Model-reference robust tuning of 2DoF PI controllers for first- and second-order plus dead-time controlled processes

The aim of this paper is to present a robust tuning method for two-degree-of-freedom (2DoF) proportional integral (PI) controllers. This is based on the use of a model reference optimization procedure with servo and regulatory target closed-loop transfer functions for first- and second-order plus dead-time (FOPDT, SOPDT) controlled process models. The designer is allowed to deal with the performance/robustness trade-off of the closed-loop control system by specifying the desired robustness level by selecting a maximum sensitivity in the range from 1.4 to 2.0. In addition, a smooth servo/regulatory performance combination is obtained by forcing both closed-loop transfer functions to perform as closely as possible to non-oscillatory dynamic targets. A unified set of controller tuning equations is provided for FOPDT and SOPDT models with normalized dead-times from 0.1 to 2.0 that guarantees the achievement of the design robustness level. The robustness of the control system is analyzed as well as the robustness–fragility and performance–fragility of the optimized controllers. Comparative examples show the effectiveness of the proposed tuning method. The exact achievement of the control system robustness target for all the controlled process models considered (first- and second-order) is one of the distinctive characteristics of the proposed model reference robust tuning (MoReRT) method.     

Robustness of AI-based prognostic and systems health management

Prognostic and systems Health Management (PHM) is an integral part of a system. It is used for solving reliability problems that often manifest due to complexities in design, manufacturing, operating environment and system maintenance. For safety-critical applications, using a model-based development process for complex systems might not always be ideal but it is equally important to establish the robustness of the solution. The information revolution has allowed data-driven methods to diffuse within this field to construct the requisite process (or system models) to cope with the so-called big data phenomenon. This is supported by large datasets that help machine-learning models achieve impressive accuracy. AI technologies are now being integrated into many PHM related applications including aerospace, automotive, medical robots and even autonomous weapon systems. However, with such rapid growth in complexity and connectivity, a systems’ behaviour is influenced in unforeseen ways by cyberattacks, human errors, working with incorrect or incomplete models and even adversarial phenomena. Many of these models depend on the training data and how well the data represents the test data. These issues require fine-tuning and even retraining the models when there is even a small change in operating conditions or equipment. Yet, there is still ambiguity associated with their implementation, even if the learning algorithms classify accordingly. Uncertainties can lie in any part of the AI-based PHM model, including in the requirements, assumptions, or even in the data used for training and validation. These factors lead to sub-optimal solutions with an open interpretation as to why the requirements have not been met. This warrants the need for achieving a level of robustness in the implemented PHM, which is a challenging task in a machine learning solution.This article aims to present a framework for testing the robustness of AI-based PHM. It reviews some key milestones achieved in the AI research community to deal with three particular issues relevant for AI-based PHM in safety-critical applications: robustness to model errors, robustness to unknown phenomena and empirical evaluation of robustness during deployment. To deal with model errors, many techniques from probabilistic inference and robust optimisation are often used to provide some robustness guarantee metric. In the case of unknown phenomena, techniques include anomaly detection methods, using causal models, the construction of ensembles and reinforcement learning. It elicits from the authors’ work on fault diagnostics and robust optimisation via machine learning techniques to offer guidelines to the PHM research community. Finally, challenges and future directions are also examined; on how to better cope with any uncertainties as they appear during the operating life of an asset.     

Assessment of diagnostic procedures in symmetrical nonlinear regression models

The aim of this paper is to derive diagnostic procedures based on case-deletion model for symmetrical nonlinear regression models, which complements Galea et al. (2005) that developed local influence diagnostics under some perturbation schemes. This class of models includes all symmetric continuous distributions for errors covering both light- and heavy-tailed distributions such as Student-t, logistic-I and -II, power exponential, generalized Student-t, generalized logistic and contaminated normal, among others. Thus, these models can be checked for robustness to outliers in the response variable and diagnostic methods may be a useful tool for an appropriate choice. First, an iterative process for the parameter estimation as well as some inferential results are presented. Besides, we present the results of a simulation study in which the characteristics of heavy-tailed models are evaluated in the presence of outliers. Then, we derive some diagnostic measures such as Cook distance, W-K statistic, one-step approach and likelihood displacement, generalizing results obtained for normal nonlinear regression models. Also, we present simulation studies that illustrate the behavior of diagnostic measures proposed. Finally, we consider two real data sets previously analyzed under normal nonlinear regression models. The diagnostic analysis indicates that a Student-t nonlinear regression model seems to fit the data better than the normal nonlinear regression model as well as other symmetrical nonlinear models in the sense of robustness against extreme observations.     

