Unified Fuzzy Control of High-Order Nonlinear Systems With Multitype State Constraints.

This article presents a unified adaptive fuzzy control approach for high-order nonlinear systems (HONSs) with multitype state constraints. Existing methods always require the upper and lower constraint boundaries are strictly positive and negative functions (or constants), respectively, which is often inconsistent with the actual constraints. In this article, "multitype state constraint" means that the upper and lower constraint boundaries include multiple types, such as both being strictly positive (or negative), sometime be positive or negative, and so on (cases ①-⑥). By designing a unified mapping function (UMF), the multitype state constraints are processed under removal the feasibility conditions (FCs). Furthermore, a technical design makes the proposed method also applicable to unconstrained HONSs without changing the control structure. By means of a fuzzy-logic system (FLS) and fixed-time stability theory (FTST), the proposed algorithm can ensure that the tracking error converges to a zero-centered neighborhood within a fixed time, and the singularity which often appears in the existing fixed-time control (FTC) methods of HONSs is effectively avoided. Simulation results demonstrate the scheme developed.

Decentralized adaptive neural safe tracking control for nonlinear systems with conflicted output constraints.

The issue of decentralized adaptive safe tracking control for interconnected large-scale nonlinear systems (ILSNSs) with conflicted output constraints is discussed in this paper. By "conflicted output constraints", we mean that the output constraint functions conflict with reference signal, i.e., the reference signal is not completely constrained within the constraint range. In existing methods, it is always assumed that the reference signal is constrained within the constraint region. In practice, the constraints may be detected during system operation and conflict with the reference signal given in advance. In this particular case, the existing methods based on barrier Lyapunov function (BLF) or nonlinear transformation function (NTF) are invalid. From a new point of view, this article designs a new safety reference signal (SRS) which is completely restricted within the constraint range by using the boundary protection approach. Meanwhile, a prescribed performance function which can arbitrarily define the convergence time and tracking accuracy is introduced so that the system output can better track the SRS. Then, combining backstepping technique and radial basis function neural network (RBFNN), a new controller is constructed, under which a desired tracking trajectory can be obtained under the premise of ensuring safety performance. Furthermore, by adding a dynamic event triggering mechanism (DETM) between the actuator and the plant, the communication burden is effectively reduced. Simulation results verify the scheme developed.

Convolution-Based Model-Solving Method for Three-Dimensional, Unsteady, Partial Differential Equations.

Neural networks are increasingly used widely in the solution of partial differential equations (PDEs). This letter proposes 3D-PDE-Net to solve the three-dimensional PDE. We give a mathematical derivation of a three-dimensional convolution kernel that can approximate any order differential operator within the range of expressing ability and then conduct 3D-PDE-Net based on this theory. An optimum network is obtained by minimizing the normalized mean square error (NMSE) of training data, and L-BFGS is the optimized algorithm of second-order precision. Numerical experimental results show that 3D-PDE-Net can achieve the solution with good accuracy using few training samples, and it is of highly significant in solving linear and nonlinear unsteady PDEs.

Automatic measurement of axial vertebral rotation in 3D vertebral models.

Axial Vertebral Rotation (AVR) is a significant indicator of adolescent idiopathic scoliosis (AIS). A host of methods are provided to measure AVR on coronal plane radiographs or 3D vertebral model. This paper provides a method of automatic AVR measurement in 3D vertebral model that is based on point cloud segmentation neural network and the tip of the spinous process searching algorithm. An improved PointNet using multi-input and attention mechanism named Multi-Input PointNet is proposed, which can segment the upper and lower endplates of the vertebral model accurately to determine the transverse plane of vertebral model. An algorithm is developed to search the tip of the spinous process according to the special structure of vertebrae. AVR angle is measured automatically using the midline of vertebral model and projection ofy-axis on the transverse plane of vertebral model based on points obtained above. We compare automatic measurement results with manual measurement results on different vertebral models. The experiment shows that automatic results can achieve accuracy of manual measurement results and the correlation coefficient of them is 0.986, proving our automatic AVR measurement method performs well.

A Novel Handwritten Digit Classification System Based on Convolutional Neural Network Approach.

An enormous number of CNN classification algorithms have been proposed in the literature. Nevertheless, in these algorithms, appropriate filter size selection, data preparation, limitations in datasets, and noise have not been taken into consideration. As a consequence, most of the algorithms have failed to make a noticeable improvement in classification accuracy. To address the shortcomings of these algorithms, our paper presents the following contributions: Firstly, after taking the domain knowledge into consideration, the size of the effective receptive field (ERF) is calculated. Calculating the size of the ERF helps us to select a typical filter size which leads to enhancing the classification accuracy of our CNN. Secondly, unnecessary data leads to misleading results and this, in turn, negatively affects classification accuracy. To guarantee the dataset is free from any redundant or irrelevant variables to the target variable, data preparation is applied before implementing the data classification mission. Thirdly, to decrease the errors of training and validation, and avoid the limitation of datasets, data augmentation has been proposed. Fourthly, to simulate the real-world natural influences that can affect image quality, we propose to add an additive white Gaussian noise with σ = 0.5 to the MNIST dataset. As a result, our CNN algorithm achieves state-of-the-art results in handwritten digit recognition, with a recognition accuracy of 99.98%, and 99.40% with 50% noise.

Blind Image Deblurring Using a Non-Linear Channel Prior Based on Dark and Bright Channels.

