Set-valued observers and optimal disturbance rejection

A set-valued observer (also called guaranteed state estimator) produces a set of possible states based on output measurements and models of exogenous signals. We consider the guaranteed state estimation problem for linear time-varying systems with a priori magnitude bounds on exogenous signals. We provide an algorithm to propagate the set of possible states based on output measurements and show that the centers of these sets provide optimal estimates in an l^∞-induced norm sense. We then consider the utility of set-valued observers for disturbance rejection with output feedback and derive the following general separation structure. An optimal controller can consist of a set-valued observer followed by a static nonlinear function on the observed set of possible states. A general construction of this function is provided in the scalar control case. Furthermore, in the special case of full-control, i.e., the number of control inputs equals the number of states, optimal output feedback controllers can take the form of an optimal estimate of the full-state feedback controller     

Single-lifting Macaulay-type formulae of generalized unmixed sparse resultants

Resultants are defined in the sparse (or toric) context in order to exploit the structure of the polynomials as expressed by their Newton polytopes. Since determinantal formulae are not always possible, the most efficient general method for computing resultants is rational formulae. This is made possible by Macaulay’s famous determinantal formula in the dense homogeneous case, extended by D’Andrea to the sparse case. However, the latter requires a lifting of the Newton polytopes, defined recursively on the dimension. Our main contribution is a single-lifting function of the Newton polytopes, which avoids recursion, and yields a simpler method for computing Macaulay-type formulae of sparse resultants. We focus on the case of generalized unmixed systems, where all Newton polytopes are scaled copies of each other, and sketch how our approach may extend to mixed systems of up to four polynomials, as well as those whose Newton polytopes have a sufficiently different face structure. In the mixed subdivision used to construct the matrices, our algorithm defines significantly fewer cells than D’Andrea’s, though the matrix formulae are same. We discuss asymptotic complexity bounds and illustrate our results by fully studying a bivariate example.     

Guaranteed non-linear estimation using constraint propagation on sets

Bounded-error estimation is the estimation of the parameter or state vector of a model from experimental data, under the assumption that some suitably defined errors should belong to some prior feasible sets. When the model outputs are linear in the vector to be estimated, a number of methods are available to contain all estimates that are consistent with the data within simple sets such as ellipsoids, orthotopes or parallelotopes, thereby providing guaranteed set estimates. In the non-linear case, the situation is much less developed and there are very few methods that produce such guaranteed estimates. In this paper, the problem of characterizing the set of all state vectors that are consistent with all data in the case of non-linear discrete-time systems is cast into the more general framework of constraint satisfaction problems. The state vector at time k should be estimated either on-line from past measurement only or off-line from a series of measurements that may include measurements posterior to k . Even in the causal case, prior information on the future value of the state and output vectors, due for instance to physical constraints, is readily taken into account. Algorithms taken from the literature of interval constraint propagation are extended by replacing intervals by more general subsets of real vector spaces. This makes it possible to propose a new algorithm that contracts the feasible domain for each uncertain variable optimally (i.e. no smaller domain could be obtained) and efficiently.     

A Right-Preconditioning Process for the Formal–Algebraic Approach to Inner and Outer Estimation of AE-Solution Sets

Aright-preconditioning process for linear interval systems has been presented by Neumaier in 1987. It allows the construction of an outer estimate of the united solution set of a square linear interval system in the form of a parallelepiped. The denomination “right-preconditioning” is used to describe the preconditioning processes which involve the matrix product AC in contrast to the (usual) left-preconditioning processes which involve the matrix product AC, where A and C are respectively the interval matrix of the studied linear interval system and the preconditioning matrix.The present paper presents a new right-preconditioning process similar to the one presented by Neumaier in 1987 but in the more general context of the inner and outer estimations of linear AEsolution sets. Following the spirit of the formal-algebraic approach to AE-solution sets estimation, summarized by Shary in 2002, the new right-preconditioning process is presented in the form of two new auxiliary interval equations. Then, the resolution of these auxiliary interval equations leads to inner and outer estimates of AE-solution sets in the form of parallelepipeds. This right-preconditioning process has two advantages: on one hand, the parallelepipeds estimates are often more precise than the interval vectors estimates computed by Shary. On the other hand, in many situations, it simplifies the formal algebraic approach to inner estimation of AE-solution sets. Therefore, some AE-solution sets which were almost impossible to inner estimate with interval vectors, become simple to inner estimate using parallelepipeds. These benefits are supported by theoretical results and by some experimentations on academic examples of linear interval systems.     