Robust identification of process models from plant data

A precursor to any advanced control solution is the step of obtaining an accurate model of the process. Suitable models can be obtained from phenomenological reasoning, analysis of plant data or a combination of both. Here, we will focus on the problem of estimating (or calibrating) models from plant data. A key goal is to achieve robust identification. By robust we mean that small errors in the hypotheses should lead to small errors in the estimated models. We argue that, in some circumstances, it is essential that special precautions, including discarding some part of the data, be taken to ensure that robustness is preserved. We present several practical case studies to illustrate the results.     

Assessment of local influence in elliptical linear models with longitudinal structure

The aim of this paper is to derive local influence curvatures under various perturbation schemes for elliptical linear models with longitudinal structure. The elliptical class provides a useful generalization of the normal model since it covers both light- and heavy-tailed distributions for the errors, such as Student-t, power exponential, contaminated normal, among others. It is well known that elliptical models with longer-than-normal tails may present robust parameter estimates against outlying observations. However, little has been investigated on the robustness aspects of the parameter estimates against perturbation schemes. We use appropriate derivative operators to express the normal curvatures in tractable forms for any correlation structure. Estimation procedures for the position and variance-covariance parameters are also presented. A data set previously analyzed under a normal linear mixed model is reanalyzed under elliptical models. Local influence graphics are used to select less sensitive models with respect to some perturbation schemes.     

On robustness to noise of least squares based adaptive control

Current engineering practice for adaptive control schemes is to base the design on globally convergent schemes for simple plant models. An important class of such schemes uses least squares estimation of assumed simple input-output models and constructs the controller using the parameter estimates. This paper studies the robustness of such schemes to the presence of unmodelled plant coloured noise. Such noise is sometimes an adequate model for unmodelled plant dynamics.The theory of the paper makes a connection between the least squares parameter error equations and those associated with extended least squares using a posteriori noise estimates for which there are known global convergence results. For the case of adaptive minimum variance control of minimum phase plants, this connection permits stronger convergence results than those hitherto derived from the theory of extended least squares based on a priori noise estimates.     

Analysing the performance of bootstrap neural tests for conditional heteroskedasticity in ARCH-M models

The robustness against strongly non-linear forms for the conditional variance of tests for detecting conditional heteroskedasticity using both artificial neural network techniques and bootstrap methods combined, is analysed in the context of ARCH-M models. The size and the power properties in small samples of these tests are examined by using out Monte-Carlo experiments with various standard and non-standard models of conditional heteroskedasticity. The P value functions are explored in order to select particularly problematic cases. Graphical presentations, based on the principle of size correction, are used for presenting the true power of the tests, rather than a spurious nominal power as it is usually made in the literature. In addition, graphics linking the process dynamics with the heteroskedasticity forms are shown for analysing in which circumstances the neural networks are effective.     

An efficient method for fully automatic 3D digitization of unknown objects

Our goal is to develop a complete and automatic scanning strategy with minimum prior information about the object shape. We aim to establish a methodology for the automation of the 3D digitization process. The paper presents a novel approach to determine the Next Best View (NBV) for an efficient reconstruction of highly accurate 3D models. Our method is based on the classification of the acquired surfaces into Well Visible and Barely Visible combined with a best view selection algorithm based on mean shift, which avoids unreachable positions. Our approach is applicable to all kinds of range sensors. To prove the efficiency and the robustness of our method, test objects are first scanned manually by experts in 3D digitization from the VECTEO company. The comparison of results between manual and automatic scanning shows that our method is very efficient and faster than trained experts. The 3D models of the different objects are obtained with a strongly reduced number of acquisitions while moving efficiently the ranging device. The obtained results prove the effectiveness and the versatility of our 3D reconstruction approach for industrial applications.     

