Explicit solution to the singular discrete-time stationary linear filtering problem

A closed form solution to the stationary discrete-time linear filtering problem is obtained explicitly in terms of the system state-space matrices in the limiting singular case where the measurement noise tends to zero. Simple expressions, in closed form, are obtained for the Kalman gain matrix both for uniform and nonuniform rank systems and the explicit eigenstructure of the Kalman filter closed loop matrix is derived. The minimum error covariance matrices of the a priori and a posteriori filtered estimates are obtained using this special eigenstructure, and a remarkably different behavior of the solution in the minimum- and nonminimum-phase cases is found.     

On asymptotically efficient solutions for a class of supervisorycontrol problems

We consider restricted sequences of forbidden-string and forbidden-state problems and show the limiting problem instance defined by a sequence has a solution if and only if each problem instance in the sequence has a solution. Furthermore, the supervisor that enforces the desired behavior in the limiting problem instance also enforces the desired behavior in each problem instance in the sequence. In essence, the storage requirement for supervisory control of each instance is bounded above by that of the limiting problem instance. In addition, if the limiting problem instance is available, a similar observation can be made regarding the complexity of supervisory control. This observation coupled with recent results on the solvability of the supervisory control problem involving certain classes of nonregular behaviors and finite-representations of infinite-state systems is suggested as an effective procedure to combat the computational and storage issues in supervisory control. We conclude with an incomplete list of future research directions     

Performance limitations in the robust servomechanism problem for discrete-time LTI systems

Fundamental limitations for error tracking/regulation are obtained for the robust servomechanism problem (RSP) for a sampled system. In studying this problem, the cheap control problem for a multi-input/multi-output discrete time system is considered, and explicit expressions are obtained for the limiting steady state solution of its associated algebraic Riccati equation (ARE), as the weight on the control energy tends to zero. Application of these results is then made to obtain explicit expressions for the limiting performance costs associated with error tracking/regulation in the RSP. These limitations can be characterized by the system order, the dimension of the outputs, the number of the system's transmission zeros and the location of the system's nonminimum phase transmission zeros.     

Explicit solutions to the singular discrete finite-time linear estimation problem

The problem of optimal filtering of linear discrete-time processes is considered. It is assumed that the driving and measurement noise signals are stationary white, and that the number of measurements is finite. Explicit expressions for the filtering error covariance and the corresponding Kalman gain matrices are obtained for the limiting case where the measurement noise intensity tends uniformly to zero. Similar expressions are derived for the limiting singular smoothing problem with finite number of measurements.     

Deadbeat control and tracking of discrete-time systems

A new approach for forcing the state of a linear discrete-time system to zero in a minimum number of steps is discussed. The problem is formulated as the solution to a steady-state optimal control problem with no cost on the control. This problem in turn is set up as the solution to an associated eigenvalue problem. No special assumption on the open loop system matrix and/or the ratio of the number of states to controls is required. Stable numerical techniques are presented for solving for the feedback gain. Robust deadbeat tracking is also discussed.     

Resilient linear filtering of uncertain systems

The problem of resilient linear filtering for a class of linear continuous-time systems with norm-bounded uncertainties is investigated. We have considered additive filter gain variations to reflect the imprecision in filter implementation. The design problem of resilient linear filter is formulated as a convex optimization problem over linear matrix inequalities. As a limiting procedure, the case of resilient Kalman filter is derived. All the developed results are conveniently extended to the case of multiplicative filter gain variations. Simulation studies are carried out to support the theoretical findings.     

一种实用的模型自适应MPC策略

在丙烯精馏塔的控制中,由于被控对象的非线性特性,采用线性模型的模型预测控制器难以保持良好的控制性能。本文提出基于系统稳态模型的模型自适应MPC策略,利用稳态模型在不同操作点上被控变量对操纵变量及扰动变量的相对变化率的变化,来刷新RMPCT控制器中各通道的模型增益。在模型输出对输入的相对变化率的计算中,使用主操作区间内的变化率以替代实际操作点的变化率,并采用设定模型变化域和控制模型变化频度的方法,以解决模型变化过大,与模型变换周期和RMPCT控制周期不协调而引起的系统不稳定等问题。实际投运效果表明:采用该控制策略,塔顶、塔底温度控制偏差与传统RMPCT比下降了一个数量级,有利于稳定与提高产品质量。     

Robust ℒ︁2‐gain feedforward control of uncertain systems using dynamic IQCs

We consider the problem of robust ??₂‐gain disturbance feedforward control for uncertain systems described in the standard LFT form. We use integral quadratic constraints (IQCs) for describing the uncertainty blocks in the system. For technical reasons related to the feedforward problem, throughout the paper, we work with the duals of the constraints involved in robustness analysis using IQCs. We obtain a convex solution to the problem using a state‐space characterization of nominal stability that we have developed recently. Specifically, our solution consists of LMI conditions for the existence of a feedforward controller that guarantees a given ??₂‐gain for the closed‐loop system. We demonstrate the effectiveness of using dynamic IQCs in robust feedforward design through a numerical example. Copyright © 2008 John Wiley & Sons, Ltd.     

A linear distributed filter inspired by the Markovian jump linear system filtering problem

In this paper we introduce a consensus-based distributed filter, executed by a sensor network, inspired by the Markovian jump linear system filtering theory. We show that the optimal filtering gains of the Markovian jump linear system can be used as an approximate solution of the optimal distributed filtering problem. This parallel allows us to interpret each filtering gain corresponding to a mode of operation of the Markovian jump linear system as a filtering gain corresponding to a sensor in the network. The approximate solution can be implemented distributively and guarantees a quantifiable level of performance.     

