited understanding of the process of interest, modeling uncertainties are generally inevitable, significantly reducing the convergence rate of the control systems. However, huge amounts of measured process data interacting with model uncertainties can easily be collected. Incorporating these data into the optimal controller design could unlock new opportunities to reduce the error of the current trail optimization. Based on several existing optimal ILC methods, we incorporate the online process data into the optimal and robust optimal ILC design, respectively. Our methodology, called CIM, utilizes the process data for the first time by applying the convex cone theory and maps the data into the design of control inputs. CIM-based optimal ILC and robust optimal ILC methods are developed for uncertain systems to achieve better control performance and a faster convergence rate. Next, rigorous theoretical analyses for the two methods have been presented, respectively. Finally, two illustrative numerical examples are provided to validate our methods with improved performance. IEEE

Reconfigurable intelligent surfaces (RISs) are an emerging technology for future wireless communication. The vast majority of recent research on RIS has focused on system level optimizations. However, developing straightforward and tractable electromagnetic models that are suitable for RIS aided communication modeling remains an open issue. In this paper, we address this issue and derive communication models by using rigorous scattering parameter network analysis. We also propose new RIS architectures based on group and fully connected reconfigurable impedance networks that can adjust not only the phases but also the magnitudes of the impinging waves, which are more general and more efficient than conventional single connected reconfigurable impedance network that only adjusts the phases of the impinging waves. In addition, the scaling law of the received signal power of an RIS aided system with reconfigurable impedance networks is also derived. Compared with the single connected reconfigurable impedance network, our group and fully connected reconfigurable impedance network can increase the received signal power by up to 62%, or maintain the same received signal power with a number of RIS elements reduced by up to 21%. We also investigate the proposed architecture in deployments with distance-dependent pathloss and Rician fading channel, and show that the proposed group and fully connected reconfigurable impedance networks outperform the single connected case by up to 34% and 48%, respectively.

For linear time-invariant (LTI) systems, the design of an optimal controller is a commonly encountered problem in many applications. Among all the optimization approaches available, the linear quadratic regulator (LQR) methodology certainly garners much attention and interest. In various scenarios where some of the system state variables are not measurable, and consequent prescribed structural constraints on the controller structure arise, the optimization problem can become intractable. In such cases, there have been some first-order methods proposed to cater to these problems, but all of these first-order optimization methods are limited to only linear convergence. To speed up the convergence, a second-order approach in the matrix space is essential. Thus in this work, an efficient method is proposed in the matrix space to calculate the Hessian matrix by solving several Lyapunov equations. Then a new optimization technique is applied to deal with the indefiniteness of the Hessian matrix Subsequently, through Newton's method with linear equality constraints, a second-order optimization algorithm is developed to effectively solve the constrained SOF LQR problem. IEEE

The intrinsically underactuated and nonlinear nature of continuum soft robots makes the derivation of provably stable feedback control laws a challenging task. Most of the works so far circumvented the issue either by looking at coarse fully-actuated approximations of the dynamics or by imposing quasi-static assumptions. In this letter, we move a step in the direction of controlling generic soft robots taking explicitly into account their underactuation. A class of soft robots that have no direct elastic couplings between the dynamics of actuated and unactuated coordinates is identified. Considering the actuated variables as output, we prove that the system is minimum phase. We then propose regulators that implement different levels of model compensation. The stability of the associated closed-loop systems is formally proven by Lyapunov/LaSalle techniques, taking into account the nonlinear and underactuated dynamics. Simulation results are reported for two models of 2D and 3D soft robots., Accepted Author Manuscript, Learning & Autonomous Control

We present a learning method to learn the mapping from an input space to an action space, which is particularly suitable when the action is an optimal decision with respect to a certain unknown cost function. We use an inverse optimization approach to retrieve the cost function by introducing a new loss function and a new hypothesis class of mappings. A tractable convex reformulation of the learning problem is also presented. The method is effective for learning input-action mapping in continuous input-action space with input-output constraints, typically present in control systems. The learning approach can be effectively transformed to learn a Model Predictive Control (MPC) behaviour and a case study to mimic an MPC is presented, which is a rather computationally heavy control strategy. Simulation and experimental results show the effectiveness of the proposed approach., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team Tamas Keviczky

In this article, we investigate the problem of practical output regulation, i.e., to design a controller that brings the system output in the vicinity of a desired target value while keeping the other variables bounded. We consider uncertain systems that are possibly nonlinear and the uncertainty of their linear parts is modeled element wise through a parametric family of matrix boxes. An optimization-based design procedure is proposed that delivers a continuous-time control and estimates the maximal regulation error. We also analyze an event-triggered emulation of this controller, which can be implemented on a digital platform, along with an explicit estimate of the regulation error., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team DeSchutter, Team Peyman Mohajerin Esfahani