Set-theoretic approach to uncertain dynamical systems using convex polytopes

A set-theoretic approach to uncertain dynamical systems by using convex polytopes is treated. In some points, this approach is superior to the one using ellipsoids, though it also has some disadvantages. First, some fundamental operations are discussed, such as addition, intersection, and the linear transformation of convex polytopes. The results are applied to the analysis of reachable sets for discrete-time linear dynamical systems and also to the state estimation.     

Ellipsoidal parameter or state estimation under model uncertainty

Ellipsoidal outer-bounding of the set of all feasible state vectors under model uncertainty is a natural extension of state estimation for deterministic models with unknown-but-bounded state perturbations and measurement noise. The technique described in this paper applies to linear discrete-time dynamic systems; it can also be applied to weakly non-linear systems if non-linearity is replaced by uncertainty. Many difficulties arise because of the non-convexity of feasible sets. Combined quadratic constraints on model uncertainty and additive disturbances are considered in order to simplify the analysis. Analytical optimal or suboptimal solutions of the basic problems involved in parameter or state estimation are presented, which are counterparts in this context of uncertain models to classical approximations of the sum and intersection of ellipsoids. The results obtained for combined quadratic constraints are extended to other types of model uncertainty.     

Convergence analysis for reduced-order adaptive controller design of uncertain SISO linear systems with noisy output measurements

We consider in this article a class of uncertain SISO linear systems that are subject to system and measurement noises. Reduced-order adaptive controller designs have been proposed before for such systems by the authors and stability analysis of the closed-loop systems has been established. Here we analyse, further, the robustness properties for these reduced-order adaptive control systems by providing detailed convergence analysis results for the key closed-loop signals and parameter estimates. We rigorously prove that, whenever the exogenous disturbance input is of finite energy and bounded, and the reference trajectory and its derivatives up to rth order are bounded, r being the relative degree of the transfer function of the true system, a set of signals, including the tracking error, the estimation error between the system output and its estimate, the projection signal, are of finite energy and converge to zero; and the system states and their estimates exhibit asymptotic behaviours with certain formats. With an additional persistency of excitation condition, it is also proved that the estimate and the worst-case estimate of the state vector asymptotically track the actual state vector; and the estimate and the worst-case estimate of the unknown parameter vector converge to the true value. A numerical example is given to illustrate the theoretical findings.     

An analysis and design method for a class of nonlinear systems with nested saturations

This paper is concerned with the analysis and design of a class of nonlinear systems subject to nested saturations. The proposed controller incorporates both an extended state observer (ESO), which is utilised to estimate the nonlinear dynamics of the plant, as well as a set of observer-based feedbacks. We first present analysis results for systems with nonlinear ESOs and show that local stabilisation can be achieved in a region including the origin. Then, in the case that the ESO is in linear form (LESO), the conditions for determining the estimate of the domain of attraction of the resulting closed-loop system are formulated as a convex optimisation problem. A linear matrix inequality based algorithm is then established to design the feedback gains and the LESO gain. An illustrative example is given to show the effectiveness of the proposed approach.     

Semi-global sampled-data output feedback disturbance rejection control for a class of uncertain nonlinear systems

This paper investigates the semi-global output feedback disturbance rejection control problem for a class of uncertain nonlinear systems with additive disturbances using linear sampled-data control. Aiming to reject the adverse effects caused by the uncertainties and unknown nonlinear perturbations which may not satisfy the strict feedback or feedforward structure, a new generalised discrete-time extended state observer is proposed to estimate the disturbance at sampling points. An output feedback disturbance rejection control law is then constructed in a sampled-data form which facilitates digital implementations. By selecting adequate control gains and a sufficiently small sampling period to restrain the state growth under a zero-order-hold input, the semi-global asymptotic stability of the hybrid closed-loop system and the disturbance rejection ability are proved. Both numerical example and an application of a single-link robot arm system demonstrate the feasibility and efficacy of the proposed method.     

Optimal filtering for linear systems with state and observation delays

In this paper, the optimal filtering problem for linear systems with state and observation delays is treated proceeding from the general expression for the stochastic Ito differential of the optimal estimate, error variance, and various error covariances. As a result, the optimal estimate equation similar to the traditional Kalman–Bucy one is derived; however, it is impossible to obtain a system of the filtering equations, that is closed with respect to the only two variables, the optimal estimate and the error variance, as in the Kalman–Bucy filter. The resulting system of equations for determining the filter gain matrix consists, in the general case, of an infinite set of equations. It is however demonstrated that a finite set of the filtering equations, whose number is specified by the ratio between the current filtering horizon and the delay values, can be obtained in the particular case of equal or commensurable (τ=qh, q is natural) delays in the observation and state equations. In the example, performance of the designed optimal filter for linear systems with state and observation delays is verified against the best Kalman–Bucy filter available for linear systems without delays and two versions of the extended Kalman–Bucy filter for time delay systems. Copyright © 2005 John Wiley & Sons, Ltd.     