Blind image deblurring aims at recovering a clean image from the given blurry image without knowing the blur kernel. Recently proposed dark and extreme channel priors have shown their effectiveness in deblurring various blurry scenarios. However, these two priors fail to help the blur kernel estimation under the particular circumstance that clean images contain neither enough darkest nor brightest pixels. In this paper, we propose a novel and robust non-linear channel (NLC) prior for the blur kernel estimation to fill this gap. It is motivated by a simple idea that the blurring operation will increase the ratio of dark channel to bright channel. This change has been proved to be true both theoretically and empirically. Nonetheless, the presence of the NLC prior introduces a thorny optimization model. To handle it, an efficient algorithm based on projected alternating minimization (PAM) has been established which innovatively combines an approximate strategy, the half-quadratic splitting method, and fast iterative shrinkage-thresholding algorithm (FISTA). Extensive experimental results show that the proposed method achieves state-of-the-art results no matter when it has been applied in synthetic uniform and non-uniform benchmark datasets or in real blurry images.

Morphology-based realization of a rapid scoliosis correction simulation system.

OBJECTIVE: Scoliosis is a complex spinal deformity in 3D space that commonly occurs in teenagers, especially teenage girls, and judging the actual deformed spine situation using only CT images is difficult. However, using 3D finite element models to help doctors analyse the deformed spine is also time-consuming and laborious. Therefore, software that can quickly and easily perform scoliosis correction analysis is needed. To achieve rapid preoperative simulation of scoliosis correction in 3D space and help doctors construct surgical programmes faster, a morphology-based system was developed for simulating scoliosis correction performance. METHODS: The simulation system first takes advantage of the centre point of each vertebra on the entire spine model to fit a space curve. Then the system obtains information from the models and the space curve, and finally, uses the information to simulate scoliosis correction. The deformed spine model in the system can be corrected to a better state. RESULTS: During the simulation process, doctors can easily and clearly see how the vertebral models move, and the deformed spine parameters are also updated and shown. Using this system, doctors can easily simulate scoliosis correction according to their experience and quickly construct a surgical programme. CONCLUSIONS: The experimental results show that this system is capable of simulating scoliosis correction according to a doctor's own experience to speed up the operation and provides a scientific basis for the development of surgical programmes.

Hybrid regularizers-based adaptive anisotropic diffusion for image denoising.

To eliminate the staircasing effect for total variation filter and synchronously avoid the edges blurring for fourth-order PDE filter, a hybrid regularizers-based adaptive anisotropic diffusion is proposed for image denoising. In the proposed model, the [Formula: see text]-norm is considered as the fidelity term and the regularization term is composed of a total variation regularization and a fourth-order filter. The two filters can be adaptively selected according to the diffusion function. When the pixels locate at the edges, the total variation filter is selected to filter the image, which can preserve the edges. When the pixels belong to the flat regions, the fourth-order filter is adopted to smooth the image, which can eliminate the staircase artifacts. In addition, the split Bregman and relaxation approach are employed in our numerical algorithm to speed up the computation. Experimental results demonstrate that our proposed model outperforms the state-of-the-art models cited in the paper in both the qualitative and quantitative evaluations.

Gender difference of unconscious attentional bias in high trait anxiety individuals.

By combining binocular suppression technique and a probe detection paradigm, we investigated attentional bias to invisible stimuli and its gender difference in both high trait anxiety (HTA) and low trait anxiety (LTA) individuals. As an attentional cue, happy or fearful face pictures were presented to HTAs and LTAs for 800 ms either consciously or unconsciously (through binocular suppression). Participants were asked to judge the orientation of a gabor patch following the face pictures. Their performance was used to measure attentional effect induced by the cue. We found gender differences of attentional effect only in the unconscious condition with HTAs. Female HTAs exhibited difficulty in disengaging attention from the location where fearful faces were presented, while male HTAs showed attentional avoidance of it. Our results suggested that the failure to find attentional avoidance of threatening stimuli in many previous studies might be attributed to consciously presented stimuli and data analysis regardless of participants' gender. These findings also contributed to our understanding of gender difference in anxiety disorder.

Psychometric properties of the Padua Inventory in Chinese college samples.

The aim of the present study was to investigate the psychometric properties of the Padua Inventory in Chinese college samples. Three different samples (N1 = 1,939, N2 = 1,341, and N3 = 298) of Chinese college students were recruited. Exploratory factor analyses yielded a four-factor structure which was similar to that found in previous studies. Further, present results showed good internal consistency as well as convergent and divergent validity with the subscales of the Symptom Checklist-90 and the Penn State Worry Questionnaire. Although no sex difference was found on total score, women had significantly higher scores on Factor 2 (Urges and worries of losing control over motor behaviors) and Factor 4 (Checking), while men had significantly higher scores on Factor 3 (Contamination). Implications of ter the results and directions for research are discussed.

Continual training of attentional bias in social anxiety.

Using the dot-probe paradigm, it has been shown that high social anxiety is associated with an attentional bias toward negative information. In the present study, individuals with high social anxiety were divided into two groups randomly. One group was the attentional bias training group (Group T), and the other was the control group (Group C). For Group T, 7 days' continuous training of attentional bias was conducted using the dot-probe paradigm to make socially anxious individuals focus more on positive face pictures. The results showed that the training was effective in changing attentional bias in Group T. Scores of the Social Interaction Anxiety Scale (SIAS) in Group T were reduced compared to Group C, while the scores of Social Phobia Scale (SPS) and scores of Negative Evaluation Scale (FNE) showed no difference between the two groups, which suggested a limited reduction of social anxiety.