InRob: An approach for testing interoperability and robustness of real-time embedded software

Advances in digital technologies have contributed for significant reduction in accidents caused by hardware failures. However, the growing complexity of functions performed by embedded software has increased the number of accidents caused by software faults in critical systems. Moreover, due to the highly competitive market, software intensive subsystems are usually developed by different suppliers. Often these subsystems are required to interact with each other in order to provide a collaborative service. Testing approaches for subsystems integration support verification of the quality of service, focusing on the subsystems interfaces. The increasing complexity and tight coupling of real-time subsystems make integration testing unmanageable. The ad-hoc approach for testing is becoming less effective and more expensive. This article presents an integration testing approach denominated InRob, designed to verify the interoperability and robustness related to timing constraints of real-time embedded software. InRob guides the construction of services, based on formal models, aiming at the specifications of interoperability and robustness of test cases related to delays and time-outs of the messages exchanged in the interfaces of interconnected subsystems. The proposed formalism supports automatic test cases generation by verifying the relevant properties in the service behavioral model. As timing constraints are critical properties of aerospace systems, the feasibility of InRob is showed in the integration testing process of a telescope onboard in a satellite. The process is instantiated with existing testing tools and the case study is the software embedded in the telescope.     

Memetic Pareto Evolutionary Artificial Neural Networks to determine growth/no-growth in predictive microbiology

The main objective of this work is to automatically design neural network models with sigmoid basis units for binary classification tasks. The classifiers that are obtained achieve a double objective: a high classification level in the dataset and a high classification level for each class. We present MPENSGA2, a Memetic Pareto Evolutionary approach based on the NSGA2 multiobjective evolutionary algorithm which has been adapted to design Artificial Neural Network models, where the NSGA2 algorithm is augmented with a local search that uses the improved Resilient Backpropagation with backtracking—IRprop+ algorithm. To analyze the robustness of this methodology, it was applied to four complex classification problems in predictive microbiology to describe the growth/no-growth interface of food-borne microorganisms such as Listeria monocytogenes, Escherichia coli R31, Staphylococcus aureus and Shigella flexneri. The results obtained in Correct Classification Rate (CCR), Sensitivity (S) as the minimum of sensitivities for each class, Area Under the receiver operating characteristic Curve (AUC), and Root Mean Squared Error (RMSE), show that the generalization ability and the classification rate in each class can be more efficiently improved within a multiobjective framework than within a single-objective framework.     

面向图像识别的卷积神经网络鲁棒性研究进展

卷积神经网络是目前人工智能领域在图像识别与处理相关应用中的关键技术之一,广泛的应用使对其鲁棒性研究的重要性不断凸显。以往对于卷积神经网络鲁棒性的研究较为笼统,且多集中在对抗鲁棒性方面。这难以更深入地研究神经网络鲁棒性的发生机制,已经不适应人工智能的发展。引入神经科学的相关研究,提出了视觉鲁棒性的概念,通过研究神经网络模型与人类视觉系统的相似性,揭示了神经网络鲁棒性的内在缺陷。回顾了近年来神经网络鲁棒性的研究现状,并分析了神经网络模型缺乏鲁棒性的原因。神经网络缺乏鲁棒性体现在其对于微小扰动的敏感性,其原因在于神经网络会更倾向于学习人类难以感知的高频信息用于计算和推理。而这部分高频信息很容易被扰动所破坏,最终导致模型出现判断错误。传统鲁棒性的研究大多关注模型的数学性质,无法突破神经网络的天然局限性。视觉鲁棒性在传统鲁棒性的概念上进行拓展。传统鲁棒性概念衡量模型对于失真变形的图像样本的辨识能力,失真样本与原始干净样本在鲁棒模型上都能保持正确的输出。视觉鲁棒性衡量模型与人类判别能力的一致性。这需要将神经科学和心理学的研究方法、成果与人工智能相结合。回顾了神经科学在视觉领域的发展,讨论了认知心理...     