The existence of an optimal reverse-waterfilling algorithm to compute the nonanticipative rate distortion function (NRDF) for time-invariant vector-valued Gauss-Markov processes with a mean-squared-error distortion has been an open question since the pioneering work of Tatikonda et al. on stochastic linear control over a communication channel, in 2004. In this article, we derive strong structural properties on the time-invariant multidimensional Gauss-Markov processes that allow for an optimization problem that can be computed optimally via a reverse-waterfilling algorithm. Moreover, we propose an elegant optimal iterative scheme that computes this reverse-waterfilling algorithm. We show that the specific scheme operates much faster than any existing algorithmic approach that solves the same problem optimally and is also scalable. Finally, using our new results, we derive for the first time a nontrivial analytical solution of the asymptotic NRDF using a correlated time-invariant 2-D Gauss-Markov process., QC 20220419

In this article, we propose a distributed negotiation framework that allows a set of cooperative agents to find a common ground with their neighbors while attempting to modify their initial opinion by the least possible amount. Based on such a framework, we develop a distributed agreement approach where the effort spent in the local agreement reflects the relevance of the agents in a weighted consensus process. In particular, we assume that players whose ideas happen to satisfactory mediate the standpoint of their interlocutors will end-up being more influential in the overall decision-making process. We conclude the article by applying the proposed methodology in the context of distributed data aggregation scenarios, as a way to mitigate the effect of outliers (e.g., faulty sensors)., QC 20220916

This work investigates the fixed-time distributed coordination control for multiple Euler-Lagrange systems and, in particular, addresses containment control with stationary/dynamic leaders as well as leaderless synchronization control. For the containment control scenario with stationary leaders, the subgraph associated with followers is directed. When dynamic leaders are involved, the information transfer between neighboring followers is bidirectional, for which a novel distributed estimator is developed. For the leaderless synchronization control scenario, the communication network among agents is unidirectional. Three fixed-time distributed control schemes are designed for the aforementioned three cases by applying the fixed-time stability theory. The convergence of the coordination control objectives can be achieved in a fixed time that does not depend on any initial conditions of agents' states, and the settling times are also explicitly derived. Finally, numerical simulations are presented to demonstrate the feasibility of the developed control strategies.

The parallel connection of DC-DC converters requires the development of an appropriate control strategy that regulates load voltage and shares current amongst participating converters. This paper proposes a resilient and robust cooperative distributed control approach that simultaneously ensures voltage regulation and balanced current sharing in parallel DC-DC converters in the presence of false data injection attacks on control input channels. Based on analytical tools from network control and Lyapunov stability theory, concise stability certificates are derived. The proposed cooperative distributed control strategy guarantees resilience against unknown bounded attacks on the actuators of DC-DC converters and the robustness to uncertainties in load parameters and the physical parameters of converters. Furthermore, the control design for each converter does not require any knowledge about the number of participating converters. The detailed simulation and experimental results verify the satisfactory performance of the proposed method in voltage regulation and balanced current sharing in parallel converters, as well as resilience to bounded false data injection attacks.

In real-world applications, it is important for machine learning algorithms to be robust against data outliers or corruptions. In this paper, we focus on improving the robustness of a large class of learning algorithms that are formulated as low-rank semi-definite programming (SDP) problems. Traditional formulations use the square loss, which is notorious for being sensitive to outliers. We propose to replace this with more robust noise models, including the l(1)-loss and other nonconvex losses. However, the resultant optimization problem becomes difficult as the objective is no longer convex or smooth. To alleviate this problem, we design an efficient algorithm based on majorization-minimization. The crux is on constructing a good optimization surrogate, and we show that this surrogate can be efficiently obtained by the alternating direction method of multipliers (ADMM). By properly monitoring ADMM's convergence, the proposed algorithm is empirically efficient and also theoretically guaranteed to converge to a critical point. Extensive experiments are performed on four machine learning applications using both synthetic and real-world data sets. Results show that the proposed algorithm is not only fast but also has better performance than the state-of-the-arts.