The simultaneous optimization problem for sensitivity and gainmargin

The combined sensitivity and gain margin problem for single-input single-output linear systems is formulated and solved using a complex function interpolation technique. It is proved that this problem always has a real rational solution provided it is solvable in the complex irrational sense. The sensitivity minimization problem subject to a gain margin constraint and its dual problem are also considered. The range of the gain margin constraint is given subject to which the optimal constrained sensitivity is identical with the optimal unconstrained sensitivity. It is shown that, not unexpectedly, the gain margin maximization conflicts with the sensitivity minimization for a nonminimum phase plant     

A novel method of periodic orbit computation in circular restricted three-body problem

Periodic orbits are fundamental keys to understand the dynamical system of circular restricted three-body problem, and they play important roles in practical deep-space exploration. Current methods of periodic orbit computation need a high-order analytical approximate solution to start the iteration process, thus making the computation complicated and limiting the types of periodic orbits that can be obtained. By utilizing the symmetry of the restricted three-body problem, a special kind of flow function is...     

Pseudo-state approach to observer based optimal discrete-timeestimation

A frequency-domain-based solution of the optimal linear estimation problem is obtained using a polynomial system representation. The H₂ optimal filter can be realized using an observer based structure estimating pseudo-state variables that are fed back through a dynamic gain control block. There are normally fewer pseudo-states than the total order of the system. The dynamic gain for the model based observer is computed from diophantine equations and the observer provides both pseudo state and output estimates. The approach can deal with improper plant transfers and singular measurements     

Optimization of a Flow Shop System of Initially Controllable Machines

We consider an optimization problem for deterministic flow shop systems of traditional machines with service costs penalizing small service times. A regular completion-time cost is also included so as to complete jobs as early as possible. The service times are assumed to be initially controllable, i.e., they are set at the startup time. Assuming convexity of the cost functions, we formulate a convex optimization problem after linearization of the max constraints. The numeric solution of this problem demands a large memory limiting the solvable system sizes. In order to relieve the memory bottleneck, some waiting characteristics of jobs served in fixed-service-time flow shop systems are exploited to result in a simpler equivalent convex optimization problem. These characteristics and the benefit of CNC machines are demonstrated in a numerical example. We also show that the simplifications result in significant improvements in solvable system sizes and solution times.      

On nonlinear H∞ control under sampled measurements

This paper deals with the nonlinear H_∞ control problem with sampled measurement feedback. This problem has already been studied in Suzuki et al. (1995), where, using a certainty equivalence principle, a control solution involving a state estimator with a linear injection gain is proposed. Using the same general framework, we propose a more refined estimator with a nonlinear injection gain. This gain is shown to be connected to a periodic solution of a Hamilton-Jacobi inequality with jumps     

L1 synthesis of a static output controller for positive systems by LMI iteration

An approach to find a static output feedback gain that makes the feedback system positive and minimizes the L₁ gain is proposed. The problem of finding a static output feedback gain has 3 aspects: stabilizing the system, making the system positive, and then minimizing the L₁ gain. Each subproblem is described by bilinear matrix inequality with respect to the feedback gain and the Lyapunov matrix or vector. Linear matrix inequality (LMI) that is sufficient to satisfy bilinear matrix inequality is derived using a convex‐concave decomposition, and the feedback gain sequence is calculated by an iterative solution of LMI. The sequence of the upper bounds on the design parameter is guaranteed to be monotonically nonincreasing for each algorithm. Similarly, 2 other LMIs are derived for each subproblem using another convex‐concave decomposition and PK iteration. The effectiveness of these algorithms is illustrated via several numerical examples.     

基于LPV的航空发动机鲁棒变增益控制

随着我国现代化进程的不断加快,航天航空技术标准越来越高,对于航空发动机运转工况的鲁棒性和适应性提出了更高的要求。传统的航空发动机变增益设计步骤繁琐,不能将发动机置于整个航空器的运转去考虑设计,使发动机变增益缺乏相应的稳定性和适应性,易出现系统问题。为此,提出一般基于LPV的航空发动机鲁棒变增益控制系统,依据航空发动机结构参数,考虑到航空器在空中负载特性,计算出新的约束极点 模糊变增益,在航空器发动机工作范围连续增益,避免了传统增益切换情况,在转速控制上确定误差等因素,将非线性控制设计分解为多个线性子问题,使航空器控制系统能够沿着LPV参数轨迹保持良好的运转,保持稳定性能。仿真实验证明,提出的基于LPV的航空发动机鲁棒变增益控制系统控制效果优于传统方法,在航空器发动机转速改变时,控制精度能够满足要求 ,改变航空器负载时,有效对目标进行变增益控制。提出的控制方法对航空发动机鲁棒变增益控制问题提供了新的解决办法,具有较大应用价值。     

满意滤波在航迹辨识中的应用

利用线性矩阵不等式(LMI,linear matrix inequalities)方法,首先研究了Kalman滤波 的稳态误差方差和滤波增益的解法,并根据满意控制的思想,提出了具有误差方差上界约束的 滤波问题,然后研究了误差方差上界约束下对系统测量噪声具有最大容许强度的满意鲁棒滤 波.本文的研究成果已被应用于某型C3I系统中的航迹辨识.     

A rigorous solution of the infinite time interval LQ problem with constant state tracking

A linear quadratic constant state tracking problem is considered over an infinite time horizon. It is shown that the solution cannot be obtained as a limit from a finite time horizon problem, as in general the limiting problem is ill-posed. To obtain a rigorous solution, the problem is split in two natural well-posed subproblems. One optimal control problem addresses the transient and the other optimal control problem concerns the steady state behavior. It is shown that the transient problem and the steady state problem are solved by the same control law.