二维随机FM-II系统的状态估计

This paper is concerned with state estimation of two-dimensional (2-D) discrete stochastic systems. First, 2-D discrete stochastic system model is established by extending system matrices of the well-known Fornasini-Marchesini's second model into stochastic matrices. Each element of these stochastic matrices is second-order weakly stationary white noise sequences. Secondly, a linear and unbiased full-order state estimation problem for 2-D discrete linear stochastic model is formulated. Two estimation problems considered are the designs for the mean-square bounded estimation error and for the mean-square stochastic version of the suboptimal H∞ estimator, respectively. Our results can be seen as extensions of the 2-D linear deterministic case. Finally, illustrative examples are provided.     

输出信号量化的双率采样系统参数辨识方法

针对双率采样和信号量化(signal quantization)[BFQB]的控制系统,采用随机重复性试验测量信息,提出基于辅助模型的双率采样量化控制系统辨识方法.分析了在随机重复试验和放松估计误差方差条件下,双率采样量化系统的模型特征并给出了分两步辨识的策略,推导了进行参数辨识所满足的持续激励条件,并给出了基于辅助模型的双率采样量化控制系统量化辨识递推算法;接着分析了所给出量化辨识递推算法的收敛性,得到了双率采样量化系统参数估计误差上界的计算式,最后数字仿真验证了该算法及结论的有效性.     

Stochastic gradient algorithm for a dual-rate Box-Jenkins model based on auxiliary model and FIR model

Based on the work in Ding and Ding(2008),we develop a modifed stochastic gradient(SG)parameter estimation algorithm for a dual-rate Box-Jenkins model by using an auxiliary model.We simplify the complex dual-rate Box-Jenkins model to two fnite impulse response(FIR)models,present an auxiliary model to estimate the missing outputs and the unknown noise variables,and compute all the unknown parameters of the system with colored noises.Simulation results indicate that the proposed method is efective.     

非均匀采样数据系统的一种辨识方法

阐述了非均匀采样方案,推导了非均匀多率采样系统的状态空间模型,进一步获得了对应的传递函数模型.为解决辨识模型信息向量中存在未知变量的问题,使用辅助模型技术,用辅助模型的输出代替系统的未知变量,进而提出了非均匀采样数据系统的辅助模型随机梯度辨识算法.为了提高算法收敛速度和改善参数估计精度,在算法中引入遗忘因子,给出了相应的辅助模型带遗忘因子随机梯度算法.仿真结果表明,引入遗忘因子后,算法的收敛速度加快,参数估计精度提高.     

反馈非线性系统随机梯度辨识算法及其收敛性

反馈非线性受控自回归系统是由前向通道的受控自回归模型和反馈通道的静态非线性构成, 这类系统经过参数化后得到双线性参数辨识模型. 本文通过对辨识模型中双线性参数乘积项进行分解, 基于梯度搜索原理, 提 出了反馈非线性系统的随机梯度辨识算法. 为了改善随机梯度算法的收敛速度, 引入遗忘因子, 文章给出了遗忘因子随机梯度算法, 利用随机过程理论, 建立了随机梯度算法的参数估计收敛定理, 证明了算法的收敛性. 最后, 通过数值仿真验证了算法的有效性.     

Auxiliary model multiinnovation stochastic gradient parameter estimation methods for nonlinear sandwich systems

This article studies the identification problem of the nonlinear sandwich systems. For the sandwich system, because there are inner variables which cannot be measured in the information vector of the identification models, it is difficult to identify the nonlinear sandwich systems. In order to overcome the difficulty, an auxiliary model is built to predict the estimates of inner variables by means of the output of the auxiliary model. For the purpose of employing the real‐time observed data, a cost function with dynamical data is constructed to capture on‐line information of the nonlinear sandwich system. On this basis, an auxiliary model stochastic gradient identification approach is proposed based on the gradient optimization. Moreover, an auxiliary model multiinnovation stochastic gradient estimation method is developed, which tends to enhance estimation accuracy by introducing more observed data dynamically. The numerical simulation is provided and the simulation results show that the proposed auxiliary model identification method is effective for the nonlinear sandwich systems.     

A Deformable Template Model, with Special Reference to Elliptical Templates

This paper suggests a high-level continuous image model for planar star-shaped objects. Under this model, a planar object is a stochastic deformation of a star-shaped template. The residual process, describing the difference between the radius-vector function of the template and the object, is allowed to be non-stationary. Stationarity is obtained by a time change. A parametric model for the residual process is suggested and straightforward parameter estimation techniques are developed. The deformable template model makes it possible to detect pathologies as demonstrated by an analysis of a data set of cell nuclei from a benign and a malignant tumour, using stochastic deformations of ellipses.     

