This paper presents a novel approximation-based event-triggered control of multi-input multi-output uncertain nonlinear continuous-time systems in affine form. The controller is approximated using a ...linearly parameterized neural network (NN) in the context of event-based sampling. After revisiting the NN approximation property in the context of event-based sampling, an event-triggered condition is proposed using the Lyapunov technique to reduce the network resource utilization and to generate the required number of events for the NN approximation. In addition, a novel weight update law for aperiodic tuning of the NN weights at triggered instants is proposed to relax the knowledge of complete system dynamics and to reduce the computation when compared with the traditional NN-based control. Nonetheless, a nonzero positive lower bound for the inter-event times is guaranteed to avoid the accumulation of events or Zeno behavior. For analyzing the stability, the event-triggered system is modeled as a nonlinear impulsive dynamical system and the Lyapunov technique is used to show local ultimate boundedness of all signals. Furthermore, in order to overcome the unnecessary triggered events when the system states are inside the ultimate bound, a dead-zone operator is used to reset the event-trigger errors to zero. Finally, the analytical design is substantiated with numerical results.
Diabetic Retinopathy (DR) is a micro vascular complication caused by long-term diabetes mellitus. Unidentified diabetic retinopathy leads to permanent blindness. Early identification of this disease ...requires frequent complex diagnostic procedure which is expensive and time consuming. In this article, we propose a composite deep neural network architecture with gated-attention mechanism for automated diagnosis of diabetic retinopathy. The feature descriptors obtained from multiple pre-trained deep Convolutional Neural Networks (CNNs) are used to represent color fundus retinal images. Spatial pooling methods are introduced to get the reduced versions of these representations without loosing much information. The proposed composite DNN learns independently from each of these reduced representations through different channels and contributes to improving the model generalization. In addition, model also includes gated attention blocks which allows the model to emphasize more on lesion portions of the retinal images while reduced attention to the non-lesion regions. Our experiments on APTOS-2019 Kaggle blindness detection challenge reveal that, the proposed approach leads to improved performance when compared to the existing best models. Our empirical studies also reveal that, the proposed approach leads to more generalised predictions with multi-modal representations when compared to those of uni-modal representations. The proposed composite deep neural network model recorded an accuracy of 82.54% (
↑
2%), and a Kappa score of 79 (
↑
9
points) for diabetic retinopathy severity level prediction.
Accurate metro ridership prediction can guide passengers in efficiently selecting their departure time and transferring from station to station. An increasing number of deep learning algorithms are ...being utilized to forecast metro ridership due to the development of computational intelligence. However, limited efforts have been exerted to consider spatiotemporal features, which are important in forecasting ridership through deep learning methods, in large-scale metro networks. To fill this gap, this paper proposes a parallel architecture comprising convolutional neural network (CNN) and bi-directional long short-term memory network (BLSTM) to extract spatial and temporal features, respectively. Metro ridership data are transformed into ridership images and time series. Spatial features can be learned from ridership image data by using CNN, which demonstrates favorable performance in video detection. Time series data are input into the BLSTM which considers the historical and future impacts of ridership in temporal feature extraction. The two networks are concatenated in parallel and prevented from interfering with each other. Joint spatiotemporal features are fed into a fully connected network for metro ridership prediction. The Beijing metro network is used to demonstrate the efficiency of the proposed algorithm. The proposed model outperforms traditional statistical models, deep learning architectures, and sequential structures, and is suitable for ridership prediction in large-scale metro networks. Metro authorities can thus effectively allocate limited resources to overcrowded areas for service improvement.