Semi-global weighted output average tracking of discrete-time heterogeneousmulti-agent systems subject to input saturation and external disturbances

In this paper, we revisit the semi-global weighted output average tracking problem for a discrete-time multi-agent system subject to input saturation and external disturbances. The multi-agent system consists of multiple heterogeneous linear systems as leader agents and multiple heterogeneous linear systems as follower agents. We design both the state feedback and output feedback control protocols for each follower agent. In particular, a distributed state observer is designed for each follower agent that estimates the state of each leader agent. In the output feedback case, state observer is also designed for each follower agent to estimate its own state. With these estimates, we design low gain-based distributed control protocols, parameterized in a scalar low gain parameter. It is shown that, for any bounded set of the initial conditions, these control protocols cause the follower agents to track the weighted average of the outputs of the leader agents as long as the value of the low gain parameter is tuned sufficiently small. Simulation results illustrate the validity of the theoretical results.     

State estimation for Markovian Jump Linear Systems with bounded disturbances

In this paper, we investigate the state estimation problem for a class of Markovian Jump Linear Systems (MJLSs) in the presence of bounded polyhedral disturbances. A set-membership estimation algorithm is first proposed to find the smallest consistent set of all possible states, which is shown to be expressed by a union of multiple polytopes. The posterior probabilities of the system jumping modes are then estimated by introducing the Lebesgue measure, based on which the optimal point estimate is further provided. Moreover, a similarity relationship for polytopes is defined and an approximate method is presented to calculate the Minkowski sum of polytopes, which can help reduce the computational complexity of the overall estimation algorithm.     

Kalman filter-based multialternative method for fault detection and estimation

Proposed was a computationally efficient multialternative method for detection and estimation of faults additively involved in the right-hand sides of the linear equations of the state and measurement vectors. In distinction to the classical approach using a Kalman filter bank for each fault, the paper suggested an extended Kalman filter estimating a set of possible faults. The a posteriori probabilities and estimates of individual faults were shown to be readily calculable from the estimates and covariance matrices generated by the extended Kalman filter. Efficiency of the method is corroborated by modeling of a navigation complex with two inertial systems.     

Interval set‐membership estimation for continuous linear systems

This article proposes a mixed interval set‐membership estimation (ISME) method for continuous linear time‐invariant (LTI) systems by combining the positive system theory and the set theory. The proposed ISME method gives a new mixed interval‐set estimation framework for continuous LTI systems, whose benefit consists in that it has potential to achieve a balance of computational complexity and robust state estimation conservatism with respect to the interval observer (IO) and the set‐valued observer (SVO) for continuous LTI systems. Particularly, the proposed ISME method first uses a coordinate transformation such that the original system is transformed into an equivalent system. Second, the equivalent system is partitioned into two subsystems, where the first subsystem has a Meztler and Hurwitz subsystem matrix and then an IO is designed for the first subsystem based on the positive system theory. Because it is not guaranteed that the second subsystem also has a Meztler and Hurwitz subsystem matrix, a zonotopic SVO is further designed for the second subsystem based on the set theory. Consequently, an integration of the two steps above provides the whole SE results for the original system. At the end of this article, an example is used to illustrate the effectiveness of the proposed ISME method.     

Offline Robust Model Predictive Control for Lipschitz Non‐Linear Systems Using Polyhedral Invariant Sets

In this paper the concept of maximal admissible set (MAS) for linear systems with polytopic uncertainty is extended to non‐linear systems composed of a linear constant part followed by a non‐linear term. We characterize the maximal admissible set for the non‐linear system with unstructured uncertainty in the form of polyhedral invariant sets. A computationally efficient state‐feedback RMPC law is derived off‐line for Lipschitz non‐linear systems. The state‐feedback control law is calculated by solving a convex optimization problem within the framework of linear matrix inequalities (LMIs), which leads to guaranteeing closed‐loop robust stability. Most of the computational burdens are moved off‐line. A linear optimization problem is performed to characterize the maximal admissible set, and it is shown that an ellipsoidal invariant set is only an approximation of the true stabilizable region. This method not only remarkably extends the size of the admissible set of initial conditions but also greatly reduces the on‐line computational time. The usefulness and effectiveness of the method proposed here is verified via two simulation examples.     