双输入多率系统辅助模型增广随机梯度算法

针对有色噪声干扰的双输入多率系统,为解决辨识模型信息向量中存在未知变量和不可测噪声项的问题,结合辅助模型思想和递推增广随机梯度算法的优点,用辅助模型的输出代替系统的未知变量,用估计残差代替信息向量中的不可测噪声项,进而提出了双输入多率系统的辅助模型增广随机梯度算法。为了提高辨识算法的收敛速度和改善参数估计精度,在算法中引入遗忘因子,得到相应的辅助模型带遗忘因子增广随机梯度算法。仿真实例说明,引入遗忘因子,能加快算法的收敛性,提高参数估计精度。     

气固催化反应L—H—H—W型等温动力学模型自动识别及图示

本文依据L-H-H-W机理采用组合方式对气固催化反应的动力学模型进行自动识别以及对找出模型进行三维立体图示诊断,建立了GSKM系统。GSKM系统主要包括模型寻优、固定模型参数估值和图示诊断。我们对从等温数据建立反应速率与温度、压力的关系作了较有实用意义的探讨。应用本系统,我们对文献上所提供的数据,进行了动力学模型自动识别及图示诊断,得到了较为满意的结果。     

基于ARMA的微惯性传感器随机误差建模方法

针对微惯性传感器随机误差建模效果不理想,影响微惯性组合导航系统性能的问题,提出了采用自回归滑动平均(ARMA)对微惯性传感器随机误差进行建模的方法。通过对随机误差模型应用于微惯性器件误差建模的深入分析,将Yule-Walker方程引入线性预测问题中,实现AR功率谱密度的估计,建立了基于随机过程有理功率谱密度的ARMA模型建立方法,并给出了ARMA建模准确性的LDA验证准则。通过微惯性传感器实测数据,对随机误差建模方法进行了有效性验证。该方法为微惯性器件的随机误差建模和分析提供了一种新的途径。     

A Stochastic Deterministic Model Predictive Control: Its Accuracy

The loss in control quality due to the use of a suboptimal deterministic model predictive control for a stochastic object is evaluated. Special attention is paid to the estimation of prediction bias in stochastic deterministic-predictive control. The results are illustrated by an example of a stochastic deterministic model predictive control for missing data.     

Probabilistic prediction methods for nonlinear systems with application to stochastic model predictive control

The performance of modern control methods, such as model predictive control, depends significantly on the accuracy of the system model. In practice, however, stochastic uncertainties are commonly present, resulting from inaccuracies in the modeling or external disturbances, which can have a negative impact on the control performance. This article reviews the literature on methods for predicting probabilistic uncertainties for nonlinear systems. Since a precise prediction of probability density functions comes along with a high computational effort in the nonlinear case, the focus of this article is on approximating methods, which are of particular relevance in control engineering practice. The methods are classified with respect to their approximation type and with respect to the assumptions about the input and output distribution. Furthermore, the application of these prediction methods to stochastic model predictive control is discussed including a literature review for nonlinear systems. Finally, the most important probabilistic prediction methods are evaluated numerically. For this purpose, the estimation accuracies of the methods are investigated first and the performance of a stochastic model predictive controller with different prediction methods is examined subsequently using multiple nonlinear systems, including the dynamics of an autonomous vehicle.     

基于非稳态随机过程的近红外反射率鲁棒估计算法

反射率估计在计算机视觉、计算机图形学等领域具有重要作用.为了精确获取反射率,提出一种基于非稳态随机过程的近红外反射率鲁棒估计算法(RENA).该算法以Kinect二代传感器采集结果计算初始反射率,并建立反射率加性噪声模型,同时提出光照度鲁棒估计的概念,简化反射率图像非稳态随机过程模型.实验表明,RENA算法的反射率估计结果优于其他去噪算法,适用于室内场景的反射率图像高精度估计.     

On the indirect approaches for CARMA model identification

In this paper, we study the problem of reconstructing a continuous-time (CT) model from an identified discrete-time (DT) model for a continuous-time stochastic process. We present a new necessary and sufficient condition for the existence of the solution. We also show that the solution is unique if it exists. Our results are useful in modeling multivariable processes as well. These results are then used to develop an algorithm where the intermediate discrete-time model estimation is not necessary. The performance of our algorithm is illustrated using numerical simulations.     