For linear time-invariant (LTI) systems, the design of an optimal controller is a commonly encountered problem in many applications. Among all the optimization approaches available, the linear quadratic regulator (LQR) methodology certainly garners much attention and interest. In various scenarios where some of the system state variables are not measurable, and consequent prescribed structural constraints on the controller structure arise, the optimization problem can become intractable. In such cases, there have been some first-order methods proposed to cater to these problems, but all of these first-order optimization methods are limited to only linear convergence. To speed up the convergence, a second-order approach in the matrix space is essential. Thus in this work, an efficient method is proposed in the matrix space to calculate the Hessian matrix by solving several Lyapunov equations. Then a new optimization technique is applied to deal with the indefiniteness of the Hessian matrix Subsequently, through Newton's method with linear equality constraints, a second-order optimization algorithm is developed to effectively solve the constrained SOF LQR problem. IEEE

This article addresses the multiagent systems consensus control problem via a learning observer-based performance tuning control policy. Specifically, a novel learning observer is presented to reconstruct the compound nonlinear terms and system states simultaneously. Based on the learning observer’s reconstructed information, a novel performance tuning control policy is proposed to deal with the internal nonlinear terms and external disturbances acting on the system while providing prescribed consensus performance. The proposed learning observer can guarantee the uniformly ultimately bounded estimation while saving computing resources, which is beneficial to the multiagent system. The proposed control policy, combined with observer and performance tuning, ensures the robustness to nonideal perturbations and the high accuracy control performance simultaneously. Mathematical simulations verify the effectiveness of the control algorithm.

Reconfigurable intelligent surfaces (RISs) are an emerging technology for future wireless communication. The vast majority of recent research on RIS has focused on system level optimizations. However, developing straightforward and tractable electromagnetic models that are suitable for RIS aided communication modeling remains an open issue. In this paper, we address this issue and derive communication models by using rigorous scattering parameter network analysis. We also propose new RIS architectures based on group and fully connected reconfigurable impedance networks that can adjust not only the phases but also the magnitudes of the impinging waves, which are more general and more efficient than conventional single connected reconfigurable impedance network that only adjusts the phases of the impinging waves. In addition, the scaling law of the received signal power of an RIS aided system with reconfigurable impedance networks is also derived. Compared with the single connected reconfigurable impedance network, our group and fully connected reconfigurable impedance network can increase the received signal power by up to 62%, or maintain the same received signal power with a number of RIS elements reduced by up to 21%. We also investigate the proposed architecture in deployments with distance-dependent pathloss and Rician fading channel, and show that the proposed group and fully connected reconfigurable impedance networks outperform the single connected case by up to 34% and 48%, respectively.

In this article, we investigate the problem of practical output regulation, i.e., to design a controller that brings the system output in the vicinity of a desired target value while keeping the other variables bounded. We consider uncertain systems that are possibly nonlinear and the uncertainty of their linear parts is modeled element wise through a parametric family of matrix boxes. An optimization-based design procedure is proposed that delivers a continuous-time control and estimates the maximal regulation error. We also analyze an event-triggered emulation of this controller, which can be implemented on a digital platform, along with an explicit estimate of the regulation error., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team Bart De Schutter, Team Peyman Mohajerin Esfahani

The intrinsically underactuated and nonlinear nature of continuum soft robots makes the derivation of provably stable feedback control laws a challenging task. Most of the works so far circumvented the issue either by looking at coarse fully-actuated approximations of the dynamics or by imposing quasi-static assumptions. In this letter, we move a step in the direction of controlling generic soft robots taking explicitly into account their underactuation. A class of soft robots that have no direct elastic couplings between the dynamics of actuated and unactuated coordinates is identified. Considering the actuated variables as output, we prove that the system is minimum phase. We then propose regulators that implement different levels of model compensation. The stability of the associated closed-loop systems is formally proven by Lyapunov/LaSalle techniques, taking into account the nonlinear and underactuated dynamics. Simulation results are reported for two models of 2D and 3D soft robots., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Learning & Autonomous Control

We present a learning method to learn the mapping from an input space to an action space, which is particularly suitable when the action is an optimal decision with respect to a certain unknown cost function. We use an inverse optimization approach to retrieve the cost function by introducing a new loss function and a new hypothesis class of mappings. A tractable convex reformulation of the learning problem is also presented. The method is effective for learning input-action mapping in continuous input-action space with input-output constraints, typically present in control systems. The learning approach can be effectively transformed to learn a Model Predictive Control (MPC) behaviour and a case study to mimic an MPC is presented, which is a rather computationally heavy control strategy. Simulation and experimental results show the effectiveness of the proposed approach., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team Tamas Keviczky