The rapid development of remote sensing sensors makes the acquisition, analysis, and application of hyperspectral images (HSIs) more and more extensive. However, the limited sample sets, ...high-dimensional features, highly correlated bands, and mixing spectral information make the classification of HSIs a great challenge. In this article, an unsupervised multiscale and diverse feature learning (UMsDFL) method is proposed for HSI classification, which deeply considers the spatial-spectral features via convolutional neural networks (CNNs). Specifically, after employing the simple noniterative clustering (SNIC) algorithm with the heuristic calculation of superpixel size, the HSIs are segmented into superpixels for feature learning. The unsupervised network is designed with the convolutional encoder and decoder, the additional clustering branch, and the multilayer feature fusion to enhance the distinguishability of feature learning and the reusability of feature maps. Then, the spatial relationships and object attributes in large- and small-scale contexts are learned collaboratively through the unsupervised network to utilize the complementary multiscale characteristics. Moreover, the diverse features of hyperspectral information and nonsubsampled contourlet transform (NSCT) textures are learned simultaneously via the unsupervised network to alleviate the insufficiency of geometric representation. Finally, the random forest (RF) is adopted as the comprehensive classifier for land cover mapping based on the UMsDFL, and superpixel regularization is adopted to optimize the classification results. A series of experiments are performed on three real-world HSI datasets to demonstrate the effectiveness of our UMsDFL approach. The experimental results show that the proposed UMsDFL can achieve the overall accuracy of 79.23%, 96.49%, and 77.26% for Houston, Pavia, and Dioni datasets, respectively, when there are only five samples per class for training.
Automated machining feature recognition, a sub-discipline of solid modeling, has been an active research area for last three decades and is a critical component in digital manufacturing thread for ...detecting manufacturing information from computer aided design (CAD) models. In this paper, a novel framework using Deep 3D Convolutional Neural Networks (3D-CNNs) termed FeatureNet to learn machining features from CAD models of mechanical parts is presented. FeatureNet learns the distribution of complex manufacturing feature shapes across a large 3D model dataset and discovers distinguishing features that help in recognition process automatically. To train FeatureNet, a large-scale mechanical part datasets of 3D CAD models with labeled machining features is automatically constructed. The proposed framework can recognize manufacturing features from the low-level geometric data such as voxels with a very high accuracy. The developed framework can also recognize planar intersecting features in the 3D CAD models. Extensive numerical experiments show that FeatureNet enables significant improvements over the state-of-the-arts manufacturing feature detection techniques. The developed data-driven framework can easily be extended to identify a large variety of machining features leading to a sound foundation for real-time computer aided process planning (CAPP) systems.
•A novel deep 3D CNN framework to learn machining features from CAD models.•A large-scale labeled manufacturing features dataset with 3D CAD models.•Significant improvements over the state-of-the-arts manufacturing feature detection.
There is a need to develop an algorithm that can determine the relative activities of radioisotopes in a large data set of low-resolution gamma-ray spectra that contain a mixture of many ...radioisotopes. Low-resolution gamma-ray spectra that contain mixtures of radioisotopes often exhibit feature overlap, requiring algorithms that can analyze these features when overlap occurs. While machine learning and pattern recognition algorithms have shown promise for the problem of radioisotope identification, their ability to identify and quantify mixtures of radioisotopes has not been studied. Because machine-learning algorithms use abstract features of the spectrum, such as the shape of overlapping peaks and Compton continuum, they are a natural choice for analyzing radioisotope mixtures. An artificial neural network (ANN) has been trained to calculate the relative activities of 32 radioisotopes in a spectrum. The ANN is trained with simulated gamma-ray spectra, allowing easy expansion of the library of target radioisotopes. In this paper, we present our initial algorithms based on an ANN and evaluate them against a series of measured and simulated spectra.
Friction is one of the significant obstacles that hinders high-performance robot tracking control because accurate friction modeling and effective compensation are challenging issues. To address this ...problem, in this paper, we propose a modified neural network (NN) structure with additional jump approximation activation functions to model the inherent discontinuous friction in robotic systems, this structure allows us to improve the NN approximation accuracy without using too many NN nodes. The modeling accuracy is theoretically guaranteed by a composite learning technique, it explores both online historical data and instantaneous data to achieve NN weight convergence under a much weaker interval-excitation condition than the stringent persistent-excitation condition. Furthermore, a partitioned NN technique is used to handle a problem caused by variable substitution when formulating the prediction error for composite learning. This technique also helps us to alleviate the requirements regarding the inertial matrix inversion and joint acceleration signals. The practical exponential stability of the closed-loop system is proved under the more realizable interval-excitation condition. Experimental results demonstrate the effectiveness and superiority of the proposed approach.