Weighted and extended total variation for image restoration and decomposition

In various information processing tasks obtaining regularized versions of a noisy or corrupted image data is often a prerequisite for successful use of classical image analysis algorithms. Image restoration and decomposition methods need to be robust if they are to be useful in practice. In particular, this property has to be verified in engineering and scientific applications. By robustness, we mean that the performance of an algorithm should not be affected significantly by small deviations from the assumed model. In image processing, total variation (TV) is a powerful tool to increase robustness. In this paper, we define several concepts that are useful in robust restoration and robust decomposition. We propose two extended total variation models, weighted total variation (WTV) and extended total variation (ETV). We state generic approaches. The idea is to replace the TV penalty term with more general terms. The motivation is to increase the robustness of ROF (Rudin, Osher, Fatemi) model and to prevent the staircasing effect due to this method. Moreover, rewriting the non-convex sublinear regularizing terms as WTV, we provide a new approach to perform minimization via the well-known Chambolle's algorithm. The implementation is then more straightforward than the half-quadratic algorithm. The behavior of image decomposition methods is also a challenging problem, which is closely related to anisotropic diffusion. ETV leads to an anisotropic decomposition close to edges improving the robustness. It allows to respect desired geometric properties during the restoration, and to control more precisely the regularization process. We also discuss why compression algorithms can be an objective method to evaluate the image decomposition quality.     

Non-linear time series clustering based on non-parametric forecast densities

The problem of clustering time series is studied for a general class of non-parametric autoregressive models. The dissimilarity between two time series is based on comparing their full forecast densities at a given horizon. In particular, two functional distances are considered: L¹ and L². As the forecast densities are unknown, they are approximated using a bootstrap procedure that mimics the underlying generating processes without assuming any parametric model for the true autoregressive structure of the series. The estimated forecast densities are then used to construct the dissimilarity matrix and hence to perform clustering. Asymptotic properties of the proposed method are provided and an extensive simulation study is carried out. The results show the good behavior of the procedure for a wide variety of nonlinear autoregressive models and its robustness to non-Gaussian innovations. Finally, the proposed methodology is applied to a real dataset involving economic time series.     

Idle speed control of a F1 racing engine

This paper presents the design and implementation of an idle speed regulator for the Toyota Formula 1 racing car. The control variables are the throttle opening and spark advance, while the controlled variable is the crank shaft speed. First, a set of linear models has been identified from experimental data. Then, a nominal estimated model has been used to synthesize an idle speed regulator with the H₂ approach, and its robustness properties have been tested both in the frequency domain and in simulation. The regulator, already implemented on the engine with fully satisfactory results, has been mounted on the 2003 F1 racing car.     

On robustness of constrained discrete-time systems to state measurement errors

In this note we show that robustness with respect to additive disturbances implies robustness with respect to state measurement errors and additive disturbances for a class of discrete-time closed-loop nonlinear systems. The main result is formulated in terms of input-to-state stability and includes the possible presence of input and state constraints. Moreover, the state feedback controllers are allowed to be discontinuous and set-valued and thus the result also applies to model predictive control laws.     

A discrete-time stable controller for an omni-directional mobile robot based on an approximated model

An adaptive controller based on multi-input fuzzy rules emulated networks (MIFRENs) is introduced for omni-directional mobile robot systems in the discrete-time domain without any kinematic or dynamic models. An approximated model for unknown systems is developed by using two MIFRENs with an online learning algorithm in addition to the stability analysis. The main theorem in this model is proposed to guarantee closed-loop performance and system robustness for all adjustable parameters inside MIFRENs. The system is validated by an experimental setup with a FESTO omni-directional mobile robot called Robotino^®. The proposed algorithm is shown to have superior performance compared to that of an algorithm that uses only an embedded controller. The advantage of the MIFREN initial setting is verified comparing its results with those of a controller that is based on neural networks.     

One-step minimum Hellinger distance estimation

It is well known now that the minimum Hellinger distance estimation approach introduced by Beran (Beran, R., 1977. Minimum Hellinger distance estimators for parametric models. Ann. Statist. 5, 445-463) produces estimators that achieve efficiency at the model density and simultaneously have excellent robustness properties. However, computational difficulties and algorithmic convergence problems associated with this method have hampered its application in practice, particularly when the method is applied to models with high-dimensional parameter spaces. A one-step minimum Hellinger distance (MHD) procedure is investigated in this paper to overcome computational drawbacks of the fully iterative MHD method. The idea is to start with an initial estimator, and then iterate the Newton-Raphson equation once related to the Hellinger distance. The resulting estimator can be considered a one-step MHD estimator. We show that the proposed one-step MHD estimator has the same asymptotic behavior as the MHD estimator, as long as the initial estimators are reasonably good. Furthermore, our theoretical and numerical studies also demonstrate that the proposed one-step MHD estimator also retains excellent robustness properties of the MHD estimators. A real data example is analyzed as well.