Search

Spacecraft reorientation with attitude pointing constraints and the uncertainty of inertia and external disturbance is investigated in this paper. By introducing the potential function into the design of non-singular fixed-time sliding mode surface, the proposed controller can achieve fixed-time convergence and the convergence time of attitude error can be predetermined by selecting appropriate parameters. Meanwhile, the attitude pointing constraints can be satisfied all the time. The designed sliding surface and potential function have two equilibrium points, which guarantees the unwinding-free performance. Furthermore, an adaptive sliding mode control scheme is developed to handle the system lumped disturbance. Rigorous Lyapunov analyses are employed to ensure practical fixed-time closed-loop stability in the presence of system disturbance uncertainties and attitude pointing constraints. Therefore, the fixed-time stability, the feasibility of attitude pointing constraints and disturbance rejection are achieved simultaneously with the proposed controller. Numerical simulations are provided to demonstrate the effectiveness and superiority of the proposed method. • The PF is introduced into the design of the non-singular fixed-time sliding surface. • The fixed-time convergence and attitude pointing constraints are satisfied. • By introducing the sign function, the unwinding phenomenon is prevented. [ABSTRACT FROM AUTHOR]

Highlights • A number of Mahalanobis stationary kernels were proposed for credit risk evaluation. • The Mahalanobis kernels are suitable for SVM framework, with a theoretical illustration. • Stationary kernels with Mahalanobis distance outperform the stationary kernels with various distance measures, describing the behavior of datasets more properly. • The proposed kernels can outperform state-of-the-art models in credit risk evaluations, generating more robust models in terms of specificity and sensitivity. • The satisfying performance of the proposed kernels in credit risk evaluation may provide potential options in risk modeling. Abstract This paper proposed Mahalanobis distance induced kernels in Support Vector Machines (SVMs) with applications in credit risk evaluation. We take a particular interest in stationary ones. Compared to traditional stationary kernels, Mahalanobis kernels take into account on feature's correlation and can provide a more suitable description on the behavior of the data sets. Results on real world credit data sets show that stationary kernels with Mahalanobis distance outperform the stationary kernels with various distance measures and they can also compete with frequently used kernels in SVM. The superior performance of our proposed kernels over other classical machine learning methods and the successful application of the kernels in large scale credit risk evaluation problems may imply that we have proposed a new class of kernels appropriate for credit risk evaluations. [ABSTRACT FROM AUTHOR]

Volatility modeling is crucial for risk management and asset allocation; this is an influential area in financial econometrics. The central requirement of volatility modeling is to be able to forecast volatility accurately. The literature review of volatility modeling shows that the approaches of model averaging estimation are commonly used to reduce model uncertainty in order to achieve a satisfactory forecasting reliability. However, those approaches attempt to forecast more reliable volatilities by integrating all forecasting outcomes equally from several volatility models. Forecasting patterns generated by each model may be similar. This may cause redundant computation without improving forecasting reliability. The proposed multivariate volatility modeling method which is called the fuzzy-method-involving multivariate volatility model (abbreviated as FMVM) classifies the individual models into smaller scale clusters and selects the most representative model in each cluster. Hence, repetitive but unnecessary computational burden can be reduced, and forecasting patterns from representative models can be integrated. The proposed FMVM is benchmarked against existing multivariate volatility models on forecasting volatilities of Hong Kong Hang Seng Index constituent stocks. Numerical results show that it can obtain relatively lower forecasting errors with less model complexity. [ABSTRACT FROM AUTHOR]

Highlights: [•] We study the signal enhancement problem. [•] We propose a multi-filter system. [•] We study the multi-criteria design of the system. [•] We propose an implementation of the system with an FPGA embedded system. [ABSTRACT FROM AUTHOR]

Abstract: Feedforward neural networks are the most commonly used function approximation techniques in neural networks. By the universal approximation theorem, it is clear that a single-hidden layer feedforward neural network (FNN) is sufficient to approximate the corresponding desired outputs arbitrarily close. Some researchers use genetic algorithms (GAs) to explore the global optimal solution of the FNN structure. However, it is rather time consuming to use GA for the training of FNN. In this paper, we propose a new optimization algorithm for a single-hidden layer FNN. The method is based on the convex combination algorithm for massaging information in the hidden layer. In fact, this technique explores a continuum idea which combines the classic mutation and crossover strategies in GA together. The proposed method has the advantage over GA which requires a lot of preprocessing works in breaking down the data into a sequence of binary codes before learning or mutation can apply. Also, we set up a new error function to measure the performance of the FNN and obtain the optimal choice of the connection weights and thus the nonlinear optimization problem can be solved directly. Several computational experiments are used to illustrate the proposed algorithm, which has good exploration and exploitation capabilities in search of the optimal weight for single hidden layer FNNs. [Copyright &y& Elsevier]

Ball bonding inspections with human vision are essential in manufacturing processes of semiconductors devices and integrated circuits (ICs). The inspections are an intensive task which involves human labours to detect poor bonds. Prolonged visual inspections cause poor inspection integrity due to eye-fatigue. However, inspections nowadays are mostly conducted manually by humans which cannot satisfy the demanding productions. Motivated by the extraordinary performance of machine learning for manufacturing inspections, a detection framework integrated with machine learning and human judgement is proposed to aid bonding inspections based on visual images. The detection framework is incorporated with the convolution neural network (CNN), support vector machine (SVM) and circle hough transform algorithm (CHT); human judgement is only used when the detection uncertainty is below the threshold. The novel machine learning integration is proposed on the detection framework to improve the generalization capabilities. The CNN architecture is redeveloped by incorporating with the SVM which is generally more effective than the fully connected network in the classical CNN. Also a novel training function is proposed based on the CHT to ensure the inspection reliability; the function not only takes into account real image captures, but also locates important features using pattern analysis of the ball bondings. Experimental results show that significantly better classifications can be achieved by the proposed framework compared with the classical CNN and other commonly used classifiers. Only the machine learning determinations below the threshold are reassessed by human judgements. • Ball bonding inspections with human vision are essential in manufacturing ICs. • Inspections nowadays are mostly conducted manually by humans. • A detection framework with machine learning and human judgement. • Results show better classifications achieved by the framework. • Human judgement is used under threshold; Significant human resource is saved. [ABSTRACT FROM AUTHOR]