Counting the solutions of Presburger equations without enumerating them

The number decision diagram (NDD) has recently been introduced as a powerful representation system for sets of integer vectors. NDDs can notably be used for handling sets defined by arbitrary Presburger formulas, which makes them well suited for representing the set of reachable states of finite-state systems extended with unbounded integer variables. In this paper, we address the problem of counting the number of distinct elements in a set of numbers or, more generally, of vectors, represented by an NDD. We give an algorithm that is able to produce an exact count without enumerating explicitly the vectors, which makes it capable of handling very large sets. As an auxiliary result, we also develop an efficient projection method that allows to construct efficiently NDDs from quantified formulas, and thus makes it possible to apply our counting technique to sets specified by formulas. Our algorithms have been implemented in the verification tool LASH, and applied successfully to various counting problems.     

Classification of linearly nonseparable patterns by linearthreshold elements

Learning and convergence properties of linear threshold elements or perceptrons are well understood for the case where the input vectors (or the training sets) to the perceptron are linearly separable. Little is known, however, about the behavior of the perceptron learning algorithm when the training sets are linearly nonseparable. We present the first known results on the structure of linearly nonseparable training sets and on the behavior of perceptrons when the set of input vectors is linearly nonseparable. More precisely, we show that using the well known perceptron learning algorithm, a linear threshold element can learn the input vectors that are provably learnable, and identify those vectors that cannot be learned without committing errors. We also show how a linear threshold element can be used to learn large linearly separable subsets of any given nonseparable training set. In order to develop our results, we first establish formal characterizations of linearly nonseparable training sets and define learnable structures for such patterns. We also prove computational complexity results for the related learning problems. Next, based on such characterizations, we show that a perceptron does the best one can expect for linearly nonseparable sets of input vectors and learns as much as is theoretically possible.     

On invariant sets and closed‐loop boundedness of Lure‐type nonlinear systems by LPV‐embedding

We address the problem of achieving trajectory boundedness and computing ultimate bounds and invariant sets for Lure‐type nonlinear systems with a sector‐bounded nonlinearity. Our first contribution is to compare two systematic methods to compute invariant sets for Lure systems. In the first method, a linear‐like bound is considered for the nonlinearity, and this bound is used to compute an invariant set by regarding the nonlinear system as a linear system with a nonlinear perturbation. In the second method, the sector‐bounded nonlinearity is treated as a time‐varying parameterised linear function with bounded parameter variations, and then invariant sets are computed by embedding the nonlinear system into a convex polytopic linear parameter varying (LPV) system. We show that under some conditions on the system matrices, these approaches give identical invariant sets, the LPV‐embedding method being less conservative in the general case. The second contribution of the paper is to characterise a class of Lure systems, for which an appropriately designed linear state feedback achieves bounded trajectories of the closed‐loop nonlinear system and allows for the computation of an invariant set via a simple, closed‐form expression. The third contribution is to show that, for disturbances that are ‘aligned’ with the control input, arbitrarily small ultimate bounds on the system states can be achieved by assigning the eigenvalues of the linear part of the system with ‘large enough’ negative real part. We illustrate the results via examples of a pendulum system, a Josephson junction circuit and the well‐known Chua circuit. Copyright © 2015 John Wiley & Sons, Ltd.     

Optimal filtering for linear state delay systems

In this note, the optimal filtering problem for linear systems with state delay over linear observations is treated proceeding from the general expression for the stochastic Ito differential of the optimal estimate and the error variance. As a result, the optimal estimate equation similar to the traditional Kalman-Bucy one is derived; however, it is impossible to obtain a system of the filtering equations, that is closed with respect to the only two variables, the optimal estimate and the error variance, as in the Kalman-Bucy filter. The resulting system of equations for determining the error variance consists of a set of equations, whose number is specified by the ratio between the current filtering horizon and the delay value in the state equation and increases as the filtering horizon tends to infinity. In the example, performance of the designed optimal filter for linear systems with state delay is verified against the best Kalman-Bucy filter available for linear systems without delays and two versions of the extended Kalman-Bucy filter for time-delay systems.     

基于线性状态反馈和高阶泰勒逼近的自抗扰控制方法

针对对象模型不确定性和输入扰动问题, 设计扩张状态观测器. 提出利用高阶泰勒多项式构造综合扰动的内部模型, 将其作为系统的扩张状态, 由Luenberger 状态观测器对其进行估计. 运用线性状态反馈法, 将原系统状态估值反馈至参考输入, 再结合极点配置法和扩张状态估值得到最终的控制作用. 由于将原系统转化为积分串联型, 实现了系统线性化, 并对干扰进行了有效补偿, 使系统抗扰性能大为增强. 通过数例分析验证了所提出方法的有效性.