© 2021 IEEE. Personal use of this material is permitted. Permissíon from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertisíng or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] In this article, a parallel structured divide-and-conquer (PSDC) eigensolver is proposed for symmetric tridiagonal matrices based on ScaLAPACK and a parallel structured matrix multiplication algorithm, called PSMMA. Computing the eigenvectors via matrix-matrix multiplications is the most computationally expensive part of the divide-and-conquer algorithm, and one of the matrices involved in such multiplications is a rank-structured Cauchy-like matrix. By exploiting this particular property, PSMMA constructs the local matrices by using generators of Cauchy-like matrices without any communication, and further reduces the computation costs by using a structured low-rank approximation algorithm. Thus, both the communication and computation costs are reduced. Experimental results show that both PSMMA and PSDC are highly scalable and scale to 4096 processes at least. PSDC has better scalability than PHDC that was proposed in [16] and only scaled to 300 processes for the same matrices. Comparing with PDSTEDC in ScaLAPACK, PSDC is always faster and achieves 1.4x-1.6x speedup for some matrices with few deflations. PSDC is also comparable with ELPA, with PSDC being faster than ELPA when using few processes and a little slower when using many processes.

© 2021 IEEE. Personal use of this material is permitted. Permissíon from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertisíng or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] In this article, a parallel structured divide-and-conquer (PSDC) eigensolver is proposed for symmetric tridiagonal matrices based on ScaLAPACK and a parallel structured matrix multiplication algorithm, called PSMMA. Computing the eigenvectors via matrix-matrix multiplications is the most computationally expensive part of the divide-and-conquer algorithm, and one of the matrices involved in such multiplications is a rank-structured Cauchy-like matrix. By exploiting this particular property, PSMMA constructs the local matrices by using generators of Cauchy-like matrices without any communication, and further reduces the computation costs by using a structured low-rank approximation algorithm. Thus, both the communication and computation costs are reduced. Experimental results show that both PSMMA and PSDC are highly scalable and scale to 4096 processes at least. PSDC has better scalability than PHDC that was proposed in [16] and only scaled to 300 processes for the same matrices. Comparing with PDSTEDC in ScaLAPACK, PSDC is always faster and achieves 1.4x-1.6x speedup for some matrices with few deflations. PSDC is also comparable with ELPA, with PSDC being faster than ELPA when using few processes and a little slower when using many processes.

Reconfigurable intelligent surfaces (RISs) are an emerging technology for future wireless communication. The vast majority of recent research on RIS has focused on system level optimizations. However, developing straightforward and tractable electromagnetic models that are suitable for RIS aided communication modeling remains an open issue. In this paper, we address this issue and derive communication models by using rigorous scattering parameter network analysis. We also propose new RIS architectures based on group and fully connected reconfigurable impedance networks that can adjust not only the phases but also the magnitudes of the impinging waves, which are more general and more efficient than conventional single connected reconfigurable impedance network that only adjusts the phases of the impinging waves. In addition, the scaling law of the received signal power of an RIS aided system with reconfigurable impedance networks is also derived. Compared with the single connected reconfigurable impedance network, our group and fully connected reconfigurable impedance network can increase the received signal power by up to 62%, or maintain the same received signal power with a number of RIS elements reduced by up to 21%. We also investigate the proposed architecture in deployments with distance-dependent pathloss and Rician fading channel, and show that the proposed group and fully connected reconfigurable impedance networks outperform the single connected case by up to 34% and 48%, respectively. CCBY

In real-world applications, it is important for machine learning algorithms to be robust against data outliers or corruptions. In this paper, we focus on improving the robustness of a large class of learning algorithms that are formulated as low-rank semi-definite programming (SDP) problems. Traditional formulations use the square loss, which is notorious for being sensitive to outliers. We propose to replace this with more robust noise models, including the $\ell_1$-loss and other nonconvex losses. However, the resultant optimization problem becomes difficult as the objective is no longer convex or smooth. To alleviate this problem, we design an efficient algorithm based on majorization-minimization. The crux is on constructing a good optimization surrogate, and we show that this surrogate can be efficiently obtained by the alternating direction method of multipliers (ADMM). By properly monitoring ADMM's convergence, the proposed algorithm is empirically efficient and also theoretically guaranteed to converge to a critical point. Extensive experiments are performed on four machine learning applications using both synthetic and real-world data sets. Results show that the proposed algorithm is not only fast but also has better performance than the state-of-the-arts. IEEE

This article addresses the multiagent systems consensus control problem via a learning observer-based performance tuning control policy. Specifically, a novel learning observer is presented to reconstruct the compound nonlinear terms and system states simultaneously. Based on the learning observer’s reconstructed information, a novel performance tuning control policy is proposed to deal with the internal nonlinear terms and external disturbances acting on the system while providing prescribed consensus performance. The proposed learning observer can guarantee the uniformly ultimately bounded estimation while saving computing resources, which is beneficial to the multiagent system. The proposed control policy, combined with observer and performance tuning, ensures the robustness to nonideal perturbations and the high accuracy control performance simultaneously. Mathematical simulations verify the effectiveness of the control algorithm. IEEE