Three‐stage multi‐innovation parameter estimation for an exponential autoregressive time‐series model with moving average noise by using the data filtering technique

This paper studies the data filtering‐based identification algorithms for an exponential autoregressive time‐series model with moving average noise. By means of the data filtering technique and the hierarchical identification principle, the identification model is transformed into three sub‐identification (Sub‐ID) models, and a filtering‐based three‐stage extended stochastic gradient algorithm is derived for identifying these Sub‐ID models. In order to improve the parameter estimation accuracy, a filtering‐based three‐stage multi‐innovation extended stochastic gradient (F‐3S‐MIESG) algorithm is developed by using the multi‐innovation identification theory. The simulation results indicate that the proposed F‐3S‐MIESG algorithm can work well.     

Fault detection model‐based controller for process systems

This paper develops a model‐based control system for fault detection and controller reconfiguration using stochastic model predictive control (MPC). The system can determine online the optimal control actions, detect faults quickly, and reconfigure the controller accordingly. Such a system can perform its function correctly in the presence of internal faults. A fault detection model based (FDMB) controller consists of two main parts: the first is fault detection and diagnosis (FDD) and the second is controller reconfiguration (CR). Systems subject to such abrupt failures are modeled as stochastic hybrid systems with variable‐structure. This paper deals with three challenging issues: design of the fault‐model set; estimation of hybrid multiple models; and stochastic MPC of hybrid multiple models. For the first issue, we propose a simple scheme for designing a fault model set based on random variables. For the second issue, we consider and select a fast and reliable FDD system applied to the above model set. Finally, we develop a stochastic MPC scheme for multiple model CR with soft switching signals based on the weighted probabilities of the outputs of different models. Simulations for the proposed FDMB controller are illustrated and analyzed. Copyright © 2011 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

Estimation of First Order Plus Dead Time and Second Order Plus Dead Time models from noisy step response data

This paper proposes the estimation of first-order-plus-dead-time (FOPDT) and second-order-plus-dead-time (SOPDT) models from noisy step response data. The model parameters are estimated by computation of areas, which makes it robust in the presence of stochastic disturbances in the step response data. The efficiency of the methodology is extensively tested in various numerical examples as well as in real-life experimental tests. The results—comparing our proposed estimation method with some other methods—suggest that the novel algorithm can be used with noisy step response data and adequately approximates high-order systems. Moreover, it does not require any system identification expertise, making it readily accessible for the nonexperienced user in industrial practice. The method is successfully validated for overdamped, reasonably underdamped, as well as highly oscillatory processes, hence offering a comprehensive estimation method.     

Point Light Source Estimation from Two Images and Its Limits

In this paper, the performance of parameter estimation of a single static distant point light source from two video images is analyzed in terms of estimation theory. The illumination parameters are the intensity and direction of the light source.In the first part of this paper, estimators from the literature are reviewed. Most recent estimators evaluate as input data two video images as well as the 3D shape and the 3D motion of the visible moving objects.In the second part of the paper, the performance of these recent methods is analyzed. The input data to estimation as well as the inherent input data errors are described by a stochastic observation model. Based on this model, the performance is analyzed regarding the Cramér-Rao theoretical lower bound of estimation error variances. The bound is derived for a variety of cases of scene illumination, object motion and errors in input data. For simplification purpose, the bound is valid only for object motions with the rotation axis lying in the image plane. The analysis shows in which cases which estimation accuracy can be expected with current methods.Finally, a comparison of the bound with one of the recent estimators shows that recent estimators are suboptimal in case of errors in the 3D shape of the objects. In future work, the stochastic observation model presented in this paper can be used to improve illumination estimation.     

Fractionally Delayed Kalman Filter

The conventional Kalman filter is based on the assumption of non-delayed measurements. Several modifications appear to address this problem, but they are constrained by two crucial assumptions: 1) the delay is an integer multiple of the sampling interval, and 2) a stochastic model representing the relationship between delayed measurements and a sequence of possible non-delayed measurements is known. Practical problems often fail to satisfy these assumptions, leading to poor estimation accuracy and frequent track-failure. This paper introduces a new variant of the Kalman filter, which is free from the stochastic model requirement and addresses the problem of fractional delay.The proposed algorithm fixes the maximum delay(problem specific), which can be tuned by the practitioners for varying delay possibilities. A sequence of hypothetically defined intermediate instants characterizes fractional delays while maximum likelihood based delay identification could preclude the stochastic model requirement. Fractional delay realization could help in improving estimation accuracy. Moreover, precluding the need of a stochastic model could enhance the practical applicability. A comparative analysis with ordinary Kalman filter shows the high estimation accuracy of the proposed method in the presence of delay.