This article proposes an adaptive type-2 fuzzy neural network control system to enhance the performance of power quality improvement. First, the dynamic model of APF with lumped uncertainties caused ...by parameter perturbation of ac inductor and dc capacitor is briefly introduced. Then, an integral-type terminal sliding mode control (TSMC) is developed for the finite-time reference signal tracking. Meanwhile, in terms of the considered chattering problem, saturation function is utilized in the proposed TSMC. Moreover, an adaptive type-2 fuzzy neural network (T2RFSFNN) is derived to achieve the model-free design, by applying the recurrent feature selection algorithm in the type-2 fuzzy neural network. To enhance the capacity to represent the uncertainties, adaptive learning mechanisms for updating the parameters of T2RFSFNN are derived by the Lyapunov theorem. Furthermore, a robust compensator with an adaptive uncertainty estimation law is investigated to relax the requirement for lumped uncertainties. Finally, the control performance using the developed T2RFSFNN is evaluated by some comparative experimental results.
• CNN-based HMMs are proposed to detect faults in rolling element bearings.• This model takes advantage of both the CNN and HMMs for their strong ability in data feature learning and pattern ...recognition.• The agreeable classification accuracy and stability are tested by benchmark data and experimental data investigations.• The average classification accuracy ratios are 98.125% and 98% for two data series, which is better than those of CNN model alone, SVM and BP neural network.
Vibration signals of faulty rolling element bearings usually exhibit non-linear and non-stationary characteristics caused by the complex working environment. It is difficult to develop a robust method to detect faults in bearings based on signal processing techniques. In this paper, convolutional neural network -based hidden Markov models (CNNHMMs) are presented to classify multi-faults in mechanical systems. In CNNHMMs, a CNN model is first employed to learn data features automatically from raw vibration signals. By utilizing the t-distributed stochastic neighbor embedding (t-SNE) technique, feature visualization is constructed to manifest the powerful learning ability of CNN. Then, HMMs are employed as a strong stability tool to classify faults. Both the benchmark data and experimental data are applied to the CNNHMMs. Classification results confirm the superior performance of the present combination model by comparing with CNN model alone, support vector machine (SVM) and back propagation (BP) neural network. It is shown that the average classification accuracy ratios are 98.125% and 98% for two data series with agreeable error rate reductions.
An energy-efficient convolutional neural network (CNN) accelerator is proposed for the video application. Previous works exploited the sparsity of differential (Diff) frame activation, but the ...improvement is limited as many Diff-frame data is small but non-zero. Processing of irregular sparse data also leads to low hardware utilization. To solve these problems, two key innovations are proposed in this article. First, we implement a hybrid-precision inter-frame-reuse architecture which takes advantage of both low bit-width and high sparsity of Diff-frame data. This technology can accelerate 3.2<inline-formula> <tex-math notation="LaTeX">\times </tex-math></inline-formula> inference speed with no accuracy loss. Second, we design a conv-pattern-aware processing array that achieves the 2.48<inline-formula> <tex-math notation="LaTeX">\times </tex-math></inline-formula>-14.2<inline-formula> <tex-math notation="LaTeX">\times </tex-math></inline-formula> PE utilization rate to process sparse data for different convolution kernels. The accelerator chip was implemented in 65-nm CMOS technology. To the best of our knowledge, it is the first silicon-proven CNN accelerator that supports inter-frame data reuse. Attributed to the inter-frame similarity, this video CNN accelerator reaches the minimum energy consumption of 24.7 <inline-formula> <tex-math notation="LaTeX">\mu \text{J} </tex-math></inline-formula>/frame in the MobileNet-slim model, which is 76.3% less than the baseline.
PDF