© 2021 IEEE. Personal use of this material is permitted. Permissíon from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertisíng or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., [EN] In this article, a parallel structured divide-and-conquer (PSDC) eigensolver is proposed for symmetric tridiagonal matrices based on ScaLAPACK and a parallel structured matrix multiplication algorithm, called PSMMA. Computing the eigenvectors via matrix-matrix multiplications is the most computationally expensive part of the divide-and-conquer algorithm, and one of the matrices involved in such multiplications is a rank-structured Cauchy-like matrix. By exploiting this particular property, PSMMA constructs the local matrices by using generators of Cauchy-like matrices without any communication, and further reduces the computation costs by using a structured low-rank approximation algorithm. Thus, both the communication and computation costs are reduced. Experimental results show that both PSMMA and PSDC are highly scalable and scale to 4096 processes at least. PSDC has better scalability than PHDC that was proposed in [16] and only scaled to 300 processes for the same matrices. Comparing with PDSTEDC in ScaLAPACK, PSDC is always faster and achieves 1.4x-1.6x speedup for some matrices with few deflations. PSDC is also comparable with ELPA, with PSDC being faster than ELPA when using few processes and a little slower when using many processes.

In real-world applications, it is important for machine learning algorithms to be robust against data outliers or corruptions. In this paper, we focus on improving the robustness of a large class of learning algorithms that are formulated as low-rank semi-definite programming (SDP) problems. Traditional formulations use the square loss, which is notorious for being sensitive to outliers. We propose to replace this with more robust noise models, including the $\ell_1$-loss and other nonconvex losses. However, the resultant optimization problem becomes difficult as the objective is no longer convex or smooth. To alleviate this problem, we design an efficient algorithm based on majorization-minimization. The crux is on constructing a good optimization surrogate, and we show that this surrogate can be efficiently obtained by the alternating direction method of multipliers (ADMM). By properly monitoring ADMM's convergence, the proposed algorithm is empirically efficient and also theoretically guaranteed to converge to a critical point. Extensive experiments are performed on four machine learning applications using both synthetic and real-world data sets. Results show that the proposed algorithm is not only fast but also has better performance than the state-of-the-arts. IEEE

This article addresses the multiagent systems consensus control problem via a learning observer-based performance tuning control policy. Specifically, a novel learning observer is presented to reconstruct the compound nonlinear terms and system states simultaneously. Based on the learning observer’s reconstructed information, a novel performance tuning control policy is proposed to deal with the internal nonlinear terms and external disturbances acting on the system while providing prescribed consensus performance. The proposed learning observer can guarantee the uniformly ultimately bounded estimation while saving computing resources, which is beneficial to the multiagent system. The proposed control policy, combined with observer and performance tuning, ensures the robustness to nonideal perturbations and the high accuracy control performance simultaneously. Mathematical simulations verify the effectiveness of the control algorithm. IEEE

This article deals with networked control systems (NCSs) whose shared networks have limited communication capacity and are prone to data losses. Our contributions are two-fold. First, we present necessary and sufficient conditions on the plant and controller dynamics and the network parameters such that there exist purely time-dependent periodic scheduling sequences under which the stability of each plant is preserved for all admissible data loss signals. Second, given a period for the scheduling sequence, we identify sufficient conditions on the plant dynamics and the network parameters such that the plants admit static state-feedback controllers that are favorable for the stability conditions presented. The main apparatus for our analysis is a switched system representation of the individual plants in an NCS whose switching signals are time-inhomogeneous Markov chains. Our stability conditions involve the existence of sets of symmetric and positive definite matrices that satisfy certain (in)equalities. We identify the existence of stabilizing periodic scheduling sequences and their corresponding favorable state-feedback controllers under ideal communication between the plants and their controllers as special cases of our results. A numerical experiment is presented to demonstrate the proposed techniques.

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 of symmetrically structured matrices, as well as the modes of any numerical realization. Subsequently, we provide graph-theoretic necessary and sufficient conditions for the structural target controllability of undirected networks with multiple control nodes. In addition, we show that these results can be extended and lead to a necessary and sufficient condition of the structural output controllability. However, different from structural target controllability, we prove that verifying the proposed conditions on structural output controllability in undirected networks is NP-hard., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Team Tamas Keviczky